Posts in category: Google AI

Posted by Shayne Longpre, Student Researcher, and Adam Roberts, Senior Staff Software Engineer, Google Research, Brain Team

Language models are now capable of performing many new natural language processing (NLP) tasks by reading instructions, often that they hadn’t seen before. The ability to reason on new tasks is mostly credited to training models on a wide variety of unique instructions, known as “instruction tuning”, which was introduced by FLAN and extended in T0, Super-Natural Instructions, MetaICL, and InstructGPT. However, much of the data that drives these advances remain unreleased to the broader research community. 

In “The Flan Collection: Designing Data and Methods for Effective Instruction Tuning”, we closely examine and release a newer and more extensive publicly available collection of tasks, templates, and methods for instruction tuning to advance the community’s ability to analyze and improve instruction-tuning methods. This collection was first used in Flan-T5 and Flan-PaLM, for which the latter achieved significant improvements over PaLM. We show that training a model on this collection yields improved performance over comparable public collections on all tested evaluation benchmarks, e.g., a 3%+ improvement on the 57 tasks in the Massive Multitask Language Understanding (MMLU) evaluation suite and 8% improvement on BigBench Hard (BBH). Analysis suggests the improvements stem both from the larger and more diverse set of tasks and from applying a set of simple training and data augmentation techniques that are cheap and easy to implement: mixing zero-shot, few-shot, and chain of thought prompts at training, enriching tasks with input inversion, and balancing task mixtures. Together, these methods enable the resulting language models to reason more competently over arbitrary tasks, even those for which it hasn’t seen any fine-tuning examples. We hope making these findings and resources publicly available will accelerate research into more powerful and general-purpose language models.

Public instruction tuning data collections

Since 2020, several instruction tuning task collections have been released in rapid succession, shown in the timeline below. Recent research has yet to coalesce around a unified set of techniques, with different sets of tasks, model sizes, and input formats all represented. This new collection, referred to below as “Flan 2022”, combines prior collections from FLAN, P3/T0, and Natural Instructions with new dialog, program synthesis, and complex reasoning tasks.

A timeline of public instruction tuning collections, including: UnifiedQA, CrossFit, Natural Instructions, FLAN, P3/T0, MetaICL, ExT5, Super-Natural Instructions, mT0, Unnatural Instructions, Self-Instruct, and OPT-IML Bench. The table describes the release date, the task collection name, the model name, the base model(s) that were finetuned with this collection, the model size, whether the resulting model is Public (green) or Not Public (red), whether they train with zero-shot prompts (“ZS”), few-shot prompts (“FS”), chain-of-thought prompts (“CoT”) together (“+”) or separately (“/”), the number of tasks from this collection in Flan 2022, the total number of examples, and some notable methods, related to the collections, used in these works. Note that the number of tasks and examples vary under different assumptions and so are approximations. Counts for each are reported using task definitions from the respective works.

In addition to scaling to more instructive training tasks, The Flan Collection combines training with different types of input-output specifications, including just instructions (zero-shot prompting), instructions with examples of the task (few-shot prompting), and instructions that ask for an explanation with the answer (chain of thought prompting). Except for InstructGPT, which leverages a collection of proprietary data, Flan 2022 is the first work to publicly demonstrate the strong benefits of mixing these prompting settings together during training. Instead of a trade-off between the various settings, mixing prompting settings during training improves all prompting settings at inference time, as shown below for both tasks held-in and held-out from the set of fine-tuning tasks.

Training jointly with zero-shot and few-shot prompt templates improves performance on both held-in and held-out tasks. The stars indicate the peak performance in each setting. Red lines denote the zero-shot prompted evaluation, lilac denotes few-shot prompted evaluation.

Evaluating instruction tuning methods

To understand the overall effects of swapping one instruction tuning collection for another, we fine-tune equivalently-sized T5 models on popular public instruction-tuning collections, including Flan 2021, T0++, and Super-Natural Instructions. Each model is then evaluated on a set of tasks that are already included in each of the instruction tuning collections, a set of five chain-of-thought tasks, and then a set of 57 diverse tasks from the MMLU benchmark, both with zero-shot and few-shot prompts. In each case, the new Flan 2022 model, Flan-T5, outperforms these prior works, demonstrating a more powerful general-purpose NLP reasoner.

Comparing public instruction tuning collections on held-in, chain-of-thought, and held-out evaluation suites, such as BigBench Hard and MMLU. All models except OPT-IML-Max (175B) are trained by us, using T5-XL with 3B parameters. Green text indicates improvement over the next best comparable T5-XL (3B) model.

Single task fine-tuning

In applied settings, practitioners usually deploy NLP models fine-tuned specifically for one target task, where training data is already available. We examine this setting to understand how Flan-T5 compares to T5 models as a starting point for applied practitioners. Three settings are compared: fine-tuning T5 directly on the target task, using Flan-T5 without further fine-tuning on the target task, and fine-tuning Flan-T5 on the target task. For both held-in and held-out tasks, fine-tuning Flan-T5 offers an improvement over fine-tuning T5 directly. In some instances, usually where training data is limited for a target task, Flan-T5 without further fine-tuning outperforms T5 with direct fine-tuning.

Flan-T5 outperforms T5 on single-task fine-tuning. We compare single-task fine-tuned T5 (blue bars), single-task fine-tuned Flan-T5 (red), and Flan-T5 without any further fine-tuning (beige).

An additional benefit of using Flan-T5 as a starting point is that training is significantly faster and cheaper, converging more quickly than T5 fine-tuning, and usually peaking at higher accuracies. This suggests less task-specific training data may be necessary to achieve similar or better results on a particular task.

Flan-T5 converges faster than T5 on single-task fine-tuning, for each of five held-out tasks from Flan fine-tuning. Flan-T5’s learning curve is indicated with the solid lines, and T5’s learning curve with the dashed line. All tasks are held-out during Flan finetuning.

There are significant energy efficiency benefits for the NLP community to adopt instruction-tuned models like Flan-T5 for single task fine-tuning, rather than conventional non-instruction-tuned models. While pre-training and instruction fine-tuning are financially and computationally expensive, they are a one-time cost, usually amortized over millions of subsequent fine-tuning runs, which can become more costly in aggregate, for the most prominent models. Instruction-tuned models offer a promising solution in significantly reducing the amount of fine-tuning steps needed to achieve the same or better performance.

Conclusion

The new Flan instruction tuning collection unifies the most popular prior public collections and their methods, while adding new templates and simple improvements like training with mixed prompt settings. The resulting method outperforms Flan, P3, and Super-Natural Instructions on held-in, chain of thought, MMLU, and BBH benchmarks by 3–17% across zero-shot and few-shot variants. Results suggest this new collection serves as a more performant starting point for researchers and practitioners interested in both generalizing to new instructions or fine-tuning on a single new task.

Acknowledgements

It was a privilege to work with Jason Wei, Barret Zoph, Le Hou, Hyung Won Chung, Tu Vu, Albert Webson, Denny Zhou, and Quoc V Le on this project.

Read More

Posted by Sreenivas Gollapudi, Senior Staff Research Scientist, and Kostas Kollias, Staff Research Scientist, Google Research, Algorithms & Optimization Team

In many computing applications the system needs to make decisions to serve requests that arrive in an online fashion. Consider, for instance, the example of a navigation app that responds to driver requests. In such settings there is inherent uncertainty about important aspects of the problem. For example, the preferences of the driver with respect to features of the route are often unknown and the delays of road segments can be uncertain. The field of online machine learning studies such settings and provides various techniques for decision-making problems under uncertainty.

A navigation engine has to decide how to route this user’s request. The satisfaction of the user will depend on the (uncertain) congestion of the two routes and unknown preferences of the user on various features, such as how scenic, safe, etc., the route is.

A very well known problem in this framework is the multi-armed bandit problem, in which the system has a set of n available options (arms) from which it is asked to choose in each round (user request), e.g., a set of precomputed alternative routes in navigation. The user’s satisfaction is measured by a reward that depends on unknown factors such as user preferences and road segment delays. An algorithm’s performance over T rounds is compared against the best fixed action in hindsight by means of the regret (the difference between the reward of the best arm and the reward obtained by the algorithm over all T rounds). In the experts variant of the multi-armed bandit problem, all rewards are observed after each round and not just the one played by the algorithm.

An instance of the experts problem. The table presents the rewards obtained by following each of the 3 experts at each round = 1, 2, 3, 4. The best expert in hindsight (and hence the benchmark to compare against) is the middle one, with total reward 21. If, for example, we had selected expert 1 in the first two rounds and expert 3 in the last two rounds (recall that we need to select before observing the rewards of each round), we would have extracted reward 17, which would give a regret equal to 21 – 17 = 4.

These problems have been extensively studied, and existing algorithms can achieve sublinear regret. For example, in the multi-armed bandit problem, the best existing algorithms can achieve regret that is of the order √T. However, these algorithms focus on optimizing for worst-case instances, and do not account for the abundance of available data in the real world that allows us to train machine learned models capable of aiding us in algorithm design.

In “Online Learning and Bandits with Queried Hints” (presented at ITCS 2023), we show how an ML model that provides us with a weak hint can significantly improve the performance of an algorithm in bandit-like settings. Many ML models are trained accurately using relevant past data. In the routing application, for example, specific past data can be used to estimate road segment delays and past feedback from drivers can be used to learn the quality of certain routes. Models trained with such data can, in certain cases, give very accurate feedback. However, our algorithms achieve strong guarantees even when the feedback from the model is in the form of a less explicit weak hint. Specifically, we merely ask that the model predict which of two options will be better. In the navigation application this is equivalent to having the algorithm pick two routes and query an ETA model for which of the two is faster, or presenting the user with two routes with different characteristics and letting them pick the one that is best for them. By designing algorithms that leverage such a hint we can: Improve the regret of the bandits setting on an exponential scale in terms of dependence on T and improve the regret of the experts setting from order of √T to become independent of T. Specifically, our upper bound only depends on the number of experts n and is at most log(n).

Algorithmic Ideas

Our algorithm for the bandits setting utilizes the well known upper confidence bound (UCB) algorithm. The UCB algorithm maintains, as a score for each arm, the average reward observed on that arm so far and adds to it an optimism parameter that becomes smaller with the number of times the arm has been pulled, thus balancing between exploration and exploitation. Our algorithm applies the UCB scores on pairs of arms, mainly in an effort to utilize the available pairwise comparison model that can designate the better of two arms. Each pair of arms i and j is grouped as a meta-arm (i, j) whose reward in each round is equal to the maximum reward between the two arms. Our algorithm observes the UCB scores of the meta-arms and picks the pair (i, j) that has the highest score. The pair of arms are then passed as a query to the ML auxiliary pairwise prediction model, which responds with the best of the two arms. This response is the arm that is finally used by the algorithm.

The decision problem considers three candidate routes. Our algorithm instead considers all pairs of the candidate routes. Suppose pair 2 is the one with the highest score in the current round. The pair is given to the auxiliary ML pairwise prediction model, which outputs whichever of the two routes is better in the current round.

Our algorithm for the experts setting takes a follow-the-regularized-leader (FtRL) approach, which maintains the total reward of each expert and adds random noise to each, before picking the best for the current round. Our algorithm repeats this process twice, drawing random noise two times and picking the highest reward expert in each of the two iterations. The two selected experts are then used to query the auxiliary ML model. The model’s response for the best between the two experts is the one played by the algorithm.

Results

Our algorithms utilize the concept of weak hints to achieve strong improvements in terms of theoretical guarantees, including an exponential improvement in the dependence of regret on the time horizon or even removing this dependence altogether. To illustrate how the algorithm can outperform existing baseline solutions, we present a setting where 1 of the n candidate arms is consistently marginally better than the n-1 remaining arms. We compare our ML probing algorithm against a baseline that uses the standard UCB algorithm to pick the two arms to submit to the pairwise comparison model. We observe that the UCB baseline keeps accumulating regret whereas the probing algorithm quickly identifies the best arm and keeps playing it, without accumulating regret.

An example in which our algorithm outperforms a UCB based baseline. The instance considers n arms, one of which is always marginally better than the remaining n-1.

Conclusion

In this work we explore how a simple pairwise comparison ML model can provide simple hints that prove very powerful in settings such as the experts and bandits problems. In our paper we further present how these ideas apply to more complex settings such as online linear and convex optimization. We believe our model of hints can have more interesting applications in ML and combinatorial optimization problems.

Acknowledgements

We thank our co-authors Aditya Bhaskara (University of Utah), Sungjin Im (University of California, Merced), and Kamesh Munagala (Duke University).

Read More

Posted by Posted by Alvin Rajkomar, Research Scientist, and Eric Loreaux, Software Engineer, Google Research

Today many people have digital access to their medical records, including their doctor’s clinical notes. However, clinical notes are hard to understand because of the specialized language that clinicians use, which contains unfamiliar shorthand and abbreviations. In fact, there are thousands of such abbreviations, many of which are specific to certain medical specialities and locales or can mean multiple things in different contexts. For example, a doctor might write in their clinical notes, “pt referred to pt for lbp“, which is meant to convey the statement: “Patient referred to physical therapy for low back pain.” Coming up with this translation is tough for laypeople and computers because some abbreviations are uncommon in everyday language (e.g., “lbp” means “low back pain”), and even familiar abbreviations, such as “pt” for “patient”, can have alternate meanings, such as “physical therapy.” To disambiguate between multiple meanings, the surrounding context must be considered. It’s no easy task to decipher all the meanings, and prior research suggests that expanding the shorthand and abbreviations can help patients better understand their health, diagnoses, and treatments.

In “Deciphering clinical abbreviations with a privacy protecting machine learning system”, published in Nature Communications, we report our findings on a general method that deciphers clinical abbreviations in a way that is both state-of-the-art and is on-par with board certified physicians in this task. We built the model using only public data on the web that wasn’t associated with any patient (i.e., no potentially sensitive data) and evaluated performance on real, de-identified notes from inpatient and outpatient clinicians from different health systems. To enable the model to generalize from web-data to notes, we created a way to algorithmically re-write large amounts of internet text to look as if it were written by a doctor (called web-scale reverse substitution), and we developed a novel inference method, (called elicitive inference).

The model input is a string that may or may not contain medical abbreviations. We trained a model to output a corresponding string in which all abbreviations are simultaneously detected and expanded. If the input string does not contain an abbreviation, the model will output the original string. By Rajkomar et al used under CC BY 4.0/ Cropped from original.

Rewriting Text to Include Medical Abbreviations

Building a system to translate doctors’ notes would usually start with a large, representative dataset of clinical text where all abbreviations are labeled with their meanings. But no such dataset for general use by researchers exists. We therefore sought to develop an automated way to create such a dataset but without the use of any actual patient notes, which might include sensitive data. We also wanted to ensure that models trained on this data would still work well on real clinical notes from multiple hospital sites and types of care, such as both outpatient and inpatient.

To do this, we referenced a dictionary of thousands of clinical abbreviations and their expansions, and found sentences on the web that contained uses of the expansions from this dictionary. We then “rewrote” those sentences by abbreviating each expansion, resulting in web data that looked like it was written by a doctor. For instance, if a website contained the phrase “patients with atrial fibrillation can have chest pain,” we would rewrite this sentence to “pts with af can have cp.” We then used the abbreviated text as input to the model, with the original text serving as the label. This approach provided us with large amounts of data to train our model to perform abbreviation expansion.

The idea of “reverse substituting” the long-forms for their abbreviations was introduced in prior research, but our distributed algorithm allows us to extend the technique to large, web-sized datasets. Our algorithm, called web-scale reverse substitution (WSRS), is designed to ensure that rare terms occur more frequently and common terms are down-sampled across the public web to derive a more balanced dataset. With this data in-hand, we trained a series of large transformer-based language models to expand the web text.

We generate text to train our model on the decoding task by extracting phrases from public web pages that have corresponding medical abbreviations (shaded boxes on the left) and then substituting in the appropriate abbreviations (shaded dots, right). Since some words are found much more frequently than others (“patient” more than “posterior tibialis”, both of which can be abbreviated “pt”), we downsampled common expansions to derive a more balanced dataset across the thousands of abbreviations. By Rajkomar et al used under CC BY 4.0.

Adapting Protein Alignment Algorithms to Unstructured Clinical Text

Evaluation of these models on the particular task of abbreviation expansion is difficult. Because they produce unstructured text as output, we had to figure out which abbreviations in the input correspond to which expansion in the output. To achieve this, we created a modified version of the Needleman Wunsch algorithm, which was originally designed for divergent sequence alignment in molecular biology, to align the model input and output and extract the corresponding abbreviation-expansion pairs. Using this alignment technique, we were able to evaluate the model’s capacity to detect and expand abbreviations accurately. We evaluated Text-to-Text Transfer Transformer (T5) models of various sizes (ranging from 60 million to over 60 billion parameters) and found that larger models performed translation better than smaller models, with the biggest model achieving the best performance.

Creating New Model Inference Techniques to Coax the Model

However, we did find something unexpected. When we evaluated the performance on multiple external test sets from real clinical notes, we found the models would leave some abbreviations unexpanded, and for larger models, the problem of incomplete expansion was even worse. This is mainly due to the fact that while we substitute expansions on the web for their abbreviations, we have no way of handling the abbreviations that are already present. This means that the abbreviations appear in both the original and rewritten text used as respective labels and input, and the model learns not to expand them.

To address this, we developed a new inference-chaining technique in which the model output is fed again as input to coax the model to make further expansions as long as the model is confident in the expansion. In technical terms, our best-performing technique, which we call elicitive inference, involves examining the outputs from a beam search above a certain log-likelihood threshold. Using elicitive inference, we were able to achieve state-of-the-art capability of expanding abbreviations in multiple external test sets.

Real example of the model’s input (left) and output (right).

Comparative Performance

We also sought to understand how patients and doctors currently perform at deciphering clinical notes, and how our model compared. We found that lay people (people without specific medical training) demonstrated less than 30% comprehension of the abbreviations present in the sample medical texts. When we allowed them to use Google Search, their comprehension increased to nearly 75%, still leaving 1 out of 5 abbreviations indecipherable. Unsurprisingly, medical students and trained physicians performed much better at the task with an accuracy of 90%. We found that our largest model was capable of matching or exceeding experts, with an accuracy of 98%.

How does the model perform so well compared to physicians in this task? There are two important factors in the model’s high comparative performance. Part of the discrepancy is that there were some abbreviations that clinicians did not even attempt to expand (such as “cm” for centimeter), which partly lowered the measured performance. This might seem unimportant, but for non-english speakers, these abbreviations may not be familiar, and so it may be helpful to have them written out. In contrast, our model is designed to comprehensively expand abbreviations. In addition, clinicians are familiar with abbreviations they commonly see in their speciality, but other specialists use shorthand that are not understood by those outside their fields. Our model is trained on thousands of abbreviations across multiple specialities and therefore can decipher a breadth of terms.

Towards Improved Health Literacy

We think there are numerous avenues in which large language models (LLMs) can help advance the health literacy of patients by augmenting the information they see and read. Most LLMs are trained on data that does not look like clinical note data, and the unique distribution of this data makes it challenging to deploy these models in an out-of-the-box fashion. We have demonstrated how to overcome this limitation. Our model also serves to “normalize” clinical note data, facilitating additional capabilities of ML to make the text easier for patients of all educational and health-literacy levels to understand.

Acknowledgements

This work was carried out in collaboration with Yuchen Liu, Jonas Kemp, Benny Li, Ming-Jun Chen, Yi Zhang, Afroz Mohiddin, and Juraj Gottweis. We thank Lisa Williams, Yun Liu, Arelene Chung, and Andrew Dai for many useful conversations and discussions about this work.

Read More

Posted by Marian Croak, VP, Google Research, Responsible AI and Human-Centered Technology <!–

–>

The last year showed tremendous breakthroughs in artificial intelligence (AI), particularly in large language models (LLMs) and text-to-image models. These technological advances require that we are thoughtful and intentional in how they are developed and deployed. In this blogpost, we share ways we have approached Responsible AI across our research in the past year and where we’re headed in 2023. We highlight four primary themes covering foundational and socio-technical research, applied research, and product solutions, as part of our commitment to build AI products in a responsible and ethical manner, in alignment with our AI Principles.

Theme 1: Responsible AI Research Advancements

Machine Learning Research

When machine learning (ML) systems are used in real world contexts, they can fail to behave in expected ways, which reduces their realized benefit. Our research identifies situations in which unexpected behavior may arise, so that we can mitigate undesired outcomes.

Across several types of ML applications, we showed that models are often underspecified, which means they perform well in exactly the situation in which they are trained, but may not be robust or fair in new situations, because the models rely on “spurious correlations” — specific side effects that are not generalizable. This poses a risk to ML system developers, and demands new model evaluation practices.

We surveyed evaluation practices currently used by ML researchers and introduced improved evaluation standards in work addressing common ML pitfalls. We identified and demonstrated techniques to mitigate causal “shortcuts”, which lead to a lack of ML system robustness and dependency on sensitive attributes, such as age or gender.

Shortcut learning: Age impacts correct medical diagnosis.

To better understand the causes of and mitigations for robustness issues, we decided to dig deeper into model design in specific domains. In computer vision, we studied the robustness of new vision transformer models and developed new negative data augmentation techniques to improve their robustness. For natural language tasks, we similarly investigated how different data distributions improve generalization across different groups and how ensembles and pre-trained models can help.

Another key part of our ML work involves developing techniques to build models that are more inclusive. For example, we look to external communities to guide understanding of when and why our evaluations fall short using participatory systems, which explicitly enable joint ownership of predictions and allow people to choose whether to disclose on sensitive topics.

Sociotechnical Research

In our quest to include a diverse range of cultural contexts and voices in AI development and evaluation, we have strengthened community-based research efforts, focusing on particular communities who are less represented or may experience unfair outcomes of AI. We specifically looked at evaluations of unfair gender bias, both in natural language and in contexts such as gender-inclusive health. This work is advancing more accurate evaluations of unfair gender bias so that our technologies evaluate and mitigate harms for people with queer and non-binary identities.

Alongside our fairness advancements, we also reached key milestones in our larger efforts to develop culturally-inclusive AI. We championed the importance of cross-cultural considerations in AI — in particular, cultural differences in user attitudes towards AI and mechanisms for accountability — and built data and techniques that enable culturally-situated evaluations, with a focus on the global south. We also described user experiences of machine translation, in a variety of contexts, and suggested human-centered opportunities for their improvement.

Human-Centered Research

At Google, we focus on advancing human-centered research and design. Recently, our work showed how LLMs can be used to rapidly prototype new AI-based interactions. We also published five new interactive explorable visualizations that introduce key ideas and guidance to the research community, including how to use saliency to detect unintended biases in ML models, and how federated learning can be used to collaboratively train a model with data from multiple users without any raw data leaving their devices.

Our interpretability research explored how we can trace the behavior of language models back to the training data itself, suggested new ways to compare differences in what models pay attention to, how we can explain emergent behavior, and how to identify human-understandable concepts learned by models. We also proposed a new approach for recommender systems that uses natural language explanations to make it easier for people to understand and control their recommendations.

Creativity and AI Research

We initiated conversations with creative teams on the rapidly changing relationship between AI technology and creativity. In the creative writing space, Google’s PAIR and Magenta teams developed a novel prototype for creative writing, and facilitated a writers’ workshop to explore the potential and limits of AI to assist creative writing. The stories from a diverse set of creative writers were published as a collection, along with workshop insights. In the fashion space, we explored the relationship between fashion design and cultural representation, and in the music space, we started examining the risks and opportunities of AI tools for music.

Top

Theme 2: Responsible AI Research in Products

The ability to see yourself reflected in the world around you is important, yet image-based technologies often lack equitable representation, leaving people of color feeling overlooked and misrepresented. In addition to efforts to improve representation of diverse skin tones across Google products, we introduced a new skin tone scale designed to be more inclusive of the range of skin tones worldwide. Partnering with Harvard professor and sociologist, Dr. Ellis Monk, we released the Monk Skin Tone (MST) Scale, a 10-shade scale that is available for the research community and industry professionals for research and product development. Further, this scale is being incorporated into features on our products, continuing a long line of our work to improve diversity and skin tone representation on Image Search and filters in Google Photos.

The 10 shades of the Monk Skin Tone Scale.

This is one of many examples of how Responsible AI in Research works closely with products across the company to inform research and develop new techniques. In another example, we leveraged our past research on counterfactual data augmentation in natural language to improve SafeSearch, reducing unexpected shocking Search results by 30%, especially on searches related to ethnicity, sexual orientation, and gender. To improve video content moderation, we developed new approaches for helping human raters focus their attention on segments of long videos that are more likely to contain policy violations. And, we’ve continued our research on developing more precise ways of evaluating equal treatment in recommender systems, accounting for the broad diversity of users and use cases.

In the area of large models, we incorporated Responsible AI best practices as part of the development process, creating Model Cards and Data Cards (more details below), Responsible AI benchmarks, and societal impact analysis for models such as GLaM, PaLM, Imagen, and Parti. We also showed that instruction fine-tuning results in many improvements for Responsible AI benchmarks. Because generative models are often trained and evaluated on human-annotated data, we focused on human-centric considerations like rater disagreement and rater diversity. We also presented new capabilities using large models for improving responsibility in other systems. For example, we have explored how language models can generate more complex counterfactuals for counterfactual fairness probing. We will continue to focus on these areas in 2023, also understanding the implications for downstream applications.

Top

Theme 3: Tooling and Techniques

Responsible Data

Data Documentation:

Extending our earlier work on Model Cards and the Model Card Toolkit, we released Data Cards and the Data Cards Playbook, providing developers with methods and tools to document appropriate uses and essential facts related to a model or dataset. We have also advanced research on best practices for data documentation, such as accounting for a dataset’s origins, annotation processes, intended use cases, ethical considerations, and evolution. We also applied this to healthcare, creating “healthsheets” to underlie the foundation of our international Standing Together collaboration, bringing together patients, health professionals, and policy-makers to develop standards that ensure datasets are diverse and inclusive and to democratize AI.

New Datasets:

Fairness: We released a new dataset to assist in ML fairness and adversarial testing tasks, primarily for generative text datasets. The dataset contains 590 words and phrases that show interactions between adjectives, words, and phrases that have been shown to have stereotypical associations with specific individuals and groups based on their sensitive or protected characteristics.

A partial list of the sensitive characteristics in the dataset denoting their associations with adjectives and stereotypical associations.

Toxicity: We constructed and publicly released a dataset of 10,000 posts to help identify when a comment’s toxicity depends on the comment it’s replying to. This improves the quality of moderation-assistance models and supports the research community working on better ways to remedy online toxicity.

Societal Context Data: We used our experimental societal context repository (SCR) to supply the Perspective team with auxiliary identity and connotation context data for terms relating to categories such as ethnicity, religion, age, gender, or sexual orientation — in multiple languages. This auxiliary societal context data can help augment and balance datasets to significantly reduce unintended biases, and was applied to the widely used Perspective API toxicity models.

Learning Interpretability Tool (LIT)

An important part of developing safer models is having the tools to help debug and understand them. To support this, we released a major update to the Learning Interpretability Tool (LIT), an open-source platform for visualization and understanding of ML models, which now supports images and tabular data. The tool has been widely used in Google to debug models, review model releases, identify fairness issues, and clean up datasets. It also now lets you visualize 10x more data than before, supporting up to 100s of thousands of data points at once.

A screenshot of the Language Interpretability Tool displaying generated sentences on a data table.

Counterfactual Logit Pairing

ML models are sometimes susceptible to flipping their prediction when a sensitive attribute referenced in an input is either removed or replaced. For example, in a toxicity classifier, examples such as “I am a man” and “I am a lesbian” may incorrectly produce different outputs. To enable users in the Open Source community to address unintended bias in their ML models, we launched a new library, Counterfactual Logit Pairing (CLP), which improves a model’s robustness to such perturbations, and can positively influence a model’s stability, fairness, and safety.

Illustration of fairness predictions that can be mitigated using counterfactual logit pairing.

Top

Theme 4: Demonstrating AI’s Societal Benefit

We believe that AI can be used to explore and address hard, unanswered questions around humanitarian and environmental issues. Our research and engineering efforts span many areas, including accessibility, health, and media representation, with the end goal of promoting inclusion and meaningfully improving people’s lives.

Accessibility

Following many years of research, we launched Project Relate, an Android app that uses a personalized AI-based speech recognition model to enable people with non-standard speech to communicate more easily with others. The app is available to English speakers 18+ in Australia, Canada, Ghana, India, New Zealand, the UK, and the US.

To help catalyze advances in AI to benefit people with disabilities, we also launched the Speech Accessibility Project. This project represents the culmination of a collaborative, multi-year effort between researchers at Google, Amazon, Apple, Meta, Microsoft, and the University of Illinois Urbana-Champaign. Together, this group built a large dataset of impaired speech that is available to developers to empower research and product development for accessibility applications. This work also complements our efforts to assist people with severe motor and speech impairments through improvements to techniques that make use of a user’s eye gaze.

Health

We’re also focused on building technology to better the lives of people affected by chronic health conditions, while addressing systemic inequities, and allowing for transparent data collection. As consumer technologies — such as fitness trackers and mobile phones — become central in data collection for health, we’ve explored use of technology to improve interpretability of clinical risk scores and to better predict disability scores in chronic diseases, leading to earlier treatment and care. And, we advocated for the importance of infrastructure and engineering in this space.

Many health applications use algorithms that are designed to calculate biometrics and benchmarks, and generate recommendations based on variables that include sex at birth, but might not account for users’ current gender identity. To address this issue, we completed a large, international study of trans and non-binary users of consumer technologies and digital health applications to learn how data collection and algorithms used in these technologies can evolve to achieve fairness.

Media

We partnered with the Geena Davis Institute on Gender in Media (GDI) and the Signal Analysis and Interpretation Laboratory (SAIL) at the University of Southern California (USC) to study 12 years of representation in TV. Based on an analysis of over 440 hours of TV programming, the report highlights findings and brings attention to significant disparities in screen and speaking time for light and dark skinned characters, male and female characters, and younger and older characters. This first-of-its-kind collaboration uses advanced AI models to understand how people-oriented stories are portrayed in media, with the ultimate goal to inspire equitable representation in mainstream media.

MUSE demo Source: Video Collection / Getty Images.

Top

Plans for 2023 and Beyond

We’re committed to creating research and products that exemplify positive, inclusive, and safe experiences for everyone. This begins by understanding the many aspects of AI risks and safety inherent in the innovative work that we do, and including diverse sets of voices in coming to this understanding.

• Responsible AI Research Advancements: We will strive to understand the implications of the technology that we create, through improved metrics and evaluations, and devise methodology to enable people to use technology to become better world citizens.
• Responsible AI Research in Products: As products leverage new AI capabilities for new user experiences, we will continue to collaborate closely with product teams to understand and measure their societal impacts and to develop new modeling techniques that enable the products to uphold Google’s AI Principles.
• Tools and Techniques: We will develop novel techniques to advance our ability to discover unknown failures, explain model behaviors, and to improve model output through training, responsible generation, and failure mitigation.
• Demonstrating AI’s Social Benefit: We plan to expand our efforts on AI for the Global Goals, bringing together research, technology, and funding to accelerate progress on the Sustainable Development Goals. This commitment will include $25 million to support NGOs and social enterprises. We will further our work on inclusion and equity by forming more collaborations with community-based experts and impacted communities. This includes continuing the Equitable AI Research Roundtables (EARR), focused on the potential impacts and downstream harms of AI with community based experts from the Othering and Belonging Institute at UC Berkeley, PolicyLink, and Emory University School of Law.

Building ML models and products in a responsible and ethical manner is both our core focus and core commitment.

Acknowledgements

This work reflects the efforts from across the Responsible AI and Human-Centered Technology community, from researchers and engineers to product and program managers, all of whom contribute to bringing our work to the AI community.

Google Research, 2022 & Beyond

This was the second blog post in the “Google Research, 2022 & Beyond” series. Other posts in this series are listed in the table below:

Read More

Here are nine ways we use AI today to make our products even more helpful.Read More

Posted by Jeff Dean, Senior Fellow and SVP of Google Research, on behalf of the Google Research community

<!–

–>

I’ve always been interested in computers because of their ability to help people better understand the world around them. Over the last decade, much of the research done at Google has been in pursuit of a similar vision — to help people better understand the world around them and get things done. We want to build more capable machines that partner with people to accomplish a huge variety of tasks. All kinds of tasks. Complex, information-seeking tasks. Creative tasks, like creating music, drawing new pictures, or creating videos. Analysis and synthesis tasks, like crafting new documents or emails from a few sentences of guidance, or partnering with people to jointly write software together. We want to solve complex mathematical or scientific problems. Transform modalities, or translate the world’s information into any language. Diagnose complex diseases, or understand the physical world. Accomplish complex, multi-step actions in both the virtual software world and the physical world of robotics.

We’ve demonstrated early versions of some of these capabilities in research artifacts, and we’ve partnered with many teams across Google to ship some of these capabilities in Google products that touch the lives of billions of users. But the most exciting aspects of this journey still lie ahead!

With this post, I am kicking off a series in which researchers across Google will highlight some exciting progress we’ve made in 2022 and present our vision for 2023 and beyond. I will begin with a discussion of language, computer vision, multi-modal models, and generative machine learning models. Over the next several weeks, we will discuss novel developments in research topics ranging from responsible AI to algorithms and computer systems to science, health and robotics. Let’s get started!

<!–

–>

Language Models

The progress on larger and more powerful language models has been one of the most exciting areas of machine learning (ML) research over the last decade. Important advances along the way have included new approaches like sequence-to-sequence learning and our development of the Transformer model, which underlies most of the advances in this space in the last few years. Although language models are trained on surprisingly simple objectives, like predicting the next token in a sequence of text given the preceding tokens, when large models are trained on sufficiently large and diverse corpora of text, the models can generate coherent, contextual, natural-sounding responses, and can be used for a wide range of tasks, such as generating creative content, translating between languages, helping with coding tasks, and answering questions in a helpful and informative way. Our ongoing work on LaMDA explores how these models can be used for safe, grounded, and high-quality dialog to enable contextual multi-turn conversations.

Natural conversations are clearly an important and emergent way for people to interact with computers. Rather than contorting ourselves to interact in ways that best accommodate the limitations of computers, we can instead have natural conversations to accomplish a wide variety of tasks. I’m excited about the progress we’ve made in making LaMDA useful and factual.

In April, we described our work on PaLM, a large, 540 billion parameter language model built using our Pathways software infrastructure and trained on multiple TPU v4 Pods. The PaLM work demonstrated that, despite being trained solely on the objective of predicting the next token, large-scale language models trained on large amounts of multi-lingual data and source code are capable of improving the state-of-the-art across a wide variety of natural language, translation, and coding tasks, despite never having been trained to specifically perform those tasks. This work provided additional evidence that increasing the scale of the model and training data can significantly improve capabilities.

Performance comparison between the PaLM 540B parameter model and the prior state-of-the-art (SOTA) on 58 tasks from the Big-bench suite. (See paper for details.)

We have also seen significant success in using large language models (LLMs) trained on source code (instead of natural language text data) that can assist our internal developers, as described in ML-Enhanced Code Completion Improves Developer Productivity. Using a variety of code completion suggestions from a 500 million parameter language model for a cohort of 10,000 Google software developers using this model in their IDE, we’ve seen that 2.6% of all code comes from suggestions generated by the model, reducing coding iteration time for these developers by 6%. We are working on enhanced versions of this and hope to roll it out to even more developers.

One of the broad key challenges in artificial intelligence is to build systems that can perform multi-step reasoning, learning to break down complex problems into smaller tasks and combining solutions to those to address the larger problem. Our recent work on Chain of Thought prompting, whereby the model is encouraged to “show its work” in solving new problems (similar to how your fourth-grade math teacher encouraged you to show the steps involved in solving a problem, rather than just writing down the answer you came up with), helps language models follow a logical chain of thought and generate more structured, organized and accurate responses. Like the fourth-grade math student that shows their work, not only does this make the problem-solving approach much more interpretable, it is also more likely that the correct answer will be found for complex problems that require multiple steps of reasoning.

Models that use standard prompting directly provide the answer to a multi-step reasoning problem. In contrast, chain of thought prompting teaches the model to deconstruct the problem into intermediate reasoning steps, better enabling it to reach the correct final answer.

One of the areas where multi-step reasoning is most clearly beneficial and measurable is in the ability of models to solve complex mathematical reasoning and scientific problems. A key research question is whether ML models can learn to solve complex problems using multi-step reasoning. By taking the general-purpose PaLM language model and fine-tuning it on a large corpus of mathematical documents and scientific research papers from arXiv, and then using Chain of Thought prompting and majority voting, the Minerva effort was able to demonstrate substantial improvements over the state-of-the-art for mathematical reasoning and scientific problems across a wide variety of scientific and mathematical benchmark suites.

Chain of Thought prompting is one way of better-expressing natural language prompts and examples to a model to improve its ability to tackle new tasks. The similar learned prompt tuning, in which a large language model is fine-tuned on a corpus of problem-domain–specific text, has shown great promise. In “Large Language Models Encode Clinical Knowledge”, we demonstrated that learned prompt tuning can adapt a general-purpose language model to the medical domain with relatively few examples and that the resulting model can achieve 67.6% accuracy on US Medical License Exam questions (MedQA), surpassing the prior ML state-of-the-art by over 17%. While still short compared to the abilities of clinicians, comprehension, recall of knowledge and medical reasoning all improve with model scale and instruction prompt tuning, suggesting the potential utility of LLMs in medicine. Continued work can help to create safe, helpful language models for clinical application.

Large language models trained on multiple languages can also help with translation from one language to another, even when they have never been taught to explicitly translate text. Traditional machine translation systems usually rely on parallel (translated) text to learn to translate from one language to another. However, since parallel text exists for a relatively small number of languages, many languages are often not supported in machine translation systems. In “Unlocking Zero-Resource Machine Translation to Support New Languages in Google Translate” and the accompanying papers “Building Machine Translation Systems for the Next Thousand Languages” and “Towards the Next 1000 Languages in Multilingual Machine Translation: Exploring the Synergy Between Supervised and Self-Supervised Learning”, we describe a set of techniques that use massively multilingual language models trained on monolingual (non-parallel) datasets to add 24 new languages spoken by 300 million people to Google Translate.

The amount of monolingual data per language versus the amount of parallel (translated) data per language. A small number of languages have large amounts of parallel data, but there is a long tail of languages with only monolingual data.

Another approach is represented with learned soft prompts, where instead of constructing new input tokens to represent a prompt, we add a small number of tunable parameters per task that can be learned from a few task examples. This approach generally yields high performance on tasks for which we have learned soft prompts, while allowing the large pre-trained language model to be shared across thousands of different tasks. This is a specific example of the more general technique of task adaptors, which allow a large portion of the parameters to be shared across tasks while still allowing task-specific adaptation and tuning.

As scale increases, prompt tuning, which conditions frozen models using tunable soft prompts, matches the performance of model tuning, despite using 25,000 fewer parameters.

Interestingly, the utility of language models can grow significantly as their sizes increase due to the emergence of new capabilities. “Characterizing Emergent Phenomena in Large Language Models” examines the sometimes surprising characteristic that these models are not able to perform particular complex tasks very effectively until reaching a certain scale. But then, once a critical amount of learning has happened (which varies by task), they suddenly show large jumps in the ability to perform a complex task accurately (as shown below). This raises the question of what new tasks will become feasible when these models are trained further.

The ability to perform multi-step arithmetic (left), succeed on college-level exams (middle), and identify the intended meaning of a word in context (right) all emerge only for models of sufficiently large scale. The models shown include LaMDA, GPT-3, Gopher, Chinchilla, and PaLM.

Additionally, language models of sufficient scale have the ability to learn and adapt to new information and tasks, which makes them even more versatile and powerful. As these models continue to improve and become more sophisticated, they will likely play an increasingly important role in many aspects of our lives.

Top

Computer Vision

Computer vision continues to evolve and make rapid progress. One trend that started with our work on Vision Transformers in 2020 is to use the Transformer architecture in computer vision models rather than convolutional neural networks. Although the localized feature-building abstraction of convolutions is a strong approach for many computer vision problems, it is not as flexible as the general attention mechanism in transformers, which can utilize both local and non-local information about the image throughout the model. However, the full attention mechanism is challenging to apply to higher resolution images, since it scales quadratically with image size.

In “MaxViT: Multi-Axis Vision Transformer”, we explore an approach that combines both local and non-local information at each stage of a vision model, but scales more efficiently than the full attention mechanism present in the original Vision Transformer work. This approach outperforms other state-of-the-art models on the ImageNet-1k classification task and various object detection tasks, but with significantly lower computational costs.

In MaxViT, a multi-axis attention mechanism conducts blocked local and dilated global attention sequentially followed by a FFN, with only a linear complexity. The pixels in the same colors are attended together.

In “Pix2Seq: A Language Modeling Framework for Object Detection”, we explore a simple and generic method that tackles object detection from a completely different perspective. Unlike existing approaches that are task-specific, we cast object detection as a language modeling task conditioned on the observed pixel inputs with the model trained to “read out” the locations and other attributes about the objects of interest in the image. Pix2Seq achieves competitive results on the large-scale object detection COCO dataset compared to existing highly-specialized and well-optimized detection algorithms, and its performance can be further improved by pre-training the model on a larger object detection dataset.

The Pix2Seq framework for object detection. The neural network perceives an image, and generates a sequence of tokens for each object, which correspond to bounding boxes and class labels.

Another long-standing challenge in computer vision is to better understand the 3-D structure of real-world objects from one or a few 2-D images. We have been trying multiple approaches to make progress in this area. In “Large Motion Frame Interpolation”, we demonstrated that short slow-motion videos can be created by interpolating between two pictures that were taken many seconds apart, even when there might have been significant movement in some parts of the scene. In “View Synthesis with Transformers”, we show how to combine two new techniques, light field neural rendering (LFNR) and generalizable patch-based neural rendering (GPNR), to synthesize novel views of a scene, a long-standing challenge in computer vision. LFNR is a technique that can accurately reproduce view-dependent effects by using transformers that learn to combine reference pixel colors. While LFNR works well on single scenes, its ability to generalize to novel scenes is limited. GPNR overcomes this by using a sequence of transformers with canonicalized positional encodings that can be trained on a set of scenes to synthesize views of new scenes. Together, these techniques enable high-quality view synthesis of novel scenes from just a couple of images of the scene, as shown below:

By combining LFNR and GPNR, models are able to produce new views of a scene given only a few images of it. These models are particularly effective when handling view-dependent effects like the refractions and translucency on the test tubes. Source: Still images from the NeX/Shiny dataset.

Going even further, in “LOLNerf: Learn from One Look”, we explore the ability to learn a high quality representation from just a single 2-D image. By training on many different examples of particular categories of objects (e.g., lots of single images of different cats), we can learn enough about the expected 3-D structure of objects to create a 3-D model from just a single image of a novel category (e.g., just a single image of your cat, as shown in the LOLCats clips below).

Top: Example cat images from AFHQ. Bottom: A synthesis of novel 3-D views created by LOLNeRF.

A general thrust of this work is to develop techniques that help computers have a better understanding of the 3-D world — a longstanding dream of computer vision!

Top

Multimodal Models

Most past ML work has focused on models that deal with a single modality of data (e.g., language models, image classification models, or speech recognition models). While there has been plenty of amazing progress in these areas, the future is even more exciting as we look forward to multi-modal models that can flexibly handle many different modalities simultaneously, both as model inputs and as model outputs. We have pushed in this direction in many ways over the past year.

Rather than relying on individual models tailored to specific tasks or domains, the next generation of multi-modal models can handle different modalities simultaneously by activating only the model pathways necessary for a given problem.

There are two key questions when building a multi-modal model that must be addressed to best enable cross-modality features and learning:

1. How much modality-specific processing should be done before allowing the learned representations to be merged?
2. What is the most effective way to mix the representations?

In our work on “Multi-modal Bottleneck Transformers” and the accompanying “Attention Bottlenecks for Multimodal Fusion” paper, we explore these tradeoffs and find that bringing together modalities after a few layers of modality-specific processing and then mixing the features from different modalities through a bottleneck layer is more effective than other techniques (as illustrated by the Bottleneck Mid Fusion in the figure below). This approach substantially improves accuracy on a variety of video classification tasks by learning to use multiple modalities of data to make classification decisions.

Sample attention configurations for multi-modal transformer encoders. Red and blue rows of dots represent encoder layers. Typical approaches to fusion of multi-modal transformer encoder features (“full fusion”) use pairwise self attention across hidden units in a layer (left). Bottleneck fusion (middle) restricts attention flow within a layer through tight latent units called attention bottlenecks. Bottleneck mid fusion (right) applies bottleneck fusion only to later layers in the model for optimal performance.

Combining modalities can often improve accuracy on even single-modality tasks. This is an area we have been exploring for many years, including our work on DeViSE, which combines image representations and word-embedding representations to improve image classification accuracy, even on unseen object categories. A modern variant of this general idea is found in Locked-image Tuning (LiT), a method that adds language understanding to an existing pre-trained image model. This approach contrastively trains a text encoder to match image representations from a powerful pre-trained image encoder. This simple method is data and compute efficient, and substantially improves zero-shot image classification performance compared to existing contrastive learning approaches.

LiT-tuning contrastively trains a text encoder to match a pre-trained image encoder. The text encoder learns to compute representations that align to those from the image encoder.

Another example of the uni-modal utility of multi-modal models is observed when co-training on related modalities, like images and videos. In this case, one can often improve accuracy on video action classification tasks compared to training on video data alone (especially when training data in one modality is limited).

Combining language with other modalities is a natural step for improving how users interact with computers. We have explored this direction in quite a number of ways this year. One of the most exciting is in combining language and vision inputs, either still images or videos. In “PaLI: Scaling Language-Image Learning”, we introduced a unified language-image model trained to perform many tasks in over 100 languages. These tasks span vision, language, and multimodal image and language applications, such as visual question answering, image captioning, object detection, image classification, optical character recognition, text reasoning, and others. By combining a vision transformer (ViT) with a text-based transformer encoder, and then a transformer-based decoder to generate textual answers, and training the whole system end-to-end on many different tasks simultaneously, the system achieves state-of-the-art results across many different benchmarks.

For example, PaLI achieves state-of-the-art results on the CrossModal-3600 benchmark, a diverse test of multilingual, multi-modal capabilities with an average CIDEr score of 53.4 across 35 languages (improving on the previous best score of 28.9). As the figure below shows, having a single model that can simultaneously understand multiple modalities and many languages and handle many tasks, such as captioning and question answering, will lead to computer systems where you can have a natural conversation about other kinds of sensory inputs, asking questions and getting answers to your needs in a wide variety of languages (“In Thai, can you say what is above the table in this image?”, “How many parakeets do you see sitting on the branches?”, “Describe this image in Swahili”, “What Hindi text is in this image?”).

The PaLI model addresses a wide range of tasks in the language-image, language-only and image-only domain using the same API (e.g., visual-question answering, image captioning, scene-text understanding, etc.). The model is trained to support over 100 languages and tuned to perform multilingually for multiple language-image tasks.

In a similar vein, our work on FindIt enables natural language questions about visual images to be answered through a unified, general-purpose and multitask visual grounding model that can flexibly answer different types of grounding and detection queries.

FindIt is a unified model for referring expression comprehension (first column), text-based localization (second), and the object detection task (third). FindIt can respond accurately when tested on object types and classes not known during training, e.g., “Find the desk” (fourth). We show the MattNet results for comparison.

The area of video question answering (e.g., given a baking video, being able to answer a question like “What is the second ingredient poured into the bowl?”) requires the ability to comprehend both textual inputs (the question) and video inputs (the relevant video) to produce a textual answer. In “Efficient Video-Text Learning with Iterative Co-tokenization”, multi-stream video inputs, which are versions of the same video input (e.g., a high resolution, low frame-rate video and a low resolution, high frame-rate video), are efficiently fused together with the text input to produce a text-based answer by the decoder. Instead of processing the inputs directly, the video-text iterative co-tokenization model learns a reduced number of useful tokens from the fused video-language inputs. This process is done iteratively, allowing the current feature tokenization to affect the selection of tokens at the next iteration, thus refining the selection.

An example input question for the video question answering task “What is the second ingredient poured into the bowl?” which requires deeper understanding of both the visual and text inputs. The video is an example from the 50 Salads dataset, used under the Creative Commons license.

The process of creating high-quality video content often includes several stages, from video capturing to video and audio editing. In some cases, dialogue is re-recorded in a studio (referred to as dialog replacement, post-sync or dubbing) to achieve high quality and replace original audio that might have been recorded in noisy or other suboptimal conditions. However, the dialog replacement process can be difficult and tedious because the newly recorded audio needs to be well synced with the video, often requiring several edits to match the exact timing of mouth movements. In “VDTTS: Visually-Driven Text-To-Speech”, we explore a multi-modal model for accomplishing this task more easily. Given desired text and the original video frames of a speaker, the model can generate speech output of the text that matches the video while also recovering aspects of prosody, such as timing or emotion. The system shows substantial improvements on a variety of metrics related to video-sync, speech quality, and speech pitch. Interestingly, the model can produce video-synchronized speech without any explicit constraints or losses in the model training to promote this.

In “Look and Talk: Natural Conversations with Google Assistant”, we show how an on-device multi-modal model can use both video and audio input to make interacting with Google Assistant much more natural. The model learns to use a number of visual and auditory cues, such as gaze direction, proximity, face matching, voice matching and intent classification, to more accurately determine if a nearby person is actually trying to talk to the Google Assistant device, or merely happens to be talking near the device without the intent of causing the device to take any action. With just the audio or visual features alone, this determination would be much more difficult.

Multi-modal models don’t have to be limited to just combining human-oriented modalities like natural language or imagery, and they are increasingly important for real-world autonomous vehicle and robotics applications. In this context, such models can take the raw output of sensors that are unlike any human senses, such as 3-D point cloud data from Lidar units on autonomous vehicles, and can combine this with data from other sensors, like vehicle cameras, to better understand the environment around them and to make better decisions. In “4D-Net for Learning Multi-Modal Alignment for 3D and Image Inputs in Time”, the 3-D point cloud data from Lidar is fused with the RGB data from the camera in real-time, with a self-attention mechanism controlling how the features are mixed together and weighted at different layers. The combination of the different modalities and the use of time-oriented features gives substantially improved accuracy in 3-D object recognition over using either modality on its own. More recent work on Lidar-camera fusion introduced learnable alignment and better geometric processing through inverse augmentation to further improve the accuracy of 3-D object recognition.

4D-Net effectively combines 3D LiDAR point clouds in time with RGB images, also streamed in time as video, learning the connections between different sensors and their feature representations.

Having single models that understand many different modalities fluidly and contextually and that can generate many different kinds of outputs (e.g., language, images or speech) in that context, is a much more useful, general purpose framing of ML. We’re excited about where this will take us because it will enable new exciting applications in many Google products and also advance the fields of health, science, creativity, robotics and more!

Top

Generative Models

The quality and capabilities of generative models for imagery, video, and audio has shown truly stunning and extraordinary advances in 2022. There are a wide variety of approaches for generative models, which must learn to model complex data sets (e.g., natural images). Generative adversarial networks, developed in 2014, set up two models working against each other. One is a generator, which tries to generate a realistic looking image (perhaps conditioned on an input to the model, like the category of image to generate), and the other is a discriminator, which is given the generated image and a real image and tries to determine which of the two is generated and which is real, hence the adversarial aspect. Each model is trying to get better and better at winning the competition against the other, resulting in both models getting better and better at their task, and in the end, the generative model can be used in isolation to generate images.

Advances in generative image model capabilities over the past decade. Left: From I. Goodfellow, et al. 2014. Middle: From M. Lucic, et al. 2019. Right: From Imagen.

Diffusion models, introduced in “Deep Unsupervised Learning using Nonequilibrium Thermodynamics” in 2015, systematically and slowly destroy structure in a data distribution through an iterative forward diffusion process. They then learn a reverse diffusion process that can restore the structure in the data that has been lost, even given high levels of noise. The forward process can be used to generate noisy starting points for the reverse diffusion process conditioned on various useful, controllable inputs to the model, so that the reverse diffusion (generative) process becomes controllable. This means that it is possible to ask the model to “generate an image of a grapefruit”, a much more useful capability than just “generate an image” if what you are after is indeed a sampling of images of grapefruits.

Various forms of autoregressive models have also been applied to the task of image generation. In 2016, “Pixel Recurrent Neural Networks” introduced PixelRNN, a recurrent architecture, and PixelCNN, a similar but more efficient convolutional architecture that was also investigated in “Conditional Image Generation with PixelCNN Decoders”. These two architectures helped lay the foundation for pixel-level generation using deep neural networks. They were followed in 2017 by VQ-VAE, proposed in “Neural Discrete Representation Learning”, a vector-quantized variational autoencoder. Combining this with PixelCNN yielded high-quality images. Then, in 2018 Image Transformer used the autoregressive Transformer model to generate images.

Until relatively recently, all of these image generation techniques were capable of generating images that are relatively low quality compared to real world images. However, several recent advances have opened the door for much better image generation performance. One is Contrastic Language-Image Pre-training (CLIP), a pre-training approach for jointly training an image encoder and a text decoder to predict [image, text] pairs. This pre-training task of predicting which caption goes with which image proved to be an efficient and scalable way to learn image representation and yielded good zero-shot performance on datasets like ImageNet.

In addition to CLIP, the toolkit of generative image models has recently grown. Large language model encoders have been shown to effectively condition image generation on long natural language descriptions rather than just a limited number of pre-set categories of images. Significantly larger training datasets of images and accompanying captions (which can be reversed to serve as text→image exemplars) have improved overall performance. All of these factors together have given rise to a range of models able to generate high-resolution images with strong adherence even to very detailed and fantastic prompts.

We focus here on two recent advances from teams in Google Research, Imagen and Parti.

Imagen is based on the Diffusion work discussed above. In their 2022 paper “Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding”, the authors show that a generic large language model (e.g., T5), pre-trained on text-only corpora, is surprisingly effective at encoding text for image synthesis. Somewhat surprisingly, increasing the size of the language model in Imagen boosts both sample fidelity and image-text alignment much more than increasing the size of the image diffusion model. The work offers several advances to Diffusion-based image generation, including a new memory-efficient architecture called Efficient U-Net and Classifier-Free Diffusion Guidance, which improves performance by occasionally “dropping out” conditioning information during training. Classifier-free guidance forces the model to learn to generate from the input data alone, thus helping it avoid problems that arise from over-relying on the conditioning information. “Guidance: a cheat code for diffusion models” provides a nice explanation.

Parti uses an autoregressive Transformer architecture to generate image pixels based on a text input. In “Vector-quantized Image Modeling with Improved VQGAN”, released in 2021, an encoder based on Vision Transformer is shown to significantly improve the output of a vector-quantized GAN model, VQGAN. This is extended in “Scaling Autoregressive Models for Content-Rich Text-to-Image Generation”, released in 2022, where much better results are obtained by scaling the Transformer encoder-decoder to 20B parameters. Parti also uses classifier-free guidance, described above, to sharpen the generated images. Perhaps not surprising given that it is a language model, Parti is particularly good at picking up on subtle cues in the prompt.

User Control

The advances described above make it possible to generate realistic still images based on text descriptions. However, sometimes text alone is not sufficient to enable you to create what you want — e.g., consider “A dog being chased by a unicorn on the beach” vs. “My dog being chased by a unicorn on the beach”. So, we have done subsequent research in providing new ways for users to control the generation process. In “DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generation”, users are able to fine-tune a trained model like Imagen or Parti to generate new images based on a combination of text and user-furnished images. This allows users to place images of themselves (or e.g., their pets) into generated images, thus allowing for much more user control. This is exemplified in “Prompt-to-Prompt Image Editing with Cross Attention Control”, where users are able to edit images using text prompts like “make the car into a bicycle” and in Imagen Editor, which allows users to iteratively edit images by filling in masked areas using text prompts.

Generative Video

One of the next research challenges we are tackling is to create generative models for video that can produce high resolution, high quality, temporally consistent videos with a high level of controllability. This is a very challenging area because unlike images, where the challenge was to match the desired properties of the image with the generated pixels, with video there is the added dimension of time. Not only must all the pixels in each frame match what should be happening in the video at the moment, they must also be consistent with other frames, both at a very fine-grained level (a few frames away, so that motion looks smooth and natural), but also at a coarse-grained level (if we asked for a two minute video of a plane taking off, circling, and landing, we must make thousands of frames that are consistent with this high-level video objective). This year we’ve made quite a lot of exciting progress on this lofty goal through two efforts, Imagen Video and Phenaki, each using somewhat different approaches.

Imagen Video generates high resolution videos with Cascaded Diffusion Models (described in more detail in “Imagen Video: High Definition Video Generation from Diffusion Models”). The first step is to take an input text prompt (“A happy elephant wearing a birthday hat walking under the sea”) and encode it into textual embeddings with a T5 text encoder. A base video diffusion model then generates a very rough sketch 16 frame video at 40×24 resolution and 3 frames per second. This is then followed by multiple temporal super-resolution (TSR) and spatial super-resolution (SSR) models to upsample and generate a final 128 frame video at 1280×768 resolution and 24 frames per second — resulting in 5.3s of high definition video. The resulting videos are high resolution, and are spatially and temporally consistent, but still quite short at ~5 seconds long.

<!–

Imagen Videos, each 192×320, 32 frames, 24 fps.

–>

“Phenaki: Variable Length Video Generation From Open Domain Textual Description”, released in 2022, introduces a new Transformer-based model for learning video representations, which compresses the video to a small representation of discrete tokens. Text conditioning is achieved by training a bi-directional Transformer model to generate video tokens based on a text description. These generated video tokens are then decoded to create the actual video. Because the model is causal in time, it can be used to generate variable-length videos. This opens the door to multi-prompt storytelling as illustrated in the video below.

Phenaki video generated from the complex prompt, “A photorealistic teddy bear is swimming in the ocean at San Francisco. The teddy bear goes under water. The teddy bear keeps swimming under the water with colorful fishes. A panda bear is swimming under water.”

It is possible to combine the Imagen Video and Phenaki models to benefit from both the high-resolution individual frames from Imagen and the long-form videos from Phenaki. The most straightforward way to do this is to use Imagen Video to handle superresolution of short video segments, while relying on the auto-regressive Phenaki model to generate the long-timescale video information.

Generative Audio

In addition to visual-oriented generative models, we have made significant progress on generative models for audio. In “AudioLM, a Language Modeling Approach to Audio Generation” (and the accompanying paper), we describe how to leverage advances in language modeling to generate audio without being trained on annotated data. Using a language-modeling approach for raw audio data instead of textual data introduces a number of challenges that need to be addressed.

First, the data rate for audio is significantly higher, leading to much longer sequences — while a written sentence can be represented by a few dozen characters, its audio waveform typically contains hundreds of thousands of values. Second, there is a one-to-many relationship between text and audio. This means that the same sentence can be uttered differently by different speakers with different speaking styles, emotional content and other audio background conditions.

To deal with this, we separate the audio generation process into two steps. The first involves a sequence of coarse, semantic tokens that capture both local dependencies (e.g., phonetics in speech, local melody in piano music) and global long-term structure (e.g., language syntax and semantic content in speech, harmony and rhythm in piano music), while heavily downsampling the audio signal to allow for modeling long sequences. One part of the model generates a sequence of coarse semantic tokens conditioned on the past sequence of such tokens. We then rely on a portion of the model that can use a sequence of coarse tokens to generate fine-grained audio tokens that are close to the final generated waveform.

When trained on speech, and without any transcript or annotation, AudioLM generates syntactically and semantically plausible speech continuations while also maintaining speaker identity and prosody for unseen speakers. AudioLM can also be used to generate coherent piano music continuations, despite being trained without any symbolic representation of music. You can listen to more samples here.

Concluding Thoughts on Generative Models

2022 has brought exciting advances in media generation. Computers can now interact with natural language and better understand your creative process and what you might want to create. This unlocks exciting new ways for computers to help users create images, video, and audio — in ways that surpass the limits of traditional tools!

This has inspired more research interest in how users can control the generative process. Advances in text-to-image and text-to-video have unlocked language as a powerful way to control generation, while work like Dream Booth has made it possible for users to kickstart the generative process with their own images. 2023 and beyond will surely be marked by advances in the quality and speed of media generation itself. Alongside these advances, we will also see new user experiences, allowing for more creative expression.

It is also worth noting that although these creative tools have tremendous possibilities for helping humans with creative tasks, they introduce a number of concerns — they could potentially generate harmful content of various kinds, or generate fake imagery or audio content that is difficult to distinguish from reality.  These are all issues we consider carefully when deciding when and how to deploy these models responsibly. 

Top

Responsible AI

AI must be pursued responsibly. Powerful language models can help people with many tasks, but without care they can also generate misinformation or toxic text. Generative models can be used for amazing creative purposes, enabling people to manifest their imagination in new and amazing ways, but they can also be used to create harmful imagery or realistic-looking images of events that never occurred.

These are complex topics to grapple with. Leaders in ML and AI must lead not only in state-of-the-art technologies, but also in state-of-the-art approaches to responsibility and implementation. In 2018, we were one of the first companies to articulate AI Principles that put beneficial use, users, safety, and avoidance of harms above all, and we have pioneered many best practices, like the use of model and data cards. More than words on paper, we apply our AI Principles in practice. You can see our latest AI Principles progress update here, including case studies on text-to-image generation models, techniques for avoiding gender bias in translations, and more inclusive and equitable evaluation skin tones. Similar updates were published in 2021, 2020, and 2019. As we pursue AI both boldly and responsibly, we continue to learn from users, other researchers, affected communities, and our experiences.

Our responsible AI approach includes the following:

• Focus on AI that is useful and benefits users and society.
• Intentionally apply our AI Principles (which are grounded in beneficial uses and avoidance of harm), processes, and governance to guide our work in AI, from research priorities to productization and uses.
• Apply the scientific method to AI R&D with research rigor, peer review, readiness reviews, and responsible approaches to access and externalization.
• Collaborate with multidisciplinary experts, including social scientists, ethicists, and other teams with socio-technical expertise.
• Listen, learn and improve based on feedback from developers, users, governments, and representatives of affected communities.
• Conduct regular reviews of our AI research and application development, including use cases. Provide transparency on what we’ve learned.
• Stay on top of current and evolving areas of concern and risk (e.g., safety, bias and toxicity) and address, research and innovate to respond to challenges and risks as they emerge.
• Lead on and help shape responsible governance, accountability, and regulation that encourages innovation and maximizes the benefits of AI while mitigating risks.
• Help users and society understand what AI is (and is not) and how to benefit from its potential.

In a subsequent blog post, leaders from our Responsible AI team will discuss work from 2022 in more detail and their vision for the field in the next few years.

Concluding Thoughts

We’re excited by the transformational advances discussed above, many of which we’re applying to make Google products more helpful to billions of users — including Search, Assistant, Ads, Cloud, Gmail, Maps, YouTube, Workspace, Android, Pixel, Nest, and Translate. These latest advances are making their way into real user experiences that will dramatically change how we interact with computers.

In the domain of language models, thanks to our invention of the Transformer model and advances like sequence-to-sequence learning, people can have a natural conversation (with a computer!) — and get surprisingly good responses (from a computer!). Thanks to new approaches in computer vision, computers can help people create and interact in 3D, rather than 2D. And thanks to new advances in generative models, computers can help people create images, videos, and audio — in ways they weren’t able to before with traditional tools (e.g., a keyboard and mouse). Combined with advances like natural language understanding, computers can understand what you’re trying to create — and help you realize surprisingly good results!

Another transformation changing how people interact with computers is the increasing capabilities of multi-modal models. We are working towards being able to create a single model that can understand many different modalities fluidly — understanding what each modality represents in context — and then actually generate different modes in that context. We’re excited by progress towards this goal! For example, we introduced a unified language model that can perform vision, language, question answering and object detection tasks in over 100 languages with state-of-the-art results across various benchmarks. In future applications, people can engage more senses to get computers to do what they want — e.g., “Describe this image in Swahili.” We’ve shown that on-device multi-modal models can make interacting with Google Assistant more natural. And we’ve demonstrated models that can, in various combinations, generate images, video, and audio controlled by natural language, images, and audio. More exciting things to come in this space!

As we innovate, we have a responsibility to users and society to thoughtfully pursue and develop these new technologies in accordance with our AI Principles. It’s not enough for us to develop state-of-the-art technologies, but we must also ensure that they are safe before broadly releasing them into the world, and we take this responsibility very seriously.

New advances in AI present an exciting horizon of new ways computers can help people get things done. For Google, many will enhance or transform our longstanding mission to organize the world’s information and make it universally accessible and useful. Over 20 years later, we believe this mission is as bold as ever. Today, what excites us is how we’re applying many of these advances in AI to enhance and transform user experiences — helping more people better understand the world around them and get more things done. My own longstanding vision of computers!

Acknowledgements

Thank you to the entire Research Community at Google for their contributions to this work! In addition, I would especially like to thank the many Googlers who provided helpful feedback in the writing of this post and who will be contributing to the other posts in this series, including Martin Abadi, Ryan Babbush, Vivek Bandyopadhyay, Kendra Byrne, Esmeralda Cardenas, Alison Carroll, Zhifeng Chen, Charina Chou, Lucy Colwell, Greg Corrado, Corinna Cortes, Marian Croak, Tulsee Doshi, Toju Duke, Doug Eck, Sepi Hejazi Moghadam, Pritish Kamath, Julian Kelly, Sanjiv Kumar, Ronit Levavi Morad, Pasin Manurangsi, Yossi Matias, Kathy Meier-Hellstern, Vahab Mirrokni, Hartmut Neven, Adam Paszke, David Patterson, Mangpo Phothilimthana, John Platt, Ben Poole, Tom Small, Vadim Smelyanskiy, Vincent Vanhoucke, and Leslie Yeh.

Read More

Google is using AI to help people facing disease and natural disasters, and to provide new opportunities for underserved communities.Read More

Today we published a paper outlining why we pursue AI. Read on for a preview, or visit ai.google to see the paper in full.Read More