Posts in category: Google AI

Posted by Lucas Dixon and Michael Terry, co-leads, PAIR, Google Research

PAIR (People + AI Research) first launched in 2017 with the belief that “AI can go much further — and be more useful to all of us — if we build systems with people in mind at the start of the process.” We continue to focus on making AI more understandable, interpretable, fun, and usable by more people around the world. It’s a mission that is particularly timely given the emergence of generative AI and chatbots.

Today, PAIR is part of the Responsible AI and Human-Centered Technology team within Google Research, and our work spans this larger research space: We advance foundational research on human-AI interaction (HAI) and machine learning (ML); we publish educational materials, including the PAIR Guidebook and Explorables (such as the recent Explorable looking at how and why models sometimes make incorrect predictions confidently); and we develop software tools like the Learning Interpretability Tool to help people understand and debug ML behaviors. Our inspiration this year is “changing the way people think about what THEY can do with AI.” This vision is inspired by the rapid emergence of generative AI technologies, such as large language models (LLMs) that power chatbots like Bard, and new generative media models like Google’s Imagen, Parti, and MusicLM. In this blog post, we review recent PAIR work that is changing the way we engage with AI.

Generative AI research

Generative AI is creating a lot of excitement, and PAIR is involved in a range of related research, from using language models to simulate complex community behaviors to studying how artists adopted generative image models like Imagen and Parti. These latter “text-to-image” models let a person input a text-based description of an image for the model to generate (e.g., “a gingerbread house in a forest in a cartoony style”). In a forthcoming paper titled “The Prompt Artists” (to appear in Creativity and Cognition 2023), we found that users of generative image models strive not only to create beautiful images, but also to create unique, innovative styles. To help achieve these styles, some would even seek unique vocabulary to help develop their visual style. For example, they may visit architectural blogs to learn what domain-specific vocabulary they can adopt to help produce distinctive images of buildings.

We are also researching solutions to challenges faced by prompt creators who, with generative AI, are essentially programming without using a programming language. As an example, we developed new methods for extracting semantically meaningful structure from natural language prompts. We have applied these structures to prompt editors to provide features similar to those found in other programming environments, such as semantic highlighting, autosuggest, and structured data views.

The growth of generative LLMs has also opened up new techniques to solve important long-standing problems. Agile classifiers are one approach we’re taking to leverage the semantic and syntactic strengths of LLMs to solve classification problems related to safer online discourse, such as nimbly blocking newer types of toxic language as quickly as it may evolve online. The big advance here is the ability to develop high quality classifiers from very small datasets — as small as 80 examples. This suggests a positive future for online discourse and better moderation of it: instead of collecting millions of examples to attempt to create universal safety classifiers for all use cases over months or years, more agile classifiers might be created by individuals or small organizations and tailored for their specific use cases, and iterated on and adapted in the time-span of a day (e.g., to block a new kind of harassment being received or to correct unintended biases in models). As an example of their utility, these methods recently won a SemEval competition to identify and explain sexism.

We’ve also developed new state-of-the-art explainability methods to identify the role of training data on model behaviors and misbehaviours. By combining training data attribution methods with agile classifiers, we also found that we can identify mislabelled training examples. This makes it possible to reduce the noise in training data, leading to significant improvements on model accuracy.

Collectively, these methods are critical to help the scientific community improve generative models. They provide techniques for fast and effective content moderation and dialogue safety methods that help support creators whose content is the basis for generative models’ amazing outcomes. In addition, they provide direct tools to help debug model misbehavior which leads to better generation.

Visualization and education

To lower barriers in understanding ML-related work, we regularly design and publish highly visual, interactive online essays, called AI Explorables, that provide accessible, hands-on ways to learn about key ideas in ML. For example, we recently published new AI Explorables on the topics of model confidence and unintended biases. In our latest Explorable, “From Confidently Incorrect Models to Humble Ensembles,” we discuss the problem with model confidence: models can sometimes be very confident in their predictions… and yet completely incorrect. Why does this happen and what can be done about it? Our Explorable walks through these issues with interactive examples and shows how we can build models that have more appropriate confidence in their predictions by using a technique called ensembling, which works by averaging the outputs of multiple models. Another Explorable, “Searching for Unintended Biases with Saliency”, shows how spurious correlations can lead to unintended biases — and how techniques such as saliency maps can detect some biases in datasets, with the caveat that it can be difficult to see bias when it’s more subtle and sporadic in a training set.

PAIR designs and publishes AI Explorables, interactive essays on timely topics and new methods in ML research, such as “From Confidently Incorrect Models to Humble Ensembles,” which looks at how and why models offer incorrect predictions with high confidence, and how “ensembling” the outputs of many models can help avoid this.

Transparency and the Data Cards Playbook

Continuing to advance our goal of helping people to understand ML, we promote transparent documentation. In the past, PAIR and Google Cloud developed model cards. Most recently, we presented our work on Data Cards at ACM FAccT’22 and open-sourced the Data Cards Playbook, a joint effort with the Technology, AI, Society, and Culture team (TASC). The Data Cards Playbook is a toolkit of participatory activities and frameworks to help teams and organizations overcome obstacles when setting up a transparency effort. It was created using an iterative, multidisciplinary approach rooted in the experiences of over 20 teams at Google, and comes with four modules: Ask, Inspect, Answer and Audit. These modules contain a variety of resources that can help you customize Data Cards to your organization’s needs:

• 18 Foundations: Scalable frameworks that anyone can use on any dataset type
• 19 Transparency Patterns: Evidence-based guidance to produce high-quality Data Cards at scale
• 33 Participatory Activities: Cross-functional workshops to navigate transparency challenges for teams
• Interactive Lab: Generate interactive Data Cards from markdown in the browser

The Data Cards Playbook is accessible as a learning pathway for startups, universities, and other research groups.

Software Tools

Our team thrives on creating tools, toolkits, libraries, and visualizations that expand access and improve understanding of ML models. One such resource is Know Your Data, which allows researchers to test a model’s performance for various scenarios through interactive qualitative exploration of datasets that they can use to find and fix unintended dataset biases.

Recently, PAIR released a new version of the Learning Interpretability Tool (LIT) for model debugging and understanding. LIT v0.5 provides support for image and tabular data, new interpreters for tabular feature attribution, a “Dive” visualization for faceted data exploration, and performance improvements that allow LIT to scale to 100k dataset entries. You can find the release notes and code on GitHub.

PAIR’s Learning Interpretability Tool (LIT), an open-source platform for visualization and understanding of ML models.

PAIR has also contributed to MakerSuite, a tool for rapid prototyping with LLMs using prompt programming. MakerSuite builds on our earlier research on PromptMaker, which won an honorable mention at CHI 2022. MakerSuite lowers the barrier to prototyping ML applications by broadening the types of people who can author these prototypes and by shortening the time spent prototyping models from months to minutes. 

A screenshot of MakerSuite, a tool for rapidly prototyping new ML models using prompt-based programming, which grew out of PAIR’s prompt programming research.

Ongoing work

As the world of AI moves quickly ahead, PAIR is excited to continue to develop new tools, research, and educational materials to help change the way people think about what THEY can do with AI.

For example, we recently conducted an exploratory study with five designers (presented at CHI this year) that looks at how people with no ML programming experience or training can use prompt programming to quickly prototype functional user interface mock-ups. This prototyping speed can help inform designers on how to integrate ML models into products, and enables them to conduct user research sooner in the product design process.

Based on this study, PAIR’s researchers built PromptInfuser, a design tool plugin for authoring LLM-infused mock-ups. The plug-in introduces two novel LLM-interactions: input-output, which makes content interactive and dynamic, and frame-change, which directs users to different frames depending on their natural language input. The result is more tightly integrated UI and ML prototyping, all within a single interface.

Recent advances in AI represent a significant shift in how easy it is for researchers to customize and control models for their research objectives and goals.These capabilities are transforming the way we think about interacting with AI, and they create lots of new opportunities for the research community. PAIR is excited about how we can leverage these capabilities to make AI easier to use for more people.

Acknowledgements

Thanks to everyone in PAIR and to all our collaborators.

Read More

The University of Tokyo, the University of Chicago and Google establish a strategic partnership on quantum information science and engineering.Read More

Posted by Deborah Cohen, Staff Research Scientist, and Craig Boutilier, Principal Scientist, Google Research

As virtual assistants become ubiquitous, users increasingly interact with them to learn about new topics or obtain recommendations and expect them to deliver capabilities beyond narrow dialogues of one or two turns. Dynamic planning, namely the capability to look ahead and replan based on the flow of the conversation, is an essential ingredient for the making of engaging conversations with the deeper, open-ended interactions that users expect.

While large language models (LLMs) are now beating state-of-the-art approaches in many natural language processing benchmarks, they are typically trained to output the next best response, rather than planning ahead, which is required for multi-turn interactions. However, in the past few years, reinforcement learning (RL) has delivered incredible results addressing specific problems that involve dynamic planning, such as winning games and protein folding.

Today, we are sharing our recent advances in dynamic planning for human-to-assistant conversations, in which we enable an assistant to plan a multi-turn conversation towards a goal and adapt that plan in real-time by adopting an RL-based approach. Here we look at how to improve long interactions by applying RL to compose answers based on information extracted from reputable sources, rather than relying on content generated by a language model. We expect that future versions of this work could combine LLMs and RL in multi-turn dialogues. The deployment of RL “in the wild” in a large-scale dialogue system proved a formidable challenge due to the modeling complexity, tremendously large state and action spaces, and significant subtlety in designing reward functions.

What is dynamic planning?

Many types of conversations, from gathering information to offering recommendations, require a flexible approach and the ability to modify the original plan for the conversation based on its flow. This ability to shift gears in the middle of a conversation is known as dynamic planning, as opposed to static planning, which refers to a more fixed approach. In the conversation below, for example, the goal is to engage the user by sharing interesting facts about cool animals. To begin, the assistant steers the conversation to sharks via a sound quiz. Given the user’s lack of interest in sharks, the assistant then develops an updated plan and pivots the conversation to sea lions, lions, and then cheetahs.

The assistant dynamically modifies its original plan to talk about sharks and shares facts about other animals.

Dynamic composition

To cope with the challenge of conversational exploration, we separate the generation of assistant responses into two parts: 1) content generation, which extracts relevant information from reputable sources, and 2) flexible composition of such content into assistant responses. We refer to this two-part approach as dynamic composition. Unlike LLM methods, this approach gives the assistant the ability to fully control the source, correctness, and quality of the content that it may offer. At the same time, it can achieve flexibility via a learned dialogue manager that selects and combines the most appropriate content.

In an earlier paper, “Dynamic Composition for Conversational Domain Exploration”, we describe a novel approach which consists of: (1) a collection of content providers, which offer candidates from different sources, such as news snippets, knowledge graph facts, and questions; (2) a dialogue manager; and (3) a sentence fusion module. Each assistant response is incrementally constructed by the dialogue manager, which selects candidates proposed by the content providers. The selected sequence of utterances is then fused into a cohesive response.

Dynamic planning using RL

At the core of the assistant response composition loop is a dialogue manager trained using off-policy RL, namely an algorithm that evaluates and improves a policy that is different from the policy used by the agent (in our case, the latter is based on a supervised model). Applying RL to dialogue management presents several challenges, including a large state space (as the state represents the conversation state, which needs to account for the whole conversation history) and an effectively unbounded action space (that may include all existing words or sentences in natural language).

We address these challenges using a novel RL construction. First, we leverage powerful supervised models — specifically, recurrent neural networks (RNNs) and transformers — to provide a succinct and effective dialogue state representation. These state encoders are fed with the dialogue history, composed of a sequence of user and assistant turns, and output a representation of the dialogue state in the form of a latent vector.

Second, we use the fact that a relatively small set of reasonable candidate utterances or actions can be generated by content providers at each conversation turn, and limit the action space to these. Whereas the action space is typically fixed in RL settings, because all states share the same action space, ours is a non-standard space in which the candidate actions may differ with each state, since content providers generate different actions depending on the dialogue context. This puts us in the realm of stochastic action sets, a framework that formalizes cases where the set of actions available in each state is governed by an exogenous stochastic process, which we address using Stochastic Action Q-Learning, a variant of the Q-learning approach. Q-learning is a popular off-policy RL algorithm, which does not require a model of the environment to evaluate and improve the policy. We trained our model on a corpus of crowd-compute–rated conversations obtained using a supervised dialogue manager.

Given the current dialogue history and a new user query, content providers generate candidates from which the assistant selects one. This process runs in a loop, and at the end the selected utterances are fused into a cohesive response.

Reinforcement learning model evaluation

We compared our RL dialogue manager with a launched supervised transformer model in an experiment using Google Assistant, which conversed with users about animals. A conversation starts when a user triggers the experience by asking an animal-related query (e.g., “How does a lion sound?”). The experiment was conducted using an A/B testing protocol, in which a small percentage of Assistant users were randomly sampled to interact with our RL-based assistant while other users interacted with the standard assistant.

We found that the RL dialogue manager conducts longer, more engaging conversations. It increases conversation length by 30% while improving user engagement metrics. We see an increase of 8% in cooperative responses to the assistant’s questions — e.g., “Tell me about lions,” in response to “Which animal do you want to hear about next?” Although there is also a large increase in nominally “non-cooperative” responses (e.g., “No,” as a reply to a question proposing additional content, such as “Do you want to hear more?”), this is expected as the RL agent takes more risks by asking pivoting questions. While a user may not be interested in the conversational direction proposed by the assistant (e.g., pivoting to another animal), the user will often continue to engage in a dialogue about animals.

From the non-cooperative user response in the 3rd turn (“No.”) and the query “Make a dog sound,” in the 5th turn, the assistant recognizes that the user is mostly interested in animal sounds and modifies its plan, providing sounds and sound quizzes.

In addition, some user queries contain explicit positive (e.g., “Thank you, Google,” or “I’m happy.”) or negative (e.g., “Shut up,” or “Stop.”) feedback. While an order of magnitude fewer than other queries, they offer a direct measure of user (dis)satisfaction. The RL model increases explicit positive feedback by 32% and reduces negative feedback by 18%.

Learned dynamic planning characteristics and strategies

We observe several characteristics of the (unseen) RL plan to improve user engagement while conducting longer conversations. First, the RL-based assistant ends 20% more turns in questions, prompting the user to choose additional content. It also better harnesses content diversity, including facts, sounds, quizzes, yes/no questions, open questions, etc. On average, the RL assistant uses 26% more distinct content providers per conversation than the supervised model.

Two observed RL planning strategies are related to the existence of sub-dialogues with different characteristics. Sub-dialogues about animal sounds are poorer in content and exhibit entity pivoting at every turn (i.e., after playing the sound of a given animal, we can either suggest the sound of a different animal or quiz the user about other animal sounds). In contrast, sub-dialogues involving animal facts typically contain richer content and have greater conversation depth. We observe that RL favors the richer experience of the latter, selecting 31% more fact-related content. Lastly, when restricting analysis to fact-related dialogues, the RL assistant exhibits 60% more focus-pivoting turns, that is, conversational turns that change the focus of the dialogue.

Below, we show two example conversations, one conducted by the supervised model (left) and the second by the RL model (right), in which the first three user turns are identical. With a supervised dialogue manager, after the user declined to hear about “today’s animal”, the assistant pivots back to animal sounds to maximize the immediate user satisfaction. While the conversation conducted by the RL model begins identically, it exhibits a different planning strategy to optimize the overall user engagement, introducing more diverse content, such as fun facts.

In the left conversation, conducted by the supervised model, the assistant maximizes the immediate user satisfaction. The right conversation, conducted by the RL model, shows different planning strategies to optimize the overall user engagement.

Future research and challenges

In the past few years, LLMs trained for language understanding and generation have demonstrated impressive results across multiple tasks, including dialogue. We are now exploring the use of an RL framework to empower LLMs with the capability of dynamic planning so that they can dynamically plan ahead and delight users with a more engaging experience.

Acknowledgements

The work described is co-authored by: Moonkyung Ryu, Yinlam Chow, Orgad Keller, Ido Greenberg, Avinatan Hassidim, Michael Fink, Yossi Matias, Idan Szpektor and Gal Elidan. We would like to thank: Roee Aharoni, Moran Ambar, John Anderson, Ido Cohn, Mohammad Ghavamzadeh, Lotem Golany, Ziv Hodak, Adva Levin, Fernando Pereira, Shimi Salant, Shachar Shimoni, Ronit Slyper, Ariel Stolovich, Hagai Taitelbaum, Noam Velan, Avital Zipori and the CrowdCompute team led by Ashwin Kakarla. We thank Sophie Allweis for her feedback on this blogpost and Tom Small for the visualization.

Read More

Posted by Jerry Wei, Student Researcher, and Denny Zhou, Principal Scientist, Google Research

There have recently been tremendous advances in language models, partly because they can perform tasks with strong performance via in-context learning (ICL), a process whereby models are prompted with a few examples of input-label pairs before performing the task on an unseen evaluation example. In general, models’ success at in-context learning is enabled by:

• Their use of semantic prior knowledge from pre-training to predict labels while following the format of in-context examples (e.g., seeing examples of movie reviews with “positive sentiment” and “negative sentiment” as labels and performing sentiment analysis using prior knowledge).
• Learning the input-label mappings in context from the presented examples (e.g., finding a pattern that positive reviews should be mapped to one label, and negative reviews should be mapped to a different label).

In “Larger language models do in-context learning differently”, we aim to learn about how these two factors (semantic priors and input-label mappings) interact with each other in ICL settings, especially with respect to the scale of the language model that’s used. We investigate two settings to study these two factors — ICL with flipped labels (flipped-label ICL) and ICL with semantically-unrelated labels (SUL-ICL). In flipped-label ICL, labels of in-context examples are flipped so that semantic priors and input-label mappings disagree with each other. In SUL-ICL, labels of in-context examples are replaced with words that are semantically unrelated to the task presented in-context. We found that overriding prior knowledge is an emergent ability of model scale, as is the ability to learn in-context with semantically-unrelated labels. We also found that instruction tuning strengthens the use of prior knowledge more than it increases the capacity to learn input-label mappings.

An overview of flipped-label ICL and semantically-unrelated label ICL (SUL-ICL), compared with regular ICL, for a sentiment analysis task. Flipped-label ICL uses flipped labels, forcing the model to override semantic priors in order to follow the in-context examples. SUL-ICL uses labels that are not semantically related to the task, which means that models must learn input-label mappings in order to perform the task because they can no longer rely on the semantics of natural language labels.

Experiment design

For a diverse dataset mixture, we experiment on seven natural language processing (NLP) tasks that have been widely used: sentiment analysis, subjective/objective classification, question classification, duplicated-question recognition, entailment recognition, financial sentiment analysis, and hate speech detection. We test five language model families, PaLM, Flan-PaLM, GPT-3, InstructGPT, and Codex.

Flipped labels

In this experiment, labels of in-context examples are flipped, meaning that prior knowledge and input-label mappings disagree (e.g., sentences containing positive sentiment labeled as “negative sentiment”), thereby allowing us to study whether models can override their priors. In this setting, models that are able to override prior knowledge and learn input-label mappings in-context should experience a decrease in performance (since ground-truth evaluation labels are not flipped).

The ability to override semantic priors when presented with flipped in-context example labels emerges with model scale. Smaller models cannot flip predictions to follow flipped labels (performance only decreases slightly), while larger models can do so (performance decreases to well below 50%).

We found that when no labels are flipped, larger models have better performance than smaller models (as expected). But when we flip more and more labels, the performance of small models stays relatively flat, but large models experience large performance drops to well-below random guessing (e.g., 90% → 22.5% for code-davinci-002).

These results indicate that large models can override prior knowledge from pre-training when contradicting input-label mappings are presented in-context. Small models can’t do this, making this ability an emergent phenomena of model scale.

Semantically-unrelated labels

In this experiment, we replace labels with semantically-irrelevant ones (e.g., for sentiment analysis, we use “foo/bar” instead of “negative/positive”), which means that the model can only perform ICL by learning from input-label mappings. If a model mostly relies on prior knowledge for ICL, then its performance should decrease after this change since it will no longer be able to use semantic meanings of labels to make predictions. A model that can learn input–label mappings in-context, on the other hand, would be able to learn these semantically-unrelated mappings and should not experience a major drop in performance.

Small models rely more on semantic priors than large models do, as indicated by the greater decrease in performance for small models than for large models when using semantically-unrelated labels (i.e., targets) instead of natural language labels. For each plot, models are shown in order of increasing model size (e.g., for GPT-3 models, a is smaller than b, which is smaller than c).

Indeed, we see that using semantically-unrelated labels results in a greater performance drop for small models. This suggests that smaller models primarily rely on their semantic priors for ICL rather than learning from the presented input-label mappings. Large models, on the other hand, have the ability to learn input-label mappings in-context when the semantic nature of labels is removed.

We also find that including more in-context examples (i.e., exemplars) results in a greater performance improvement for large models than it does for small models, indicating that large models are better at learning from in-context examples than small models are.

In the SUL-ICL setup, larger models benefit more from additional examples than smaller models do.

Instruction tuning

Instruction tuning is a popular technique for improving model performance, which involves tuning models on various NLP tasks that are phrased as instructions (e.g., “Question: What is the sentiment of the following sentence, ‘This movie is great.’ Answer: Positive”). Since the process uses natural language labels, however, an open question is whether it improves the ability to learn input-label mappings or whether it strengthens the ability to recognize and apply semantic prior knowledge. Both of these would lead to an improvement in performance on standard ICL tasks, so it’s unclear which of these occur.

We study this question by running the same two setups as before, only this time we focus on comparing standard language models (specifically, PaLM) with their instruction-tuned variants (Flan-PaLM).

First, we find that Flan-PaLM is better than PaLM when we use semantically-unrelated labels. This effect is very prominent in small models, as Flan-PaLM-8B outperforms PaLM-8B by 9.6% and almost catches up to PaLM-62B. This trend suggests that instruction tuning strengthens the ability to learn input-label mappings, which isn’t particularly surprising.

Instruction-tuned language models are better at learning input–label mappings than pre-training–only language models are.

More interestingly, we saw that Flan-PaLM is actually worse than PaLM at following flipped labels, meaning that the instruction tuned models were unable to override their prior knowledge (Flan-PaLM models don’t reach below random guessing with 100% flipped labels, but PaLM models without instruction tuning can reach 31% accuracy in the same setting). These results indicate that instruction tuning must increase the extent to which models rely on semantic priors when they’re available.

Instruction-tuned models are worse than pre-training–only models at learning to override semantic priors when presented with flipped labels in-context.

Combined with the previous result, we conclude that although instruction tuning improves the ability to learn input-label mappings, it strengthens the usage of semantic prior knowledge more.

Conclusion

We examined the extent to which language models learn in-context by utilizing prior knowledge learned during pre-training versus input-label mappings presented in-context.

We first showed that large language models can learn to override prior knowledge when presented with enough flipped labels, and that this ability emerges with model scale. We then found that successfully doing ICL using semantically-unrelated labels is another emergent ability of model scale. Finally, we analyzed instruction-tuned language models and saw that instruction tuning improves the capacity to learn input-label mappings but also strengthens the use of semantic prior knowledge even more.

Future work

These results underscore how the ICL behavior of language models can change depending on their scale, and that larger language models have an emergent ability to map inputs to many types of labels, a form of reasoning in which input-label mappings can potentially be learned for arbitrary symbols. Future research could help provide insights on why these phenomena occur with respect to model scale.

Read More

Posted by Candice Schumann, Software Engineer, and Gbolahan O. Olanubi, User Experience Researcher, Google Research

Skin tone is an observable characteristic that is subjective, perceived differently by individuals (e.g., depending on their location or culture) and thus is complicated to annotate. That said, the ability to reliably and accurately annotate skin tone is highly important in computer vision. This became apparent in 2018, when the Gender Shades study highlighted that computer vision systems struggled to detect people with darker skin tones, and performed particularly poorly for women with darker skin tones. The study highlights the importance for computer researchers and practitioners to evaluate their technologies across the full range of skin tones and at intersections of identities. Beyond evaluating model performance on skin tone, skin tone annotations enable researchers to measure diversity and representation in image retrieval systems, dataset collection, and image generation. For all of these applications, a collection of meaningful and inclusive skin tone annotations is key.

Monk Skin Tone (MST) Scale See more at skintone.google.

Last year, in a step toward more inclusive computer vision systems, Google’s Responsible AI and Human-Centered Technology team in Research partnered with Dr. Ellis Monk to openly release the Monk Skin Tone (MST) Scale, a skin tone scale that captures a broad spectrum of skin tones. In comparison to an industry standard scale like the Fitzpatrick Skin-Type Scale designed for dermatological use, the MST offers a more inclusive representation across the range of skin tones and was designed for a broad range of applications, including computer vision.

Today we’re announcing the Monk Skin Tone Examples (MST-E) dataset to help practitioners understand the MST scale and train their human annotators. This dataset has been made publicly available to enable practitioners everywhere to create more consistent, inclusive, and meaningful skin tone annotations. Along with this dataset, we’re providing a set of recommendations, noted below, around the MST scale and MST-E dataset so we can all create products that work well for all skin tones.

Since we launched the MST, we’ve been using it to improve Google’s computer vision systems to make equitable image tools for everyone and to improve representation of skin tone in Search. Computer vision researchers and practitioners outside of Google, like the curators of MetaAI’s Casual Conversations dataset, are recognizing the value of MST annotations to provide additional insight into diversity and representation in datasets. Incorporation into widely available datasets like these are essential to give everyone the ability to ensure they are building more inclusive computer vision technologies and can test the quality of their systems and products across a wide range of skin tones.

Our team has continued to conduct research to understand how we can continue to advance our understanding of skin tone in computer vision. One of our core areas of focus has been skin tone annotation, the process by which human annotators are asked to review images of people and select the best representation of their skin tone. MST annotations enable a better understanding of the inclusiveness and representativeness of datasets across a wide range of skin tones, thus enabling researchers and practitioners to evaluate quality and fairness of their datasets and models. To better understand the effectiveness of MST annotations, we’ve asked ourselves the following questions:

• How do people think about skin tone across geographic locations?
• What does global consensus of skin tone look like?
• How do we effectively annotate skin tone for use in inclusive machine learning (ML)?

The MST-E dataset

The MST-E dataset contains 1,515 images and 31 videos of 19 subjects spanning the 10 point MST scale, where the subjects and images were sourced through TONL, a stock photography company focusing on diversity. The 19 subjects include individuals of different ethnicities and gender identities to help human annotators decouple the concept of skin tone from race. The primary goal of this dataset is to enable practitioners to train their human annotators and test for consistent skin tone annotations across various environment capture conditions.

The MST-E image set contains 1,515 images and 31 videos featuring 19 models taken under various lighting conditions and facial expressions. Images by TONL. Copyright TONL.CO 2022 ALL RIGHTS RESERVED. Used with permission.

All images of a subject were collected in a single day to reduce variation of skin tone due to seasonal or other temporal effects. Each subject was photographed in various poses, facial expressions, and lighting conditions. In addition, Dr. Monk annotated each subject with a skin tone label and then selected a “golden” image for each subject that best represents their skin tone. In our research we compare annotations made by human annotators to those made by Dr. Monk, an academic expert in social perception and inequality.

Terms of use

Each model selected as a subject provided consent for their images and videos to be released. TONL has given permission for these images to be released as part of MST-E and used for research or human-annotator-training purposes only. The images are not to be used to train ML models.

Challenges with forming consensus of MST annotations

Although skin tone is easy for a person to see, it can be challenging to systematically annotate across multiple people due to issues with technology and the complexity of human social perception.

On the technical side, things like the pixelation, lighting conditions of an image, or a person’s monitor settings can affect how skin tone appears on a screen. You might notice this yourself the next time you change the display setting while watching a show. The hue, saturation, and brightness could all affect how skin tone is displayed on a monitor. Despite these challenges, we find that human annotators are able to learn to become invariant to lighting conditions of an image when annotating skin tone.

On the social perception side, aspects of a person’s life like their location, culture, and lived experience may affect how they annotate various skin tones. We found some evidence for this when we asked photographers in the United States and photographers in India to annotate the same image. The photographers in the United States viewed this person as somewhere between MST-5 & MST-7. However, the photographers in India viewed this person as somewhere between MST-3 & MST-5.

The distribution of Monk Skin Tone Scale annotations for this image from a sample of 5 photographers in the U.S. and 5 photographers in India.

Continuing this exploration, we asked trained annotators from five different geographical regions (India, Philippines, Brazil, Hungary, and Ghana) to annotate skin tone on the MST scale. Within each market each image had 5 annotators who were drawn from a broader pool of annotators in that region. For example, we could have 20 annotators in a market, and select 5 to review a particular image.

With these annotations we found two important details. First, annotators within a region had similar levels of agreement on a single image. Second, annotations between regions were, on average, significantly different from each other. (p<0.05). This suggests that people from the same geographic region may have a similar mental model of skin tone, but this mental model is not universal.

However, even with these regional differences, we also find that the consensus between all five regions falls close to the MST values supplied by Dr. Monk. This suggests that a geographically diverse group of annotators can get close to the MST value annotated by an MST expert. In addition, after training, we find no significant difference between annotations on well-lit images, versus poorly-lit images, suggesting that annotators can become invariant to different lighting conditions in an image — a non-trivial task for ML models.

The MST-E dataset allows researchers to study annotator behavior across curated subsets controlling for potential confounders. We observed similar regional variation when annotating much larger datasets with many more subjects.

Skin Tone annotation recommendations

Our research includes four major findings. First, annotators within a similar geographical region have a consistent and shared mental model of skin tone. Second, these mental models differ across different geographical regions. Third, the MST annotation consensus from a geographically diverse set of annotators aligns with the annotations provided by an expert in social perception and inequality. And fourth, annotators can learn to become invariant to lighting conditions when annotating MST.

Given our research findings, there are a few recommendations for skin tone annotation when using the MST.

1. Having a geographically diverse set of annotators is important to gain accurate, or close to ground truth, estimates of skin tone.
2. Train human annotators using the MST-E dataset, which spans the entire MST spectrum and contains images in a variety of lighting conditions. This will help annotators become invariant to lighting conditions and appreciate the nuance and differences between the MST points.
3. Given the wide range of annotations we suggest having at least two annotators in at least five different geographical regions (10 ratings per image).

Skin tone annotation, like other subjective annotation tasks, is difficult but possible. These types of annotations allow for a more nuanced understanding of model performance, and ultimately help us all to create products that work well for every person across the broad and diverse spectrum of skin tones.

Acknowledgements

We wish to thank our colleagues across Google working on fairness and inclusion in computer vision for their contributions to this work, especially Marco Andreetto, Parker Barnes, Ken Burke, Benoit Corda, Tulsee Doshi, Courtney Heldreth, Rachel Hornung, David Madras, Ellis Monk, Shrikanth Narayanan, Utsav Prabhu, Susanna Ricco, Sagar Savla, Alex Siegman, Komal Singh, Biao Wang, and Auriel Wright. We also would like to thank Annie Jean-Baptiste, Florian Koenigsberger, Marc Repnyek, Maura O’Brien, and Dominique Mungin and the rest of the team who help supervise, fund, and coordinate our data collection.

Read More

Posted by Weicheng Kuo and Anelia Angelova, Research Scientists, Google Research

Detection is a fundamental vision task that aims to localize and recognize objects in an image. However, the data collection process of manually annotating bounding boxes or instance masks is tedious and costly, which limits the modern detection vocabulary size to roughly 1,000 object classes. This is orders of magnitude smaller than the vocabulary people use to describe the visual world and leaves out many categories. Recent vision and language models (VLMs), such as CLIP, have demonstrated improved open-vocabulary visual recognition capabilities through learning from Internet-scale image-text pairs. These VLMs are applied to zero-shot classification using frozen model weights without the need for fine-tuning, which stands in stark contrast to the existing paradigms used for retraining or fine-tuning VLMs for open-vocabulary detection tasks.

Intuitively, to align the image content with the text description during training, VLMs may learn region-sensitive and discriminative features that are transferable to object detection. Surprisingly, features of a frozen VLM contain rich information that are both region sensitive for describing object shapes (second column below) and discriminative for region classification (third column below). In fact, feature grouping can nicely delineate object boundaries without any supervision. This motivates us to explore the use of frozen VLMs for open-vocabulary object detection with the goal to expand detection beyond the limited set of annotated categories.

We explore the potential of frozen vision and language features for open-vocabulary detection. The K-Means feature grouping reveals rich semantic and region-sensitive information where object boundaries are nicely delineated (column 2). The same frozen features can classify groundtruth (GT) regions well without fine-tuning (column 3).

In “F-VLM: Open-Vocabulary Object Detection upon Frozen Vision and Language Models”, presented at ICLR 2023, we introduce a simple and scalable open-vocabulary detection approach built upon frozen VLMs. F-VLM reduces the training complexity of an open-vocabulary detector to below that of a standard detector, obviating the need for knowledge distillation, detection-tailored pre-training, or weakly supervised learning. We demonstrate that by preserving the knowledge of pre-trained VLMs completely, F-VLM maintains a similar philosophy to ViTDet and decouples detector-specific learning from the more task-agnostic vision knowledge in the detector backbone. We are also releasing the F-VLM code along with a demo on our project page.

Learning upon frozen vision and language models

We desire to retain the knowledge of pretrained VLMs as much as possible with a view to minimize effort and cost needed to adapt them for open-vocabulary detection. We use a frozen VLM image encoder as the detector backbone and a text encoder for caching the detection text embeddings of offline dataset vocabulary. We take this VLM backbone and attach a detector head, which predicts object regions for localization and outputs detection scores that indicate the probability of a detected box being of a certain category. The detection scores are the cosine similarity of region features (a set of bounding boxes that the detector head outputs) and category text embeddings. The category text embeddings are obtained by feeding the category names through the text model of pretrained VLM (which has both image and text models)r.

The VLM image encoder consists of two parts: 1) a feature extractor and 2) a feature pooling layer. We adopt the feature extractor for detector head training, which is the only step we train (on standard detection data), to allow us to directly use frozen weights, inheriting rich semantic knowledge (e.g., long-tailed categories like martini, fedora hat, pennant) from the VLM backbone. The detection losses include box regression and classification losses.

At training time, F-VLM is simply a detector with the last classification layer replaced by base-category text embeddings.

Region-level open-vocabulary recognition

The ability to perform open-vocabulary recognition at region level (i.e., bounding box level as opposed to image level) is integral to F-VLM. Since the backbone features are frozen, they do not overfit to the training categories (e.g., donut, zebra) and can be directly cropped for region-level classification. F-VLM performs this open-vocabulary classification only at test time. To obtain the VLM features for a region, we apply the feature pooling layer on the cropped backbone output features. Because the pooling layer requires fixed-size inputs, e.g., 7×7 for ResNet50 (R50) CLIP backbone, we crop and resize the region features with the ROI-Align layer (shown below). Unlike existing open-vocabulary detection approaches, we do not crop and resize the RGB image regions and cache their embeddings in a separate offline process, but train the detector head in one stage. This is simpler and makes more efficient use of disk storage space.. In addition, we do not crop VLM region features during training because the backbone features are frozen.

Despite never being trained on regions, the cropped region features maintain good open-vocabulary recognition capability. However, we observe the cropped region features are not sensitive enough to the localization quality of the regions, i.e., a loosely vs. tightly localized box both have similar features. This may be good for classification, but is problematic for detection because we need the detection scores to reflect localization quality as well. To remedy this, we apply the geometric mean to combine the VLM scores with the detection scores for each region and category. The VLM scores indicate the probability of a detection box being of a certain category according to the pretrained VLM. The detection scores indicate the class probability distribution of each box based on the similarity of region features and input text embeddings.

At test time, F-VLM uses the region proposals to crop out the top-level features of the VLM backbone and compute the VLM score per region. The trained detector head provides the detection boxes and masks, while the final detection scores are a combination of detection and VLM scores.

Evaluation

We apply F-VLM to the popular LVIS open-vocabulary detection benchmark. At the system-level, the best F-VLM achieves 32.8 average precision (AP) on rare categories (APr), which outperforms the state of the art by 6.5 mask APr and many other approaches based on knowledge distillation, pre-training, or joint training with weak supervision. F-VLM shows strong scaling property with frozen model capacity, while the number of trainable parameters is fixed. Moreover, F-VLM generalizes and scales well in the transfer detection tasks (e.g., Objects365 and Ego4D datasets) by simply replacing the vocabularies without fine-tuning the model. We test the LVIS-trained models on the popular Objects365 datasets and demonstrate that the model can work very well without training on in-domain detection data.

F-VLM outperforms the state of the art (SOTA) on LVIS open-vocabulary detection benchmark and transfer object detection. On the x-axis, we show the LVIS metric mask AP on rare categories (APr), and the Objects365 (O365) metric box AP on all categories. The sizes of the detector backbones are as follows: Small(R50), Base (R50x4), Large(R50x16), Huge(R50x64). The naming follows CLIP convention.

We visualize F-VLM on open-vocabulary detection and transfer detection tasks (shown below). On LVIS and Objects365, F-VLM correctly detects both novel and common objects. A key benefit of open-vocabulary detection is to test on out-of-distribution data with categories given by users on the fly. See the F-VLM paper for more visualization on LVIS, Objects365 and Ego4D datasets.

F-VLM open-vocabulary and transfer detections. Top: Open-vocabulary detection on LVIS. We only show the novel categories for clarity. Bottom: Transfer to Objects365 dataset shows accurate detection of many categories. Novel categories detected: fedora, martini, pennant, football helmet (LVIS); slide (Objects365).

Training efficiency

We show that F-VLM can achieve top performance with much less computational resources in the table below. Compared to the state-of-the-art approach, F-VLM can achieve better performance with 226x fewer resources and 57x faster wall clock time. Apart from training resource savings, F-VLM has potential for substantial memory savings at training time by running the backbone in inference mode. The F-VLM system runs almost as fast as a standard detector at inference time, because the only addition is a single attention pooling layer on the detected region features.

We provide additional results using the shorter Detectron2 training recipes (12 and 36 epochs), and show similarly strong performance by using a frozen backbone. The default setting is marked in gray.

Conclusion

We present F-VLM – a simple open-vocabulary detection method which harnesses the power of frozen pre-trained large vision-language models to provide detection of novel objects. This is done without a need for knowledge distillation, detection-tailored pre-training, or weakly supervised learning. Our approach offers significant compute savings and obviates the need for image-level labels. F-VLM achieves the new state-of-the-art in open-vocabulary detection on the LVIS benchmark at system level, and shows very competitive transfer detection on other datasets. We hope this study can both facilitate further research in novel-object detection and help the community explore frozen VLMs for a wider range of vision tasks.

Acknowledgements

This work is conducted by Weicheng Kuo, Yin Cui, Xiuye Gu, AJ Piergiovanni, and Anelia Angelova. We would like to thank our colleagues at Google Research for their advice and helpful discussions.

Read More

Posted by Bryan Wang, Student Researcher, and Yang Li, Research Scientist, Google Research

Intelligent assistants on mobile devices have significantly advanced language-based interactions for performing simple daily tasks, such as setting a timer or turning on a flashlight. Despite the progress, these assistants still face limitations in supporting conversational interactions in mobile user interfaces (UIs), where many user tasks are performed. For example, they cannot answer a user’s question about specific information displayed on a screen. An agent would need to have a computational understanding of graphical user interfaces (GUIs) to achieve such capabilities.

Prior research has investigated several important technical building blocks to enable conversational interaction with mobile UIs, including summarizing a mobile screen for users to quickly understand its purpose, mapping language instructions to UI actions and modeling GUIs so that they are more amenable for language-based interaction. However, each of these only addresses a limited aspect of conversational interaction and requires considerable effort in curating large-scale datasets and training dedicated models. Furthermore, there is a broad spectrum of conversational interactions that can occur on mobile UIs. Therefore, it is imperative to develop a lightweight and generalizable approach to realize conversational interaction.

In “Enabling Conversational Interaction with Mobile UI using Large Language Models”, presented at CHI 2023, we investigate the viability of utilizing large language models (LLMs) to enable diverse language-based interactions with mobile UIs. Recent pre-trained LLMs, such as PaLM, have demonstrated abilities to adapt themselves to various downstream language tasks when being prompted with a handful of examples of the target task. We present a set of prompting techniques that enable interaction designers and developers to quickly prototype and test novel language interactions with users, which saves time and resources before investing in dedicated datasets and models. Since LLMs only take text tokens as input, we contribute a novel algorithm that generates the text representation of mobile UIs. Our results show that this approach achieves competitive performance using only two data examples per task. More broadly, we demonstrate LLMs’ potential to fundamentally transform the future workflow of conversational interaction design.

Animation showing our work on enabling various conversational interactions with mobile UI using LLMs.

Prompting LLMs with UIs

LLMs support in-context few-shot learning via prompting — instead of fine-tuning or re-training models for each new task, one can prompt an LLM with a few input and output data exemplars from the target task. For many natural language processing tasks, such as question-answering or translation, few-shot prompting performs competitively with benchmark approaches that train a model specific to each task. However, language models can only take text input, while mobile UIs are multimodal, containing text, image, and structural information in their view hierarchy data (i.e., the structural data containing detailed properties of UI elements) and screenshots. Moreover, directly inputting the view hierarchy data of a mobile screen into LLMs is not feasible as it contains excessive information, such as detailed properties of each UI element, which can exceed the input length limits of LLMs.

To address these challenges, we developed a set of techniques to prompt LLMs with mobile UIs. We contribute an algorithm that generates the text representation of mobile UIs using depth-first search traversal to convert the Android UI’s view hierarchy into HTML syntax. We also utilize chain of thought prompting, which involves generating intermediate results and chaining them together to arrive at the final output, to elicit the reasoning ability of the LLM.

Animation showing the process of few-shot prompting LLMs with mobile UIs.

Our prompt design starts with a preamble that explains the prompt’s purpose. The preamble is followed by multiple exemplars consisting of the input, a chain of thought (if applicable), and the output for each task. Each exemplar’s input is a mobile screen in the HTML syntax. Following the input, chains of thought can be provided to elicit logical reasoning from LLMs. This step is not shown in the animation above as it is optional. The task output is the desired outcome for the target tasks, e.g., a screen summary or an answer to a user question. Few-shot prompting can be achieved with more than one exemplar included in the prompt. During prediction, we feed the model the prompt with a new input screen appended at the end.

Experiments

We conducted comprehensive experiments with four pivotal modeling tasks: (1) screen question-generation, (2) screen summarization, (3) screen question-answering, and (4) mapping instruction to UI action. Experimental results show that our approach achieves competitive performance using only two data examples per task.

Task 1: Screen question generation

Given a mobile UI screen, the goal of screen question-generation is to synthesize coherent, grammatically correct natural language questions relevant to the UI elements requiring user input.

We found that LLMs can leverage the UI context to generate questions for relevant information. LLMs significantly outperformed the heuristic approach (template-based generation) regarding question quality.

Example screen questions generated by the LLM. The LLM can utilize screen contexts to generate grammatically correct questions relevant to each input field on the mobile UI, while the template approach falls short.

We also revealed LLMs’ ability to combine relevant input fields into a single question for efficient communication. For example, the filters asking for the minimum and maximum price were combined into a single question: “What’s the price range?

We observed that the LLM could use its prior knowledge to combine multiple related input fields to ask a single question.

In an evaluation, we solicited human ratings on whether the questions were grammatically correct (Grammar) and relevant to the input fields for which they were generated (Relevance). In addition to the human-labeled language quality, we automatically examined how well LLMs can cover all the elements that need to generate questions (Coverage F1). We found that the questions generated by LLM had almost perfect grammar (4.98/5) and were highly relevant to the input fields displayed on the screen (92.8%). Additionally, LLM performed well in terms of covering the input fields comprehensively (95.8%).

Task 2: Screen summarization

Screen summarization is the automatic generation of descriptive language overviews that cover essential functionalities of mobile screens. The task helps users quickly understand the purpose of a mobile UI, which is particularly useful when the UI is not visually accessible.

Our results showed that LLMs can effectively summarize the essential functionalities of a mobile UI. They can generate more accurate summaries than the Screen2Words benchmark model that we previously introduced using UI-specific text, as highlighted in the colored text and boxes below.

Example summary generated by 2-shot LLM. We found the LLM is able to use specific text on the screen to compose more accurate summaries.

Interestingly, we observed LLMs using their prior knowledge to deduce information not presented in the UI when creating summaries. In the example below, the LLM inferred the subway stations belong to the London Tube system, while the input UI does not contain this information.

LLM uses its prior knowledge to help summarize the screens.

Human evaluation rated LLM summaries as more accurate than the benchmark, yet they scored lower on metrics like BLEU. The mismatch between perceived quality and metric scores echoes recent work showing LLMs write better summaries despite automatic metrics not reflecting it.

Task 3: Screen question-answering

Given a mobile UI and an open-ended question asking for information regarding the UI, the model should provide the correct answer. We focus on factual questions, which require answers based on information presented on the screen.

Example results from the screen QA experiment. The LLM significantly outperforms the off-the-shelf QA baseline model.

We report performance using four metrics: Exact Matches (identical predicted answer to ground truth), Contains GT (answer fully containing ground truth), Sub-String of GT (answer is a sub-string of ground truth), and the Micro-F1 score based on shared words between the predicted answer and ground truth across the entire dataset.

Our results showed that LLMs can correctly answer UI-related questions, such as “what’s the headline?”. The LLM performed significantly better than baseline QA model DistillBERT, achieving a 66.7% fully correct answer rate. Notably, the 0-shot LLM achieved an exact match score of 30.7%, indicating the model’s intrinsic question answering capability.

Task 4: Mapping instruction to UI action

Given a mobile UI screen and natural language instruction to control the UI, the model needs to predict the ID of the object to perform the instructed action. For example, when instructed with “Open Gmail,” the model should correctly identify the Gmail icon on the home screen. This task is useful for controlling mobile apps using language input such as voice access. We introduced this benchmark task previously.

Example using data from the PixelHelp dataset. The dataset contains interaction traces for common UI tasks such as turning on wifi. Each trace contains multiple steps and corresponding instructions.

We assessed the performance of our approach using the Partial and Complete metrics from the Seq2Act paper. Partial refers to the percentage of correctly predicted individual steps, while Complete measures the portion of accurately predicted entire interaction traces. Although our LLM-based method did not surpass the benchmark trained on massive datasets, it still achieved remarkable performance with just two prompted data examples.

Takeaways and conclusion

Our study shows that prototyping novel language interactions on mobile UIs can be as easy as designing a data exemplar. As a result, an interaction designer can rapidly create functioning mock-ups to test new ideas with end users. Moreover, developers and researchers can explore different possibilities of a target task before investing significant efforts into developing new datasets and models.

We investigated the feasibility of prompting LLMs to enable various conversational interactions on mobile UIs. We proposed a suite of prompting techniques for adapting LLMs to mobile UIs. We conducted extensive experiments with the four important modeling tasks to evaluate the effectiveness of our approach. The results showed that compared to traditional machine learning pipelines that consist of expensive data collection and model training, one could rapidly realize novel language-based interactions using LLMs while achieving competitive performance.

Acknowledgements

We thank our paper co-author Gang Li, and appreciate the discussions and feedback from our colleagues Chin-Yi Cheng, Tao Li, Yu Hsiao, Michael Terry and Minsuk Chang. Special thanks to Muqthar Mohammad and Ashwin Kakarla for their invaluable assistance in coordinating data collection. We thank John Guilyard for helping create animations and graphics in the blog.

Read More

Google I/O 2023 was filled with news and launches — here are 100 things announced at I/O.Read More

Project Gameface, a new open-source hands-free gaming mouse has the potential to make gaming more accessible.Read More

Just in time for Google I/O 2023, try out I/O FLIP, an online card game built with generative AI.Read More