Yang Li
Yang Li is a Staff Research Scientist at Google, and an affiliate faculty member in Computer Science & Engineering at the University of Washington. Yang’s research focuses on the intersection between deep learning and human computer interaction.
See Yang Li's personal website.
Authored Publications
Sort By
PaLI-X: On Scaling up a Multilingual Vision and Language Model
Josip Djolonga
Piotr Padlewski
Basil Mustafa
Carlos Riquelme
Sebastian Goodman
Yi Tay
Siamak Shakeri
Daniel Salz
Michael Tschannen
Mandar Joshi
Filip Pavetić
Anurag Arnab
Yuanzhong Xu
Keran Rong
Computer Vision and Pattern Recognition Conference (CVPR)(2024)
Preview abstract
We explore the boundaries of scaling up a multilingual vision and language model, both in terms of size of the components and the breadth of its training task mixture. Our model achieves new levels of performance on a wide-range of varied and complex tasks, including multiple image-based captioning and question-answering tasks, image-based document understanding and few-shot (in-context) learning, as well as object detection, video question answering, and video captioning. Our model advances the state-of-the-art on most vision-and-language benchmarks considered (20+ of them). Finally, we observe emerging capabilities, such as complex counting and multilingual object detection, tasks that are not explicitly in the training mix.
View details
Rich Human Feedback for Text to Image Generation
Katherine Collins
Nicholas Carolan
Youwei Liang
Peizhao Li
Dj Dvijotham
Sarah Young
Jiao Sun
Arseniy Klimovskiy
Preview abstract
Recent Text-to-Image (T2I) generation models such as Stable Diffusion and Imagen have made significant progress in generating high-resolution images based on text descriptions. However, many generated images still suffer from issues such as artifacts/implausibility, misalignment with text descriptions, and low aesthetic quality.
Inspired by the success of Reinforcement Learning with Human Feedback (RLHF) for large language models, prior work collected human-provided scores as feedback on generated images and trained a reward model to improve the T2I generation.
In this paper, we enrich the feedback signal by (i) marking image regions that are implausible or misaligned with the text, and (ii) annotating which keywords in the text prompt are not represented in the image.
We collect such rich human feedback on 18K generated images and train a multimodal transformer to predict these rich feedback automatically.
We show that the predicted rich human feedback can be leveraged to improve image generation, for example, by selecting high-quality training data to finetune and improve the generative models, or by creating masks with predicted heatmaps to inpaint the problematic regions.
Notably, the improvements generalize to models (Muse) beyond those used to generate the images on which human feedback data were collected (Stable Diffusion variants).
View details
Towards Semantically-Aware UI Design Tools: Design, Implementation, and Evaluation of Semantic Grouping Guidelines
Peitong Duan
Bjoern Hartmann
Karina Nguyen
Marti Hearst
ICML 2023 Workshop on Artificial Intelligence and Human-Computer Interaction(2023)
Preview abstract
A coherent semantic structure, where semantically-related elements are appropriately grouped, is critical for proper understanding of a UI. Ideally, UI design tools should help designers establish coherent semantic grouping. To work towards this, we contribute five semantic grouping guidelines that capture how human designers think about semantic grouping and are amenable to implementation in design tools. They were obtained from empirical observations
on existing UIs, a literature review, and iterative refinement with UI experts’ feedback. We validated our guidelines through an expert review and heuristic evaluation; results indicate these guidelines capture valuable information about semantic structure. We demonstrate the guidelines’ use for building systems by implementing a set of computational metrics. These metrics detected many of the same severe issues that human design experts marked in a comparative study. Running our metrics on a larger UI dataset suggests many real UIs exhibit grouping violations.
View details
Preview abstract
Conversational agents show the promise to allow users to interact with mobile devices using language. However, to perform diverse UI tasks with natural language, developers typically need to create separate datasets and models for each specific task, which is expensive and effort-consuming. Recently, pre-trained large language models (LLMs) have been shown capable of generalizing to various downstream tasks when prompted with a handful of examples from the target task. This paper investigates the feasibility of enabling versatile conversational interactions with mobile UIs using a single LLM. We designed prompting techniques to adapt an LLM to mobile UIs. We experimented with four important modeling tasks that address various scenarios in conversational interaction. Our method achieved competitive performance on these challenging tasks without requiring dedicated datasets and training, offering a lightweight and generalizable approach to enable language-based mobile interaction.
View details
Preview abstract
Mobile UI understanding is important for enabling various interaction tasks such as UI automation and accessibility. Previous mobile UI modeling often depends on the view hierarchy information of a screen, which directly provides the structural data of the UI, with the hope to bypass challenging tasks of visual modeling from screen pixels. However, view hierarchies are not always available, and are often corrupted with missing object descriptions or misaligned structure information. As a result, despite the use of view hierarchies could offer short-term gains, it may ultimately hinder the applicability and performance of the model. In this paper, we propose Spotlight, a vision-only approach for mobile UI understanding. Specifically, we enhance a vision-language model that only takes the screenshot of the UI and a region of interest on the screen---the focus---as the input. This general architecture of Spotlight is easily scalable and capable of performing a range of UI modeling tasks. Our experiments show that our model establishes SoTA results on several representative UI tasks and outperforms previous methods that use both screenshots and view hierarchies as inputs. Furthermore, we explore multi-task learning and few-shot prompting capacities of the proposed models, demonstrating promising results in the multi-task learning direction.
View details
Preview abstract
User interface design is a complex task that involves designers examining a wide range of options. We present Spacewalker, a tool that allows designers to rapidly search a large design space for an optimal web UI with integrated support. Designers first annotate each attribute they want to explore in a typical HTML page, using a simple markup extension we designed. Spacewalker then parses the annotated HTML specification, and intelligently generates and distributes various configurations of the web UI to crowd workers for evaluation. We enhanced a genetic algorithm to accommodate crowd worker responses from pairwise comparison of UI designs, which is crucial for obtaining reliable feedback. Based on our experiments, Spacewalker allows designers to effectively search a large design space of a UI, using the language they are familiar with, and improve their design rapidly at a minimal cost.
View details
TapNet: The Design, Training, Implementation, and Applications of a Multi-Task Learning CNN for Off-Screen Mobile Input
Michael Xuelin Huang
Nazneen Nazneen
Alex Chao
ACM CHI Conference on Human Factors in Computing Systems, ACM(2021)
Preview abstract
Off-screen interaction offers great potential for one-handed and eyes-free mobile interaction. While a few existing studies have explored the built-in mobile phone sensors to sense off-screen signals, none met practical requirement. This paper discusses the design, training, implementation and applications of TapNet, a multi-task network that detects tapping on the smartphone using built-in accelerometer and gyroscope. With sensor location as auxiliary information, TapNet can jointly learn from data across devices and simultaneously recognize multiple tap properties, including tap direction and tap location. We developed four datasets consisting of over 180K training samples, 38K testing samples, and 87 participants in total. Experimental evaluation demonstrated the effectiveness of the TapNet design and its significant improvement over the state of the art. Along with the datasets, codebase, and extensive experiments, TapNet establishes a new technical foundation for off-screen mobile input.
View details
Preview abstract
We present a new problem: grounding natural language instructions to mobile user interface actions, and create three new datasets for it. For full task evaluation, we create PixelHelp, a corpus that pairs English instructions with actions performed by people on a mobile UI emulator. To scale training, we decouple the language and action data by (a) annotating action phrase spans in How-To instructions and (b) synthesizing grounded descriptions of actions for mobile user interfaces. We use a Transformer to extract action phrase tuples from long-range natural language instructions. A grounding Transformer then contextually represents UI objects using both their content and screen position and connects them to object descriptions. Given a starting screen and instruction, our model achieves 70.59% accuracy on predicting complete ground-truth action sequences in PixelHelp.
View details
Preview abstract
Natural language descriptions of user interface (UI) elements such as alternative text are crucial for accessibility and language-based interaction in general. Yet, these descriptions are constantly missing in mobile UIs. We propose widget captioning, a novel task for automatically generating language descriptions for UI elements from multimodal input including both the image and the structural representations of user interfaces. We collected a largescale dataset for widget captioning with crowdsourcing. Our dataset contains 162,859 language phrases created by human workers for annotating 61,285 UI elements across 21,750 unique UI screens. We thoroughly analyze the dataset, and train and evaluate a set of deep model configurations to investigate how each feature modality as well as the choice of learning strategies impact the quality of predicted captions. The task formulation and the dataset as well as our benchmark models contribute a solid basis for this novel multimodal captioning task that connects language and user interfaces.
View details
Preview abstract
Neural language models have been widely used in various NLP tasks, including
machine translation, next word prediction and conversational agents. However, it
is challenging to deploy these models on mobile devices due to their slow prediction speed, where the bottleneck is to compute top candidates in the softmax
layer. In this paper, we introduce a novel softmax layer approximation algorithm
by exploiting the clustering structure of context vectors. Our algorithm uses a
light-weight screening model to predict a much smaller set of candidate words
based on the given context, and then conducts an exact softmax only within that
subset. Training such a procedure end-to-end is challenging as traditional clustering methods are discrete and non-differentiable, and thus unable to be used with
back-propagation in the training process. Using the Gumbel softmax, we are able
to train the screening model end-to-end on the training set to exploit data distribution. The algorithm achieves an order of magnitude faster inference than the
original softmax layer for predicting top-k words in various tasks such as beam
search in machine translation or next words prediction. For example, for machine translation task on German to English dataset with around 25K vocabulary,
we can achieve 20.4 times speed up with 98.9% precision@1 and 99.3% precision@5 with the original softmax layer prediction, while state-of-the-art (Zhang
et al., 2018) only achieves 6.7x speedup with 98.7% precision@1 and 98.1% precision@5 for the same task.
View details
