Kenz Kallal, Tomoka Kan, Eric Wang
The best previous lower bounds for kissing numbers in dimensions 25 through 31 were constructed using a set $S$ with $|S| = 480$ of minimal vectors of the Leech Lattice, $Λ_{24}$, such that $\langle x, y \rangle \leq 1$ for any distinct $x, y \in S$. Then, a probabilistic argument based on applying automorphisms of $Λ_{24}$ gives more disjoint sets $S_i$ of minimal vectors of $Λ_{24}$ with the same property. Cohn, Jiao, Kumar, and Torquato proved that these subsets give kissing configurations in dimensions 25 through 31 of given size linear in the sizes of the subsets. We achieve $|S| = 488$ by applying simulated annealing. We also improve the aforementioned probabilistic argument in the general case. Finally, we greedily construct even larger $S_i$'s given our $S$ of size $488$, giving increased lower bounds on kissing numbers in $\mathbb{R}^{25}$ through $\mathbb{R}^{31}$.
Eric Wang, Pasha Khosravi, Guy Van den Broeck
Understanding the behavior of learned classifiers is an important task, and various black-box explanations, logical reasoning approaches, and model-specific methods have been proposed. In this paper, we introduce probabilistic sufficient explanations, which formulate explaining an instance of classification as choosing the "simplest" subset of features such that only observing those features is "sufficient" to explain the classification. That is, sufficient to give us strong probabilistic guarantees that the model will behave similarly when all features are observed under the data distribution. In addition, we leverage tractable probabilistic reasoning tools such as probabilistic circuits and expected predictions to design a scalable algorithm for finding the desired explanations while keeping the guarantees intact. Our experiments demonstrate the effectiveness of our algorithm in finding sufficient explanations, and showcase its advantages compared to Anchors and logical explanations.
Wesley Pegden, Eric Wang
We introduce and study two Maker-Breaker-like games for constructing planar graphs: the edge drawing game, where two players take turns drawing non-intersecting edges between points in the plane, and the circle packing game, where the players take turns placing disjoint circles in the plane. Both games produce planar graphs: the edge drawing game results in a plane graph drawing, and the circle packing game yields a planar graph via the contact graph of the packing. For both games, we give necessary conditions under which a given planar graph can be constructed. We also show that the two games are indeed different by giving a class of graphs which can be constructed in one but not the other.
Yarden Katz, Eric T. Wang, Jacob Silterra, Schraga Schwartz, Bang Wong, Jill P. Mesirov, Edoardo M. Airoldi, Christopher B. Burge
Jun 14, 2013·q-bio.GN·PDF We introduce Sashimi plots, a quantitative multi-sample visualization of mRNA sequencing reads aligned to gene annotations. Sashimi plots are made using alignments (stored in the SAM/BAM format) and gene model annotations (in GFF format), which can be custom-made by the user or obtained from databases such as Ensembl or UCSC. We describe two implementations of Sashimi plots: (1) a stand-alone command line implementation aimed at making customizable publication quality figures, and (2) an implementation built into the Integrated Genome Viewer (IGV) browser, which enables rapid and dynamic creation of Sashimi plots for any genomic region of interest, suitable for exploratory analysis of alternatively spliced regions of the transcriptome. Isoform expression estimates outputted by the MISO program can be optionally plotted along with Sashimi plots. Sashimi plots can be used to quickly screen differentially spliced exons along genomic regions of interest and can be used in publication quality figures. The Sashimi plot software and documentation is available from: http://genes.mit.edu/burgelab/miso/docs/sashimi.html
Wanrong Zhu, Xin Eric Wang, Tsu-Jui Fu, An Yan, Pradyumna Narayana, Kazoo Sone, Sugato Basu, William Yang Wang
One of the most challenging topics in Natural Language Processing (NLP) is visually-grounded language understanding and reasoning. Outdoor vision-and-language navigation (VLN) is such a task where an agent follows natural language instructions and navigates a real-life urban environment. Due to the lack of human-annotated instructions that illustrate intricate urban scenes, outdoor VLN remains a challenging task to solve. This paper introduces a Multimodal Text Style Transfer (MTST) learning approach and leverages external multimodal resources to mitigate data scarcity in outdoor navigation tasks. We first enrich the navigation data by transferring the style of the instructions generated by Google Maps API, then pre-train the navigator with the augmented external outdoor navigation dataset. Experimental results show that our MTST learning approach is model-agnostic, and our MTST approach significantly outperforms the baseline models on the outdoor VLN task, improving task completion rate by 8.7% relatively on the test set.
An Yan, Xin Eric Wang, Jiangtao Feng, Lei Li, William Yang Wang
Commanding a robot to navigate with natural language instructions is a long-term goal for grounded language understanding and robotics. But the dominant language is English, according to previous studies on vision-language navigation (VLN). To go beyond English and serve people speaking different languages, we collect a bilingual Room-to-Room (BL-R2R) dataset, extending the original benchmark with new Chinese instructions. Based on this newly introduced dataset, we study how an agent can be trained on existing English instructions but navigate effectively with another language under a zero-shot learning scenario. Without any training data of the target language, our model shows competitive results even compared to a model with full access to the target language training data. Moreover, we investigate the transferring ability of our model when given a certain amount of target language training data.
Kenan Jiang, Xuehai He, Ruize Xu, Xin Eric Wang
Contrastive Language-Image Pretraining (CLIP) has demonstrated great zero-shot performance for matching images and text. However, it is still challenging to adapt vision-lanaguage pretrained models like CLIP to compositional image and text matching -- a more challenging image and text matching task requiring the model understanding of compositional word concepts and visual components. Towards better compositional generalization in zero-shot image and text matching, in this paper, we study the problem from a causal perspective: the erroneous semantics of individual entities are essentially confounders that cause the matching failure. Therefore, we propose a novel \textbf{\textit{training-free}} compositional CLIP model (ComCLIP). ComCLIP disentangles input images into subjects, objects, and action sub-images and composes CLIP's vision encoder and text encoder to perform evolving matching over compositional text embedding and sub-image embeddings. In this way, ComCLIP can mitigate spurious correlations introduced by the pretrained CLIP models and dynamically evaluate the importance of each component. Experiments on four compositional image-text matching datasets: SVO, ComVG, Winoground, and VL-checklist, and two general image-text retrieval datasets: Flick30K, and MSCOCO demonstrate the effectiveness of our plug-and-play method, which boosts the \textbf{\textit{zero-shot}} inference ability of CLIP, SLIP, and BLIP2 even without further training or fine-tuning. Our codes can be found at https://github.com/eric-ai-lab/ComCLIP.
Tsu-Jui Fu, Xin Eric Wang, Matthew Peterson, Scott Grafton, Miguel Eckstein, William Yang Wang
Vision-and-Language Navigation (VLN) is a task where agents must decide how to move through a 3D environment to reach a goal by grounding natural language instructions to the visual surroundings. One of the problems of the VLN task is data scarcity since it is difficult to collect enough navigation paths with human-annotated instructions for interactive environments. In this paper, we explore the use of counterfactual thinking as a human-inspired data augmentation method that results in robust models. Counterfactual thinking is a concept that describes the human propensity to create possible alternatives to life events that have already occurred. We propose an adversarial-driven counterfactual reasoning model that can consider effective conditions instead of low-quality augmented data. In particular, we present a model-agnostic adversarial path sampler (APS) that learns to sample challenging paths that force the navigator to improve based on the navigation performance. APS also serves to do pre-exploration of unseen environments to strengthen the model's ability to generalize. We evaluate the influence of APS on the performance of different VLN baseline models using the room-to-room dataset (R2R). The results show that the adversarial training process with our proposed APS benefits VLN models under both seen and unseen environments. And the pre-exploration process can further gain additional improvements under unseen environments.
Gonzalo Gonzalez-Pumariega, Saaket Agashe, Jiachen Yang, Ang Li, Xin Eric Wang
Computer-use agents have rapidly improved on real-world tasks such as web navigation, desktop automation, and software interaction, in some cases surpassing human performance. Yet even when the task and model are unchanged, an agent that succeeds once may fail on a repeated execution of the same task. This raises a fundamental question: if an agent can succeed at a task once, what prevents it from doing so reliably? In this work, we study the sources of unreliability in computer-use agents through three factors: stochasticity during execution, ambiguity in task specification, and variability in agent behavior. We analyze these factors on OSWorld using repeated executions of the same task together with paired statistical tests that capture task-level changes across settings. Our analysis shows that reliability depends on both how tasks are specified and how agent behavior varies across executions. These findings suggest the need to evaluate agents under repeated execution, to allow agents to resolve task ambiguity through interaction, and to favor strategies that remain stable across runs.
Saaket Agashe, Jiuzhou Han, Shuyu Gan, Jiachen Yang, Ang Li, Xin Eric Wang
We present Agent S, an open agentic framework that enables autonomous interaction with computers through a Graphical User Interface (GUI), aimed at transforming human-computer interaction by automating complex, multi-step tasks. Agent S aims to address three key challenges in automating computer tasks: acquiring domain-specific knowledge, planning over long task horizons, and handling dynamic, non-uniform interfaces. To this end, Agent S introduces experience-augmented hierarchical planning, which learns from external knowledge search and internal experience retrieval at multiple levels, facilitating efficient task planning and subtask execution. In addition, it employs an Agent-Computer Interface (ACI) to better elicit the reasoning and control capabilities of GUI agents based on Multimodal Large Language Models (MLLMs). Evaluation on the OSWorld benchmark shows that Agent S outperforms the baseline by 9.37% on success rate (an 83.6% relative improvement) and achieves a new state-of-the-art. Comprehensive analysis highlights the effectiveness of individual components and provides insights for future improvements. Furthermore, Agent S demonstrates broad generalizability to different operating systems on a newly-released WindowsAgentArena benchmark. Code available at https://github.com/simular-ai/Agent-S.
Xiangzhi Eric Wang, Zackary P. T. Sin, Ye Jia, Daniel Archer, Wynonna H. Y. Fong, Qing Li, Chen Li
In our daily lives, we can naturally convey instructions for the spatial manipulation of objects using words and gestures. Transposing this form of interaction into virtual reality (VR) object manipulation can be beneficial. We propose VR Mover, an LLM-empowered solution that can understand and interpret the user's vocal instruction to support object manipulation. By simply pointing and speaking, the LLM can manipulate objects without structured input. Our user study demonstrates that VR Mover enhances user usability, overall experience and performance on multi-object manipulation, while also reducing workload and arm fatigue. Users prefer the proposed natural interface for broad movements and may complementarily switch to gizmos or virtual hands for finer adjustments. These findings are believed to contribute to design implications for future LLM-based object manipulation interfaces, highlighting the potential for more intuitive and efficient user interactions in VR environments.
Qianqi Yan, Yue Fan, Hongquan Li, Shan Jiang, Yang Zhao, Xinze Guan, Ching-Chen Kuo, Xin Eric Wang
Existing Multimodal Large Language Models (MLLMs) are predominantly trained and tested on consistent visual-textual inputs, leaving open the question of whether they can handle inconsistencies in real-world, layout-rich content. To bridge this gap, we propose the Multimodal Inconsistency Reasoning (MMIR) benchmark to assess MLLMs' ability to detect and reason about semantic mismatches in artifacts such as webpages, presentation slides, and posters. MMIR comprises 534 challenging samples, each containing synthetically injected errors across five reasoning-heavy categories: Factual Contradiction, Identity Misattribution, Contextual Mismatch, Quantitative Discrepancy, and Temporal/Spatial Incoherence. We evaluate six state-of-the-art MLLMs, showing that models with dedicated multimodal reasoning capabilities, such as o1, substantially outperform their counterparts while open-source models remain particularly vulnerable to inconsistency errors. Detailed error analyses further show that models excel in detecting pairwise inconsistencies but struggle with inconsistencies confined to single elements in complex layouts. Probing experiments reveal that single-modality prompting, including Chain-of-Thought (CoT) and Set-of-Mark (SoM) methods, yields marginal gains, revealing a key bottleneck in cross-modal reasoning. Our findings highlight the need for advanced multimodal reasoning and point to future research on multimodal inconsistency.
Xiangzhi Eric Wang, Zackary P. T. Sin
3D Gaussian Splatting has made a marked impact on neural rendering by achieving impressive fidelity and performance. Despite this achievement, however, it is not readily applicable to developing interactive applications. Real-time applications like XR apps and games require functions such as animation, UV-mapping, and model editing simultaneously manipulated through the usage of a 3D model. We propose a modeling that is analogous to typical 3D models, which we call 3D Gaussian Model (3DGM); it provides a manipulatable proxy for novel animation and texture transfer. By binding the 3D Gaussians in texture space and re-projecting them back to world space through implicit shell mapping, we show how our 3D modeling can serve as a valid rendering methodology for interactive applications. It is further noted that recently, 3D mesh reconstruction works have been able to produce high-quality mesh for rendering. Our work, on the other hand, only requires an approximated geometry for rendering an object in high fidelity. Applicationwise, we will show that our proxy-based 3DGM is capable of driving novel animation without animated training data and texture transferring via UV mapping of the 3D Gaussians. We believe the result indicates the potential of our work for enabling interactive applications for 3D Gaussian Splatting.
Yunchao Zhang, Zonglin Di, Kaiwen Zhou, Cihang Xie, Xin Eric Wang
Federated embodied agent learning protects the data privacy of individual visual environments by keeping data locally at each client (the individual environment) during training. However, since the local data is inaccessible to the server under federated learning, attackers may easily poison the training data of the local client to build a backdoor in the agent without notice. Deploying such an agent raises the risk of potential harm to humans, as the attackers may easily navigate and control the agent as they wish via the backdoor. Towards Byzantine-robust federated embodied agent learning, in this paper, we study the attack and defense for the task of vision-and-language navigation (VLN), where the agent is required to follow natural language instructions to navigate indoor environments. First, we introduce a simple but effective attack strategy, Navigation as Wish (NAW), in which the malicious client manipulates local trajectory data to implant a backdoor into the global model. Results on two VLN datasets (R2R and RxR) show that NAW can easily navigate the deployed VLN agent regardless of the language instruction, without affecting its performance on normal test sets. Then, we propose a new Prompt-Based Aggregation (PBA) to defend against the NAW attack in federated VLN, which provides the server with a ''prompt'' of the vision-and-language alignment variance between the benign and malicious clients so that they can be distinguished during training. We validate the effectiveness of the PBA method on protecting the global model from the NAW attack, which outperforms other state-of-the-art defense methods by a large margin in the defense metrics on R2R and RxR.
Jialu Wang, Yang Liu, Xin Eric Wang
Internet search affects people's cognition of the world, so mitigating biases in search results and learning fair models is imperative for social good. We study a unique gender bias in image search in this work: the search images are often gender-imbalanced for gender-neutral natural language queries. We diagnose two typical image search models, the specialized model trained on in-domain datasets and the generalized representation model pre-trained on massive image and text data across the internet. Both models suffer from severe gender bias. Therefore, we introduce two novel debiasing approaches: an in-processing fair sampling method to address the gender imbalance issue for training models, and a post-processing feature clipping method base on mutual information to debias multimodal representations of pre-trained models. Extensive experiments on MS-COCO and Flickr30K benchmarks show that our methods significantly reduce the gender bias in image search models.
An Yan, Xin Eric Wang, Tsu-Jui Fu, William Yang Wang
Recent advances in language and vision push forward the research of captioning a single image to describing visual differences between image pairs. Suppose there are two images, I_1 and I_2, and the task is to generate a description W_{1,2} comparing them, existing methods directly model { I_1, I_2 } -> W_{1,2} mapping without the semantic understanding of individuals. In this paper, we introduce a Learning-to-Compare (L2C) model, which learns to understand the semantic structures of these two images and compare them while learning to describe each one. We demonstrate that L2C benefits from a comparison between explicit semantic representations and single-image captions, and generalizes better on the new testing image pairs. It outperforms the baseline on both automatic evaluation and human evaluation for the Birds-to-Words dataset.
Yue Fan, Handong Zhao, Ruiyi Zhang, Yu Shen, Xin Eric Wang, Gang Wu
Graphical User Interface (GUI) action grounding is a critical step in GUI automation that maps language instructions to actionable elements on GUI screens. Most recent works of GUI action grounding leverage large GUI datasets to fine-tune MLLMs. However, the fine-tuning data always covers limited GUI environments, and we find the performance of the resulting model deteriorates in novel environments. We argue that the GUI grounding models should be further aligned to the novel environments to reveal their full potential, when the inference is known to involve novel environments, i.e., environments not used during the previous fine-tuning. To realize this, we first propose GUI-Bee, an MLLM-based autonomous agent, to collect high-quality, environment-specific data through exploration and then continuously fine-tune GUI grounding models with the collected data. Our agent leverages a novel Q-value-Incentive In-Context Reinforcement Learning (Q-ICRL) method to optimize exploration efficiency and data quality. Additionally, we introduce NovelScreenSpot, a benchmark for testing how well the data can help align GUI action grounding models to novel environments and demonstrate the effectiveness of data collected by GUI-Bee in the experiments. Furthermore, we conduct an ablation study to validate the Q-ICRL method in enhancing the efficiency of GUI-Bee. Project page: https://gui-bee.github.io
Gonzalo Gonzalez-Pumariega, Vincent Tu, Chih-Lun Lee, Jiachen Yang, Ang Li, Xin Eric Wang
Computer-use agents (CUAs) hold promise for automating everyday digital tasks, but their performance on long-horizon, complex problems remains unreliable. Single-rollout execution is brittle, with small errors compounding over time and leading to high variance in outcomes. While prior work has attempted to scale within a single rollout, such approaches have yielded limited gains. Scaling over multiple rollouts offers a more promising alternative but doing so effectively is challenging due to the difficulty of evaluating and selecting among long-horizon agent behaviors. We introduce Behavior Judge (BJudge), which addresses this challenge by representing agent executions as behavior narratives and comparing candidate behaviors at this level, substantially improving robustness and success rates. Using multiple rollouts, BJudge establishes a new state of the art (SoTA) in OSWorld at 72.6%, significantly outperforming prior methods and surpassing human-level performance at 72.36%, with comprehensive ablations validating key design choices. We further demonstrate strong generalization results to different operating systems on WindowsAgentArena and AndroidWorld. Crucially, our results highlight the strong effectiveness of scaling CUAs, when you do it right: effective scaling requires structured trajectory understanding and selection, and BJudge provides a practical framework to achieve this.
Xin Eric Wang, Vihan Jain, Eugene Ie, William Yang Wang, Zornitsa Kozareva, Sujith Ravi
Recent research efforts enable study for natural language grounded navigation in photo-realistic environments, e.g., following natural language instructions or dialog. However, existing methods tend to overfit training data in seen environments and fail to generalize well in previously unseen environments. To close the gap between seen and unseen environments, we aim at learning a generalized navigation model from two novel perspectives: (1) we introduce a multitask navigation model that can be seamlessly trained on both Vision-Language Navigation (VLN) and Navigation from Dialog History (NDH) tasks, which benefits from richer natural language guidance and effectively transfers knowledge across tasks; (2) we propose to learn environment-agnostic representations for the navigation policy that are invariant among the environments seen during training, thus generalizing better on unseen environments. Extensive experiments show that environment-agnostic multitask learning significantly reduces the performance gap between seen and unseen environments, and the navigation agent trained so outperforms baselines on unseen environments by 16% (relative measure on success rate) on VLN and 120% (goal progress) on NDH. Our submission to the CVDN leaderboard establishes a new state-of-the-art for the NDH task on the holdout test set. Code is available at https://github.com/google-research/valan.
Saaket Agashe, Kyle Wong, Vincent Tu, Jiachen Yang, Ang Li, Xin Eric Wang
Computer use agents automate digital tasks by directly interacting with graphical user interfaces (GUIs) on computers and mobile devices, offering significant potential to enhance human productivity by completing an open-ended space of user queries. However, current agents face significant challenges: imprecise grounding of GUI elements, difficulties with long-horizon task planning, and performance bottlenecks from relying on single generalist models for diverse cognitive tasks. To this end, we introduce Agent S2, a novel compositional framework that delegates cognitive responsibilities across various generalist and specialist models. We propose a novel Mixture-of-Grounding technique to achieve precise GUI localization and introduce Proactive Hierarchical Planning, dynamically refining action plans at multiple temporal scales in response to evolving observations. Evaluations demonstrate that Agent S2 establishes new state-of-the-art (SOTA) performance on three prominent computer use benchmarks. Specifically, Agent S2 achieves 18.9% and 32.7% relative improvements over leading baseline agents such as Claude Computer Use and UI-TARS on the OSWorld 15-step and 50-step evaluation. Moreover, Agent S2 generalizes effectively to other operating systems and applications, surpassing previous best methods by 52.8% on WindowsAgentArena and by 16.52% on AndroidWorld relatively. Code available at https://github.com/simular-ai/Agent-S.