27 papers
VIEWCRAFT: Evolving What Multimodal Agents See and How They Read It
Binxiao Xu, Ruichuan An, Bocheng Zou +7
Reusable skills improve multimodal agents without updating model weights, but their effectiveness depends on both the evidence presented and the procedure used to read it. We intro…
EpiCon: Collective Agent Learning through Co-Evolving Multimodal Memory
Ziyun Zeng, Hang Hua, Shaden Alshammari +3
Agents can learn from past executions, but enabling different agents to reuse and build on one another's experience remains challenging. We introduce EpiCon, a shared multimodal me…
Honeycomb: Constant-Size Scene Memory Representation for Video World Models
Jack Wei Lun Shi, Kaichen Zhou, Haoyu Chen +6
Video world models require persistent scene memory to maintain consistency during long-horizon video generation. Existing spatial memories accumulate RGB observations or latent fea…
On the Token Value Inequality in Efficient Reasoning
Runjia Zeng, Hang Hua, Yiyang Liu +5
Chain-of-Thought reasoning has enabled large language models to achieve substantial performance gains on complex tasks. However, these gains come at the cost of dramatically increa…
VideoArgus: Agentic Rubric-Grounded Unified Evaluation for Video Generation and Editing
Ziyun Zeng, Zixuan Wang, Yongsheng Yu +2
Evaluating generated videos remains challenging because existing benchmarks rely on fixed evaluation content, cover only a subset of generation and editing settings, and provide li…
Aligning Quantum Operators with Large Language Models
Rogerio Feris, Yunchao Liu, Pengyuan Li +2
Can Large Language Models (LLMs) understand and reason about quantum operators? Despite their remarkable capabilities in mathematics and symbolic reasoning, LLMs remain inherently…
MemoBench: Benchmarking World Modeling in Dynamically Changing Environments
Haoyu Chen, Kaichen Zhou, Hang Hua +11
Video generation models aspire to simulate dynamic environments, and several benchmarks now evaluate memory consistency across frames. However, most assess consistency only while t…
UXBench: Measuring the Actionability of LLM-Generated UX Critiques
Wenjie Wang, Yue Huang, Zipeng Ling +11
Large language models (LLMs) are increasingly deployed as UX judges that inspect interfaces, diagnose usability problems, and propose repairs. Yet no controlled benchmark measures…
AutoLab: Can Frontier Models Solve Long-Horizon Auto Research and Engineering Tasks?
Zhangchen Xu, Junda Chen, Yue Huang +16
Scientific and engineering progress is fundamentally a long-horizon iterative process: proposing changes, running experiments, measuring outcomes, and continuously refining artifac…
Agent Skills Should Go Beyond Text: The Case for Visual Skills
Binxiao Xu, Ruichuan An, Bocheng Zou +1
Reusable skills are a key mechanism for extending agent capabilities, allowing agents to accumulate experience and solve increasingly complex tasks. Yet most existing skill-learnin…
JobBench: Aligning Agent Work With Human Will
Yuetai Li, Yichen Feng, Zhangchen Xu +21
Current benchmarks for occupational AI agents are scoped primarily by economic values, telling a replacement story. We introduce JobBench, which evaluates AI agents on the workflow…
Aurora: Unified Video Editing with a Tool-Using Agent
Yongsheng Yu, Ziyun Zeng, Zhiyuan Xiao +4
Recent video editing models have converged on a unified conditioning design: a single diffusion transformer jointly consumes text, source video, and reference images, and one set o…
MementoGUI: Learning Agentic Multimodal Memory Control for Long-Horizon GUI Agents
Ziyun Zeng, Hang Hua, Bocheng Zou +3
Recent GUI agents have made substantial progress in visual grounding and action prediction, yet they remain brittle in long-horizon tasks that require maintaining task state across…
Visual Aesthetic Benchmark: Can Frontier Models Judge Beauty?
Yichen Feng, Yuetai Li, Chunjiang Liu +14
Multimodal large language models (MLLMs) are now routinely deployed for visual understanding, generation, and curation. A substantial fraction of these applications require an expl…
GEM-4D: Geometry-Enhanced Video World Models for Robot Manipulation
Kaichen Zhou, Yuzhen Chen, Fangneng Zhan +8
Video world models can generate realistic futures from a single instruction, but they often fail to track the same physical points consistently across time. As a result, the genera…
ChartNet: A Million-Scale, High-Quality Multimodal Dataset for Robust Chart Understanding
Jovana Kondic, Pengyuan Li, Dhiraj Joshi +24
Understanding charts requires models to jointly reason over geometric visual patterns, structured numerical data, and natural language -- a capability where current vision-language…
Guardian-as-an-Advisor: Advancing Next-Generation Guardian Models for Trustworthy LLMs
Yue Huang, Haomin Zhuang, Jiayi Ye +6
Hard-gated safety checkers often over-refuse and misalign with a vendor's model spec; prevailing taxonomies also neglect robustness and honesty, yielding safer-on-paper yet less us…
SPARC: Separating Perception And Reasoning Circuits for Test-time Scaling of VLMs
Niccolo Avogaro, Nayanika Debnath, Li Mi +6
Despite recent successes, test-time scaling -- i.e., dynamically expanding the token budget during inference as needed -- remains brittle for vision-language models (VLMs). Unstruc…
DAVE: A VLM Vision Encoder for Document Understanding and Web Agents
Brandon Huang, Hang Hua, Zhuoran Yu +3
While Vision-language models (VLMs) have demonstrated remarkable performance across multi-modal tasks, their choice of vision encoders presents a fundamental weakness: their low-le…
Video-R4: Reinforcing Text-Rich Video Reasoning with Visual Rumination
Yolo Y. Tang, Daiki Shimada, Hang Hua +4
Understanding text-rich videos requires reading small, transient textual cues that often demand repeated inspection. Yet most video QA models rely on single-pass perception over fi…
Video-LMM Post-Training: A Deep Dive into Video Reasoning with Large Multimodal Models
Yolo Y. Tang, Jing Bi, Pinxin Liu +24
Video understanding represents the most challenging frontier in computer vision, requiring models to reason about complex spatiotemporal relationships, long-term dependencies, and…
Building a Foundational Guardrail for General Agentic Systems via Synthetic Data
Yue Huang, Hang Hua, Yujun Zhou +11
While LLM agents can plan multi-step tasks, intervening at the planning stage-before any action is executed-is often the safest way to prevent harm, since certain risks can lead to…
VQualA 2025 Challenge on Engagement Prediction for Short Videos: Methods and Results
Dasong Li, Sizhuo Ma, Hang Hua +40
This paper presents an overview of the VQualA 2025 Challenge on Engagement Prediction for Short Videos, held in conjunction with ICCV 2025. The challenge focuses on understanding a…
MMPerspective: Do MLLMs Understand Perspective? A Comprehensive Benchmark for Perspective Perception, Reasoning, and Robustness
Yolo Y. Tang, Pinxin Liu, Zhangyun Tan +11
Understanding perspective is fundamental to human visual perception, yet the extent to which multimodal large language models (MLLMs) internalize perspective geometry remains uncle…