13 papers
WorkRover: Scaling Long-Horizon Cowork Agent Training Via Recursive Environment Evolution
Hang He, Linting Wang, Guanming Liu +14
Scaling interactive environments is important for training long-horizon agents through reinforcement learning. These environments often require further testing and refinement after…
CheckerBench: Can Long-Horizon Agents Synthesize Static-Analysis Checkers?
Hang He, Li Wang, Hao Chen +9
Static-analysis checker synthesis requires agents to interpret a defect specification, inspect a repository, implement analyzer-specific logic, and refine the checker through repea…
VistaHop: Benchmarking Long-Horizon Visual DeepSearch
Hang He, Chuhuai Yue, Chengqi Dong +6
Visual DeepSearch tasks require multimodal large language models (MLLMs) to resolve complex visual queries by repeatedly inspecting image regions, grounding reasoning in visual evi…
TAPO: Tool-Aware Policy Optimization via Credit Transfer for Multimodal Search Agents
Chengqi Dong, Chuhuai Yue, Hang He +6
We identify and formally characterize credit misassignment as a systematic failure mode of GRPO in tool-augmented multimodal search agents: its uniform broadcast of trajectory-leve…
VULCA-Bench: A Multicultural Vision-Language Benchmark for Evaluating Cultural Understanding
Haorui Yu, Diji Yang, Hang He +2
We introduce VULCA-Bench, a multicultural art-critique benchmark for evaluating Vision-Language Models' (VLMs) cultural understanding beyond surface-level visual perception. Existi…
A Very Big Video Reasoning Suite
Maijunxian Wang, Ruisi Wang, Juyi Lin +53
Rapid progress in video models has largely focused on visual quality, leaving their reasoning capabilities underexplored. Video reasoning grounds intelligence in spatiotemporally c…
ToolForge: A Data Synthesis Pipeline for Multi-Hop Search without Real-World APIs
Hao Chen, Zhexin Hu, Jiajun Chai +7
Training LLMs to invoke tools and leverage retrieved information necessitates high-quality, diverse data. However, existing pipelines for synthetic data generation often rely on te…
Training Multi-Image Vision Agents via End2End Reinforcement Learning
Chengqi Dong, Chuhuai Yue, Hang He +7
Recent VLM-based agents aim to replicate OpenAI O3's "thinking with images" via tool use, yet most open-source methods restrict inputs to a single image, limiting their applicabili…
LocalSearchBench: Benchmarking Agentic Search in Real-World Local Life Services
Hang He, Chuhuai Yue, Chengqi Dong +12
Recent advances in large reasoning models LRMs have enabled agentic search systems to perform complex multi-step reasoning across multiple sources. However, most studies focus on g…
RLFactory: A Plug-and-Play Reinforcement Learning Post-Training Framework for LLM Multi-Turn Tool-Use
Jiajun Chai, Guojun Yin, Zekun Xu +9
Large language models excel at basic reasoning but struggle with tasks that require interaction with external tools. We present RLFactory, a plug-and-play reinforcement learning po…
Promoting Efficient Reasoning with Verifiable Stepwise Reward
Chuhuai Yue, Chengqi Dong, Yinan Gao +4
Large reasoning models (LRMs) have recently achieved significant progress in complex reasoning tasks, aided by reinforcement learning with verifiable rewards. However, LRMs often s…
Automated detection of atomicity violations in large-scale systems
Hang He, Yixing Luo, Chengcheng Wan +3
Atomicity violations in interrupt-driven programs pose a significant threat to software reliability in safety-critical systems. These violations occur when the execution sequence o…
Probing Perceptual Constancy in Large Vision-Language Models
Haoran Sun, Bingyang Wang, Suyang Yu +14
Perceptual constancy is the ability to maintain stable perceptions of objects despite changes in sensory input, such as variations in distance, angle, or lighting. This ability is…