12 papers
RSIGym: A Flexible Environment for Recursive Self-Improvement
Fanqing Meng, Lingxiao Du, Haocheng Lu +6
Recursive self-improvement requires carrying accepted changes into later improvement cycles, while studying agent-proposed changes also requires substantial research infrastructure…
Gödel Forest: Balancing Search Depth and Breadth for Data-Centric Recursive Self-Improvement
Ziqi Zhao, Fanqing Meng, Haocheng Lu +4
Recursive self-improvement (RSI) aims to achieve compounding gains by having models improve themselves. While most existing RSI systems optimize external agent harnesses or prompts…
RSIBench-Data: Benchmarking Data-Centric Research for Recursive Self-Improvement
Fanqing Meng, Lingxiao Du, Qiguang Chen +4
Recursive self-improvement requires turning evidence of model failures into better models. Data-centric post-training research entails diagnosing capability gaps, designing and val…
ClawMark: A Living-World Benchmark for Multi-Turn, Multi-Day, Multimodal Coworker Agents
Fanqing Meng, Lingxiao Du, Zijian Wu +46
Language-model agents are increasingly used as persistent coworkers that assist users across multiple working days. During such workflows, the surrounding environment may change in…
STELLAR-E: a Synthetic, Tailored, End-to-end LLM Application Rigorous Evaluator
Alessio Sordo, Lingxiao Du, Meeka-Hanna Lenisa +2
The increasing reliance on Large Language Models (LLMs) across diverse sectors highlights the need for robust domain-specific and language-specific evaluation datasets; however, th…
Gym-V: A Unified Vision Environment System for Agentic Vision Research
Fanqing Meng, Lingxiao Du, Jiawei Gu +9
As agentic systems increasingly rely on reinforcement learning from verifiable rewards, standardized ``gym'' infrastructure has become essential for rapid iteration, reproducibilit…
MCPMark: A Benchmark for Stress-Testing Realistic and Comprehensive MCP Use
Zijian Wu, Xiangyan Liu, Xinyuan Zhang +12
MCP standardizes how LLMs interact with external systems, forming the foundation for general agents. However, existing MCP benchmarks remain narrow in scope: they focus on read-hea…
SafeWork-R1: Coevolving Safety and Intelligence under the AI-45 Law
Shanghai AI Lab, :, Yicheng Bao +115
We introduce SafeWork-R1, a cutting-edge multimodal reasoning model that demonstrates the coevolution of capabilities and safety. It is developed by our proposed SafeLadder framewo…
OWMM-Agent: Open World Mobile Manipulation With Multi-modal Agentic Data Synthesis
Junting Chen, Haotian Liang, Lingxiao Du +8
The rapid progress of navigation, manipulation, and vision models has made mobile manipulators capable in many specialized tasks. However, the open-world mobile manipulation (OWMM)…
MM-PRM: Enhancing Multimodal Mathematical Reasoning with Scalable Step-Level Supervision
Lingxiao Du, Fanqing Meng, Zongkai Liu +4
While Multimodal Large Language Models (MLLMs) have achieved impressive progress in vision-language understanding, they still struggle with complex multi-step reasoning, often prod…
CPGD: Toward Stable Rule-based Reinforcement Learning for Language Models
Zongkai Liu, Fanqing Meng, Lingxiao Du +4
Recent advances in rule-based reinforcement learning (RL) have significantly improved the reasoning capability of language models (LMs) with rule-based rewards. However, existing R…
MM-Eureka: Exploring the Frontiers of Multimodal Reasoning with Rule-based Reinforcement Learning
Fanqing Meng, Lingxiao Du, Zongkai Liu +12
DeepSeek R1, and o1 have demonstrated powerful reasoning capabilities in the text domain through stable large-scale reinforcement learning. To enable broader applications, some wor…