Benchmarks

Papers

• R-Zero: Self-Evolving Reasoning LLM from Zero Data (2025)
  Chengsong Huang et al.
  19.83
• Agentic Reinforced Policy Optimization (2025)
  Guanting Dong et al.
  19.61
• SimpleVLA-RL: Scaling VLA Training via Reinforcement Learning (2025)
  Haozhan Li et al.
  18.40
• SRPO: Enhancing Multimodal LLM Reasoning via Reflection-Aware Reinforcement Learning (2025)
  Zhongwei Wan et al.
  16.17
• QwenLong-L1: Towards Long-Context Large Reasoning Models with Reinforcement Learning (2025)
  Fanqi Wan et al.
  16.01
• EvoArena: Tracking Memory Evolution for Robust LLM Agents in Dynamic Environments (2026)
  Jundong Xu et al.
  16.01
• SpatialClaw: Rethinking Action Interface for Agentic Spatial Reasoning (2026)
  Seokju Cho et al.
  15.70
• MiniMax Sparse Attention (2026)
  Xunhao Lai et al.
  15.51
• MemAgent: Reshaping Long-Context LLM with Multi-Conv RL-based Memory Agent (2025)
  Hongli Yu et al.
  15.38
• Qwen-VLA: Unifying Vision-Language-Action Modeling across Tasks, Environments, and Robot Embodiments (2026)
  Qiuyue Wang et al.
  15.10
• APPO: Agentic Procedural Policy Optimization (2026)
  Xucong Wang et al.
  14.60
• FORT-Searcher: Synthesizing Shortcut-Resistant Search Tasks for Training Deep Search Agents (2026)
  Jia Deng et al.
  14.20
• Kimi K2.5: Visual Agentic Intelligence (2026)
  Kimi Team: Tongtong Bai et al.
  13.84
• Agent Explorative Policy Optimization for Multimodal Agentic Reasoning (2026)
  Minki Kang et al.
  13.75
• Role-Agent: Bootstrapping LLM Agents via Dual-Role Evolution (2026)
  Xucong Wang et al.
  13.67
• LabVLA: Grounding Vision-Language-Action Models in Scientific Laboratories (2026)
  Baochang Ren et al.
  13.59
• Claw-SWE-Bench: A Benchmark for Evaluating OpenClaw-style Agent Harnesses on Coding Tasks (2026)
  Mengyu Zheng et al.
  13.33
• ArcANE: Do Role-Playing Language Agents Stay in Character at the Right Time? (2026)
  Woojung Song et al.
  13.24
• Heterogeneous Agent Collaborative Reinforcement Learning (2026)
  Zhixia Zhang et al.
  13.18
• ACC: Compiling Agent Trajectories for Long-Context Training (2026)
  Qisheng Su et al.
  13.06
• MobileGym: A Verifiable and Highly Parallel Simulation Platform for Mobile GUI Agent Research (2026)
  Dingbang Wu et al.
  12.91
• AdaPlanBench: Evaluating Adaptive Planning in Large Language Model Agents under World and User Constraints (2026)
  Jiayu Liu et al.
  12.83
• AWorld: Orchestrating the Training Recipe for Agentic AI (2025)
  Chengyue Yu et al.
  12.78
• SWE-Explore: Benchmarking How Coding Agents Explore Repositories (2026)
  Shaoqiu Zhang et al.
  12.73
• DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models (2025)
  DeepSeek-AI et al.
  12.70
• K-BrowseComp: A Web Browsing Agent Benchmark Grounded in Korean Contexts (2026)
  Nahyun Lee et al.
  12.62
• SkillOpt: Executive Strategy for Self-Evolving Agent Skills (2026)
  Yifan Yang et al.
  12.58
• LatentOmni: Rethinking Omni-Modal Understanding via Unified Audio-Visual Latent Reasoning (2026)
  Yifan Dai et al.
  12.42
• ResearchMath-14K: Scaling Research-Level Mathematics via Agents (2026)
  Guijin Son et al.
  12.37
• Streaming Communication in Multi-Agent Reasoning (2026)
  Zhen Yang et al.
  12.32
• Planning, Creation, Usage: Benchmarking Llms For Comprehensive Tool Utilization In Real-world Complex Scenarios (2024)
  Shijue Huang, Wanjun Zhong, Jianqiao Lu, et al.
  12.31
• LongTraceRL: Learning Long-Context Reasoning from Search Agent Trajectories with Rubric Rewards (2026)
  Nianyi Lin et al.
  12.21
• Orchestra-o1: Omnimodal Agent Orchestration (2026)
  Fan Zhang et al.
  12.21
• DeepResearchEval: An Automated Framework for Deep Research Task Construction and Agentic Evaluation (2026)
  Yibo Wang et al.
  12.15
• Self-Improving Language Models with Bidirectional Evolutionary Search (2026)
  Guowei Xu et al.
  12.10
• HarnessX: A Composable, Adaptive, and Evolvable Agent Harness Foundry (2026)
  Tingyang Chen et al.
  11.97
• Agent models: Internalizing Chain-of-Action Generation into Reasoning models (2025)
  Yuxiang Zhang et al.
  11.87
• CHI-Bench: Can AI Agents Automate End-to-End, Long-Horizon, Policy-Rich Healthcare Workflows? (2026)
  Haolin Chen et al.
  11.86
• Flash-Searcher: Fast and Effective Web Agents via DAG-Based Parallel Execution (2025)
  Tianrui Qin et al.
  11.85
• MultiAgentBench: Evaluating the Collaboration and Competition of LLM agents (2025)
  Kunlun Zhu et al.
  11.84
• SearchSwarm: Towards Delegation Intelligence in Agentic LLMs for Long-Horizon Deep Research (2026)
  Pu Ning et al.
  11.78
• QUEST: Training Frontier Deep Research Agents with Fully Synthetic Tasks (2026)
  Jian Xie et al.
  11.71
• World Models Meet Language Models: On the Complementarity of Concrete and Abstract Reasoning (2026)
  Yucheng Zhou et al.
  11.70
• Latent Collaboration in Multi-Agent Systems (2025)
  Jiaru Zou et al.
  11.66
• Graph-R1: Towards Agentic GraphRAG Framework via End-to-end Reinforcement Learning (2025)
  Haoran Luo et al.
  11.55
• Rethinking Memory as Continuously Evolving Connectivity (2026)
  Jizhan Fang et al.
  11.49
• Personal AI Agent for Camera Roll VQA (2026)
  Thao Nguyen et al.
  11.45
• Masking Stale Observations Helps Search Agents -- Until It Doesn't: A Regime Map and Its Mechanism (2026)
  Haoxiang Zhang et al.
  11.44
• $\pi$-Bench: Evaluating Proactive Personal Assistant Agents in Long-Horizon Workflows (2026)
  Haoran Zhang et al.
  11.23
• Cosmos 3: Omnimodal World Models for Physical AI (2026)
  NVIDIA et al.
  11.10