12 papers
ASI-Bench: At the Dawn of Artificial Superintelligence
Junwei Zhou, Zhen Sun, Binyu Li +39
Artificial superintelligence (ASI) requires AI to move beyond mastering existing knowledge toward exploring the unknown, creating new knowledge, and turning new ideas into verifiab…
EnvACE: Internalizing Environment Dynamics via World Rehearsal for Agentic Reinforcement Learning
Zishan Xu, Zhiyuan Yao, Yuxin Chen +9
Training large language model agents for long-horizon tool use typically relies on interactions with real or synthesized executable environments, whose construction and verificatio…
HarnessCompass: Guiding Automatic Harness Evolution toward Generalizable and Effective Agent Harnesses
Luan Zhang, Ruochen Zhou, Dandan Song +9
Harness design plays a critical role in agent performance by shaping how large language models (LLMs) perceive, reason over, and act within executable environments. Recent work has…
BALTO: Balanced Token-Level Policy Optimization for Hallucination Mitigation
Ning Li, Zixuan Guo, Yan Xu +7
Hallucinations remain a major obstacle to deploying large language models (LLMs) in knowledge-intensive settings, where generated responses must be faithfully grounded in provided…
Anticipate and Learn: Unleashing Idle-Time Compute in Proactive Agents
Haoyi Hu, Qirong Lyu, Xianghan Kong +7
While AI agents demonstrate remarkable capabilities in reasoning and tool use, they remain fundamentally reactive: they compute responses only after explicit user prompts. This par…
SkillEvolBench: Benchmarking the Evolution from Episodic Experience to Procedural Skills
Yingtie Lei, Zhongwei Wan, Jiankun Zhang +13
Large language model (LLM) agents accumulate rich episodic trajectories while solving real-world tasks, but it remains unclear whether such experience can be distilled into reusabl…
PIVOT: Bridging Planning and Execution in LLM Agents via Trajectory Refinement
Tuo Zhang, Alin-Ionut Popa, Yan Xu +2
Large language model (LLM)-based agents frequently generate seemingly coherent plans that fail upon execution due to infeasible actions, constraint violations, and compounding erro…
Discover and Prove: An Open-source Agentic Framework for Hard Mode Automated Theorem Proving in Lean 4
Chengwu Liu, Yichun Yin, Ye Yuan +7
Most ATP benchmarks embed the final answer within the formal statement -- a convention we call "Easy Mode" -- a design that simplifies the task relative to what human competitors f…
QFFT, Question-Free Fine-Tuning for Adaptive Reasoning
Wanlong Liu, Junxiao Xu, Fei Yu +7
Recent advancements in Long Chain-of-Thought (CoT) reasoning models have improved performance on complex tasks, but they suffer from overthinking, which generates redundant reasoni…
Safe: Enhancing Mathematical Reasoning in Large Language Models via Retrospective Step-aware Formal Verification
Chengwu Liu, Ye Yuan, Yichun Yin +7
Chain-of-Thought (CoT) prompting has become the de facto method to elicit reasoning capabilities from large language models (LLMs). However, to mitigate hallucinations in CoT that…
Instruction-Tuning Data Synthesis from Scratch via Web Reconstruction
Yuxin Jiang, Yufei Wang, Chuhan Wu +8
The improvement of LLMs' instruction-following capabilities depends critically on the availability of high-quality instruction-response pairs. While existing automatic data synthet…
Teaching LLMs According to Their Aptitude: Adaptive Reasoning for Mathematical Problem Solving
Xin Xu, Yan Xu, Tianhao Chen +9
Existing approaches to mathematical reasoning with large language models (LLMs) rely on Chain-of-Thought (CoT) for generalizability or Tool-Integrated Reasoning (TIR) for precise c…