19 papers
XRepoSkill: Learning Transferable Skills for Software Engineering Agents
Yaoqi Guo, Haoyang Zhou, Jiayi Zhang +7
Software engineering agents increasingly use reusable skills distilled from prior experience to resolve repository-level issues, yet such skills often fail to transfer across repos…
Matching Supervision to the Student's Learning Capacity: A Unified Framework for On-Policy Self-Distillation
Yongkang Yang, Zhezheng Hao, Hong Zhang +8
On-policy self-distillation (OPSD) improves the reasoning abilities of LLMs by internalizing privileged context into model parameters through self-distillation. Two recent research…
Benchmarking the Residual: What Long-Horizon Evaluations Add Beyond Matched Short-Task Performance
Chao Peng, Zhiheng Lyu, Peijie Dong +2
Long-horizon benchmarks often show that agents fail more as tasks become longer. This observation is useful for deployment, but it does not by itself explain why failure occurs. Mo…
Tencent WorkBuddy Bench: A Multi-Domain Coding-Agent Benchmark with Contamination-Resistant Task Construction
Tencent WorkBuddy Bench Team, Siqi Cai, Shaopeng Chen +35
We introduce Tencent WorkBuddy Bench, a multi-domain evaluation suite for coding agents; this report documents its construction methodology, scoring protocol, and a cross-model lea…
Evolve as a Team: Collaborative Self-Evolution for LLM-based Multi-Agent Systems
Zhezheng Hao, Tianfu Wang, Huanshuo Dong +7
LLM-based multi-agent systems (MAS) have emerged as an effective paradigm for complex and long-horizon tasks. However, in real-world tasks, MAS often exhibit various failures durin…
Echo: Learning from Experience Data via User-Driven Refinement
Hande Dong, Xiaoyun Liang, Jiarui Yu +15
Static "human data" faces inherent limitations: it is expensive to scale and bounded by the knowledge of its creators. Continuous learning from "experience data" - interactions bet…
EPIC-Bench: A Perception-Centric Benchmark for Fine-Grained Embodied Visual Grounding in Vision-Language Models
Haozhe Shan, Xiancong Ren, Han Dong +10
While large vision-language models (VLMs) are increasingly adopted as the perceptual backbone for embodied agents, their fundamental vision-language grounding (VLG) capability in e…
PDEAgent-Bench: A Multi-Metric, Multi-Library Benchmark for PDE Solver Generation
Zhen Hang, Yushan Yashengjiang, Junhui Li +21
PDE-to-solver code generation aims to automatically synthesize executable numerical solvers from partial differential equation (PDE) specifications. This task requires not only und…
Scaling Human-AI Coding Collaboration Requires a Governable Consensus Layer
Tianfu Wang, Zhezheng Hao, Yin Wu +7
Vibe coding turns natural-language intent into working software, but code generation alone does not give humans and agents a shared account of what they are building. As collaborat…
LEPO: Latent Reasoning Policy Optimization for Large Language Models
Yuyan Zhou, Jiarui Yu, Hande Dong +4
Recently, latent reasoning has been introduced into large language models (LLMs) to leverage rich information within a continuous space. However, without stochastic sampling, these…
ReCreate: Reasoning and Creating Domain Agents Driven by Experience
Zhezheng Hao, Hong Wang, Jian Luo +6
Large Language Model (LLM) agents are reshaping the industrial landscape. However, most practical agents still rely on human-designed scaffolds, making their creation and adaptatio…
Accelerating IC Thermal Simulation Data Generation via Block Krylov and Operator Action
Hong Wang, Wenkai Yang, Jie Wang +6
Recent advances in data-driven approaches, such as neural operators (NOs), have shown substantial efficacy in reducing the solution time for integrated circuit (IC) thermal simulat…
Scheduling Your LLM Reinforcement Learning with Reasoning Trees
Hong Wang, Zhezheng Hao, Jian Luo +6
Using Reinforcement Learning with Verifiable Rewards (RLVR) to optimize Large Language Models (LLMs) can be conceptualized as progressively editing a query's `Reasoning Tree'. This…
GAPO: Robust Advantage Estimation for Real-World Code LLMs
Jianqing Zhang, Zhezheng Hao, Wei Xia +7
Reinforcement learning (RL) is widely used for post-training large language models (LLMs) in code editing, where group-relative methods, such as GRPO, are popular due to their crit…
Rethinking Entropy Interventions in RLVR: An Entropy Change Perspective
Zhezheng Hao, Hong Wang, Haoyang Liu +6
Reinforcement Learning with Verifiable Rewards (RLVR) serves as a cornerstone technique for enhancing the reasoning capabilities of Large Language Models (LLMs). However, its train…
AP2O-Coder: Adaptively Progressive Preference Optimization for Reducing Compilation and Runtime Errors in LLM-Generated Code
Jianqing Zhang, Wei Xia, Hande Dong +2
LLMs' code generation capabilities have yielded substantial improvements in the effectiveness of programming tasks. However, LLM-generated code still suffers from compilation and r…
Unleashing the True Potential of LLMs: A Feedback-Triggered Self-Correction with Long-Term Multipath Decoding
Jipeng Li, Zeyu Gao, Yubin Qi +3
Large Language Models (LLMs) have achieved remarkable performance across diverse tasks, yet their susceptibility to generating incorrect content during inference remains a critical…
UniSVG: A Unified Dataset for Vector Graphic Understanding and Generation with Multimodal Large Language Models
Jinke Li, Jiarui Yu, Chenxing Wei +5
Unlike bitmap images, scalable vector graphics (SVG) maintain quality when scaled, frequently employed in computer vision and artistic design in the representation of SVG code. In…
ReDit: Reward Dithering for Improved LLM Policy Optimization
Chenxing Wei, Jiarui Yu, Ying Tiffany He +3
DeepSeek-R1 has successfully enhanced Large Language Model (LLM) reasoning capabilities through its rule-based reward system. While it's a ''perfect'' reward system that effectivel…