Trace2Skill: Distill Trajectory-Local Lessons into Transferable Agent Skills
Paper • 2603.25158 • Published • 51
Heming Xia
hemingkx
AI & ML interests
Efficient and Effective NLP, Tool Learning, and Vision-Language Understanding.
Recent Activity
upvoted a paper 3 days ago
Trace2Skill: Distill Trajectory-Local Lessons into Transferable Agent Skills upvoted a paper 3 days ago
Memento-Skills: Let Agents Design Agents upvoted a paper 3 days ago
How Far Can Unsupervised RLVR Scale LLM Training?Organizations
None yet
upvoted 20
papers 3 days ago
Memento-Skills: Let Agents Design Agents
Paper • 2603.18743 • Published • 58
How Far Can Unsupervised RLVR Scale LLM Training?
Paper • 2603.08660 • Published • 59
OpenResearcher: A Fully Open Pipeline for Long-Horizon Deep Research Trajectory Synthesis
Paper • 2603.20278 • Published • 95
MetaClaw: Just Talk -- An Agent That Meta-Learns and Evolves in the Wild
Paper • 2603.17187 • Published • 139
Efficient Reasoning with Balanced Thinking
Paper • 2603.12372 • Published • 148
OpenClaw-RL: Train Any Agent Simply by Talking
Paper • 2603.10165 • Published • 152
Attention Residuals
Paper • 2603.15031 • Published • 183
DFlash: Block Diffusion for Flash Speculative Decoding
Paper • 2602.06036 • Published • 58
SkillOrchestra: Learning to Route Agents via Skill Transfer
Paper • 2602.19672 • Published • 58
MemSkill: Learning and Evolving Memory Skills for Self-Evolving Agents
Paper • 2602.02474 • Published • 63
Learning beyond Teacher: Generalized On-Policy Distillation with Reward Extrapolation
Paper • 2602.12125 • Published • 63
Experiential Reinforcement Learning
Paper • 2602.13949 • Published • 74
SkillRL: Evolving Agents via Recursive Skill-Augmented Reinforcement Learning
Paper • 2602.08234 • Published • 75
Golden Goose: A Simple Trick to Synthesize Unlimited RLVR Tasks from Unverifiable Internet Text
Paper • 2601.22975 • Published • 111
PaperBanana: Automating Academic Illustration for AI Scientists
Paper • 2601.23265 • Published • 225
Does Your Reasoning Model Implicitly Know When to Stop Thinking?
Paper • 2602.08354 • Published • 264
LightThinker++: From Reasoning Compression to Memory Management
Paper • 2604.03679 • Published • 37
Beyond Accuracy: Unveiling Inefficiency Patterns in Tool-Integrated Reasoning
Paper • 2604.05404 • Published • 42
Rethinking On-Policy Distillation of Large Language Models: Phenomenology, Mechanism, and Recipe
Paper • 2604.13016 • Published • 85