SkillAdaptor: How LLM Agents Rewrite Their Own Skills
SkillAdaptor edits an agent's skill library from failed trajectories without touching model weights, lifting WebShop score +2.3 and PinchBench +1.5 over the frozen backbone.
Institution
Ant Group
Chinese fintech company behind Alipay; its research teams work on large-scale machine learning, trustworthy AI, and foundation models, and co-authored the LLaDA diffusion language model.
Diffusion Language Models · Renmin University of China
LLaDA: An 8B Diffusion Language Model That Rivals LLaMA3
LLaDA trains an 8B language model by masked diffusion instead of next-token prediction, matches LLaMA3 8B in in-context learning, hits 70.7 on GSM8K, and beats GPT-4o on the reversal-curse poem task.