Online agents can improve future service by constructing reusable tools, guidance, or model states, but this work competes with current requests for the same GPUs. Exploiting idle compute for self-evolution faces a fundamental systems constraint: benefits arrive only after an artifact is published and used, while pausi...
Yang-Bo Wei, Jun-Hong Qian, Zhen Huang et al.· 0 citations
This work forms external-policy evolution under these two budgets and introduces SkillSmith, a conflict-driven learning system that improves over EvoSkill over OfficeQA and SealQA, and sustains improvement beyond SkillClaw on WildClawBench.
Yang-Bo Wei, Zhen Huang, Shaoqiang Lu et al.· arXiv.org· 9 citations
Long-running LLM agents are stateful: beyond the transcript they accrete compressed summaries, plaintext memory, pending tool plans, and, under every serving API, a KV cache. Yet today's"forget"operations delete a plaintext memory record and stop, leaving every artifact derived from the revoked information intact. We f...
Chao Yao, Yang-Bo Wei, Zhen Huang et al.· 2 citations
A reference-free judge is formalized as a latent solver, and a closed-form bound on discriminability (ROC-AUC) in the judge's competence and answer-space size is yields the result that the marginal AUC is confounded by item difficulty while a within-question estimator is not.
Chenleng Chen, Yangbo Wei, Chao Yao et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
