1
A Theory of Post-hoc Debate Judgement
从理论层面拆解事后辩论评判机制，给AI对齐与评估研究提供全新视角。
arXiv:2608.19002v1 Announce Type: new Abstract: Debates have recently emerged as a useful methodology for agentic AI to improve performance as well as…
2
ParEvalLayer: When Partial LLM-Agent Evaluations Support a Decision
LLM智能体评估不必等全部结果，ParEvalLayer教你用部分评估支撑决策，大幅提升效率。
arXiv:2608.02444v1 Announce Type: new Abstract: LLM-agent evaluations often produce task outcomes long before the full benchmark run is complete. A pa…