移动代理评估中LLM裁判靠谱吗？全新基准MobileJudgeBench首次系统检验其可靠性。
arXiv:2608.11434v1 Announce Type: new Abstract: Mobile agent benchmarks increasingly rely on LLM-based judges to evaluate task completion, yet the rel…
用多模型交叉审计替代提示工程去偏，为LLM裁判公正性提供新思路。
arXiv:2607.28636v1 Announce Type: new Abstract: LLMs increasingly serve as automated judges, but their judgments remain vulnerable to cognitive biases…
多角色评分标准生成，让LLM裁判与奖励模型更稳健、更对齐人类偏好。
arXiv:2607.01830v1 Announce Type: new Abstract: Reliable reward and preference signals are critical for evaluating and optimizing large language model…
从零搭建LLM裁判系统，与人类评分对比，揭露看似好用实则暗藏陷阱的评估局限。
In Part 1 the model's job was to pick one of 77 labels, so I could check it with == . But most real LLM output isn't like that — it's a paragraph, a s…
LLM裁判小组校准：有限预算下低维叠加器vs联合表，谁更优？
arXiv:2606.01034v1 Announce Type: new Abstract: We study when LLM judge panels should be calibrated with low-dimensional stackers versus joint output …
Frost Training利用嵌入空间奖励梯度优化LLM裁判策略，首次将GCG越狱信号用于提升蒙特卡洛训练效率。
arXiv:2605.27701v1 Announce Type: new Abstract: We present Frost Training, a method for improving Monte Carlo-based policy optimization for a large fa…
用Wald的序贯概率比检验动态调控LLM辩论轮数，显著提升推理效率与资源利用率
arXiv:2605.19193v1 Announce Type: new Abstract: Multi-agent LLM debate improves factuality and reasoning, but most recipes pick a fixed round count, o…