English(EN) Test-Time Confidence Adaptation under RL-Induced Trajectory Shift

新框架StaR-TTA提升了RL训练后LLM的置信度

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-27 04:00

研究人员开发了StaR-TTA，一个旨在提升大型语言模型在经过代理强化学习（RL）训练后的置信度估计的新框架。该方法解决了RL训练后推理轨迹偏移导致的置信度分数下降问题。StaR-TTA利用了这样一个观察结果：即使在RL训练后，更稳定的轨迹也倾向于产生更正确的答案。通过从标记的基础轨迹的正确性和稳定性信号中学习，StaR-TTA将置信度分数自适应到未标记的RL轨迹上，在四个问答基准测试中实现了0.721的平均AUROC。 AI

影响通过改进强化学习后LLM输出的置信度估计，增强了其可靠性。

排序理由该集群包含一篇详细介绍LLM置信度估计新方法的论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.AI 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 9 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群包含一篇详细介绍LLM置信度估计新方法的论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, model release; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Same-day; Cluster formed today. Ranking reflects the current source set at time of score.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.AI TIER_1 English(EN) · Zhenzhen Ren, Xiaoqing Zhang, GuoBiao Li, Bowen Fang, Yanxi Wang, Xinpeng Zhang, Ke Han · 2026-10-27 04:00

测试时基于RL诱导轨迹偏移的置信度自适应

arXiv:2610.35500v1 Announce Type: new Abstract: Trajectory-based confidence estimation has shown promising performance for large language models (LLMs) by leveraging rich signals from their reasoning processes. However, we find that such confidence estimators can degrade substant…

本文如何被排名

报道来源 [1]

测试时基于RL诱导轨迹偏移的置信度自适应

相关实体

相关话题