迈向大语言模型中可靠的真实性对齐不确定性估计
Towards Reliable Truth-Aligned Uncertainty Estimation in Large Language Models
AI总结：
本文提出Truth AnChoring方法，通过将原始分数映射到事实对齐分数，解决不确定性估计指标在低信息区域的非判别性问题，提升大语言模型的可靠性。
摘要：
不确定性估计（UE）旨在检测大语言模型（LLMs）的幻觉输出以提高其可靠性。然而，UE指标在不同配置下的性能不稳定，显著限制了其应用。本文将这一现象定义为代理失败，因为大多数UE指标源于模型行为，而非显式基于LLM输出的事实正确性。为此，我们提出Truth AnChoring（TAC），一种后处理校准方法，通过将原始分数映射到事实对齐分数来修复UE指标。即使在噪声和少样本监督下，TAC也能支持学习良好校准的不确定性估计，并提供实用的校准协议。我们的发现突显了将启发式UE指标视为事实不确定性直接指示器的局限性，并将TAC定位为迈向更可靠LLMs不确定性估计的必要步骤。代码库可在https://github.com/ponhvoan/TruthAnchor/上找到。
英文摘要：
Uncertainty estimation (UE) aims to detect hallucinated outputs of large language models (LLMs) to improve their reliability. However, UE metrics often exhibit unstable performance across configurations, which significantly limits their applicability. In this work, we formalise this phenomenon as proxy failure, since most UE metrics originate from model behaviour, rather than being explicitly grounded in the factual correctness of LLM outputs. With this, we show that UE metrics become non-discriminative precisely in low-information regimes. To alleviate this, we propose Truth AnChoring (TAC), a post-hoc calibration method to remedy UE metrics, by mapping the raw scores to truth-aligned scores. Even with noisy and few-shot supervision, our TAC can support the learning of well-calibrated uncertainty estimates, and presents a practical calibration protocol. Our findings highlight the limitations of treating heuristic UE metrics as direct indicators of truth uncertainty, and position our TAC as a necessary step toward more reliable uncertainty estimation for LLMs. The code repository is available at https://github.com/ponhvoan/TruthAnchor/.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
