PulseAugur
实时 06:30:02
English(EN) You Shouldn't Have Asked: A Pragmatics-Inspired Taxonomy for Evaluating LLM Refusals

新的分类法基于语用学理论评估大型语言模型拒绝

提出了一种新的评估大型语言模型(LLM)拒绝的分类法,该分类法从语用学理论中汲取灵感。该分类法旨在评估 LLM 拒绝有害或不当请求的恰当性,超越了纯粹的安全对齐视角。将此分类法应用于 16 个 LLM 在 14 个危害类别上的研究表明,虽然模型通常会明确拒绝并带有道德判断,但它们常常缺乏细致的互动修复,尤其是在敏感环境中,可能会让用户感到羞辱或被激怒。该研究主张进行对齐评估,应考虑 LLM 拒绝的上下文适应性和社交责任感。 AI

影响 这项研究可能导致更细致的 LLM 对齐评估,通过使拒绝更具上下文适应性来改善用户体验。

排序理由 该条目是一篇学术论文,提出了一种评估 LLM 拒绝的新分类法。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.CL 阅读 →

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

新的分类法基于语用学理论评估大型语言模型拒绝

本文如何被排名

Signal score
30 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
该条目是一篇学术论文,提出了一种评估 LLM 拒绝的新分类法。[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, safety
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
Breaking (< 6h)
Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准

报道来源 [1]

  1. arXiv cs.CL TIER_1 English(EN) · Ruoxuan Li, Pinqiao Wang, Sheng Li, Cameron Robert Jones ·

    你不该问:一种受语用学启发的LLM拒绝评估分类法

    arXiv:2608.30856v1 Announce Type: new Abstract: Refusals are often treated as face-threatening acts in pragmatics because they can challenge the requester's socially claimed self-image. Large language models (LLMs) are increasingly trained to refuse unsafe and inappropriate reque…