---
id: 20260611-T0-11
title: "新方法让LLM识别知识冲突，提升推理可靠性"
title_en: "Conflict-Aware Decoding Improves LLM Reliability"
url: https://ai.daily.yangsir.net/daily/20260611-T0-11
issue_date: 2026-06-11
publish_date: 2026-06-10T04:00:00.000Z
category: research
source_name: "arXiv cs.AI"
source_url: https://arxiv.org/abs/2606.10298
---

# 新方法让LLM识别知识冲突，提升推理可靠性

论文提出从上下文感知到冲突感知的新解码方法。当LLM使用检索或增强上下文时，外部信息与模型内部知识常冲突，影响可靠性。新方法通过对比解码识别冲突，比传统方法更准确。实验显示在知识密集型任务中错误率降低15%，代码已开源。

## English Version

**Conflict-Aware Decoding Improves LLM Reliability**

New research generalizes contrastive decoding from context-aware to conflict-aware for LLMs. Existing methods struggle with conflicts between external context and internal knowledge. The new approach identifies these conflicts, showing 15% lower error rates in knowledge-intensive tasks. Code is available on arXiv.

---

**来源**：[arXiv cs.AI](https://arxiv.org/abs/2606.10298)

**详情页**：https://ai.daily.yangsir.net/daily/20260611-T0-11

---

*智语观潮 · Daily — https://ai.daily.yangsir.net/llms.txt*