paper-with-me

Papers

Tug-of-War Between Knowledge: Exploring and Resolving Knowledge Conflicts in Retrieval-Augmented Language Models

2024-02-22 · Zhuoran Jin, Pengfei Cao, Yubo Chen, Kang Liu, XiaoJian Jiang, Jiexin Xu, Qiuxia Li, Jun Zhao

Retrieval-augmented language models (RALMs) have demonstrated significant potential in refining and expanding their internal memory by retrieving evidence from external sources. However, RALMs will inevitably encounter knowledge conflicts when integrating their internal memory with external sources. Knowledge conflicts can ensnare RALMs in a tug-of-war between knowledge, limiting their practical applicability. In this paper, we focus on exploring and resolving knowledge conflicts in RALMs. First, we present an evaluation framework for assessing knowledge conflicts across various dimensions. Then, we investigate the behavior and preference of RALMs from the following two perspectives: (1) Conflicts between internal memory and external sources: We find that stronger RALMs emerge with the Dunning-Kruger effect, persistently favoring their faulty internal memory even when correct evidence is provided. Besides, RALMs exhibit an availability bias towards common knowledge; (2) Conflicts between truthful, irrelevant and misleading evidence: We reveal that RALMs follow the principle of majority rule, leaning towards placing trust in evidence that appears more frequently. Moreover, we find that RALMs exhibit confirmation bias, and are more willing to choose evidence that is consistent with their internal memory. To solve the challenge of knowledge conflicts, we propose a method called Conflict-Disentangle Contrastive Decoding (CD2) to better calibrate the model's confidence. Experimental results demonstrate that our CD2 can effectively resolve knowledge conflicts in RALMs.

📄 PDF Abstract BibTeX arXiv:2402.14409

Code (0)

등록된 구현이 없습니다.

Tasks

Retrieval

Methods 이 논문이 사용한 방법론

Focus 설명 없음

Similar Papers 제목 키워드 기반

Discerning and Resolving Knowledge Conflicts through Adaptive Decoding with Contextual Information-Entropy Constraint

2024-02-19 · Xiaowei Yuan, Zhao Yang, Yequan Wang, Shengping Liu 외

Large language models internalize enormous parametric knowledge during pre-training. Concurrently, realistic applications necessitate external contextual knowledge to aid models on the underlying tasks. This raises a cru…

TruthfulRAG: Resolving Factual-level Conflicts in Retrieval-Augmented Generation with Knowledge Graphs

2025-11-13 · Shuyi Liu, Yuming Shang, Xi Zhang arxiv

Retrieval-Augmented Generation (RAG) has emerged as a powerful framework for enhancing the capabilities of Large Language Models (LLMs) by integrating retrieval-based methods with generative models. As external knowledge…

Knowledge Graphs

Where Knowledge Collides: A Mechanistic Study of Intra-Memory Knowledge Conflict in Language Models

2026-01-14 · Minh Vu Pham, Hsuvas Borkakoty, Yufang Hou arxiv

In language models (LMs), intra-memory knowledge conflict arises when inconsistent information about the same subject is encoded within the model's parametric knowledge. Prior work has primarily focused on resolving conf…

knowledge editing

Resolving Knowledge Conflicts in Large Language Models

2023-10-02 · Yike Wang, Shangbin Feng, Heng Wang, Weijia Shi 외

Large language models (LLMs) often encounter knowledge conflicts, scenarios where discrepancy arises between the internal parametric knowledge of LLMs and non-parametric information provided in the prompt context. In thi…

SegSub: Evaluating Robustness to Knowledge Conflicts and Hallucinations in Vision-Language Models

2025-02-19 · Peter Carragher, Nikitha Rao, Abhinand Jha, R Raghav 외

Vision language models (VLM) demonstrate sophisticated multimodal reasoning yet are prone to hallucination when confronted with knowledge conflicts, impeding their deployment in information-sensitive contexts. While exis…

counterfactualHallucinationMultimodal Reasoning