Seeing through the Conflict: Transparent Knowledge Conflict Handling in Retrieval-Augmented Generation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Ye, Hua, Chen, Siyuan, Zhong, Ziqi, Xiao, Canran, Zhang, Haoliang, Wu, Yuhan, Shen, Fei
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918281529524224
author Ye, Hua
Chen, Siyuan
Zhong, Ziqi
Xiao, Canran
Zhang, Haoliang
Wu, Yuhan
Shen, Fei
author_facet Ye, Hua
Chen, Siyuan
Zhong, Ziqi
Xiao, Canran
Zhang, Haoliang
Wu, Yuhan
Shen, Fei
contents Large language models (LLMs) equipped with retrieval--the Retrieval-Augmented Generation (RAG) paradigm--should combine their parametric knowledge with external evidence, yet in practice they often hallucinate, over-trust noisy snippets, or ignore vital context. We introduce TCR (Transparent Conflict Resolution), a plug-and-play framework that makes this decision process observable and controllable. TCR (i) disentangles semantic match and factual consistency via dual contrastive encoders, (ii) estimates self-answerability to gauge confidence in internal memory, and (iii) feeds the three scalar signals to the generator through a lightweight soft-prompt with SNR-based weighting. Across seven benchmarks TCR improves conflict detection (+5-18 F1), raises knowledge-gap recovery by +21.4 pp and cuts misleading-context overrides by -29.3 pp, while adding only 0.3% parameters. The signals align with human judgements and expose temporal decision patterns.
format Preprint
id arxiv_https___arxiv_org_abs_2601_06842
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Seeing through the Conflict: Transparent Knowledge Conflict Handling in Retrieval-Augmented Generation
Ye, Hua
Chen, Siyuan
Zhong, Ziqi
Xiao, Canran
Zhang, Haoliang
Wu, Yuhan
Shen, Fei
Artificial Intelligence
I.2.7; H.3.3
Large language models (LLMs) equipped with retrieval--the Retrieval-Augmented Generation (RAG) paradigm--should combine their parametric knowledge with external evidence, yet in practice they often hallucinate, over-trust noisy snippets, or ignore vital context. We introduce TCR (Transparent Conflict Resolution), a plug-and-play framework that makes this decision process observable and controllable. TCR (i) disentangles semantic match and factual consistency via dual contrastive encoders, (ii) estimates self-answerability to gauge confidence in internal memory, and (iii) feeds the three scalar signals to the generator through a lightweight soft-prompt with SNR-based weighting. Across seven benchmarks TCR improves conflict detection (+5-18 F1), raises knowledge-gap recovery by +21.4 pp and cuts misleading-context overrides by -29.3 pp, while adding only 0.3% parameters. The signals align with human judgements and expose temporal decision patterns.
title Seeing through the Conflict: Transparent Knowledge Conflict Handling in Retrieval-Augmented Generation
topic Artificial Intelligence
I.2.7; H.3.3
url https://arxiv.org/abs/2601.06842