EUVIMEDEUVIMED entdecken
Crossref

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

Hua Ye, Siyuan Chen, Ziqi Zhong, Canran Xiao, Haoliang Zhang, Yuhan Wu, Fei Shen

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (40) · 34423-34431

DOI: 10.1609/aaai.v40i40.40740 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Hua Ye und 6 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

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 percentage points and cuts misleading-context overrides by –29.3 percentage points, while adding only 0.3% parameters. The signals align with human judgements and expose temporal decision patterns.

Zitieren

APA

Ye, H., Chen, S., Zhong, Z., Xiao, C., Zhang, H., Wu, Y., & Shen, F. (2026). Seeing through the Conflict: Transparent Knowledge Conflict Handling in Retrieval-Augmented Generation. Proceedings of the AAAI Conference on Artificial Intelligence, 40(40), 34423-34431. https://doi.org/10.1609/aaai.v40i40.40740

Vancouver

Ye H, Chen S, Zhong Z, Xiao C, Zhang H, Wu Y, et al. Seeing through the Conflict: Transparent Knowledge Conflict Handling in Retrieval-Augmented Generation. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(40):34423-34431. doi:10.1609/aaai.v40i40.40740

BibTeX
@article{Ye2026,
  author = {Ye, Hua and Chen, Siyuan and Zhong, Ziqi and Xiao, Canran and Zhang, Haoliang and Wu, Yuhan and Shen, Fei},
  title = {Seeing through the Conflict: Transparent Knowledge Conflict Handling in Retrieval-Augmented Generation},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {40},
  pages = {34423--34431},
  doi = {10.1609/aaai.v40i40.40740}
}
Kostenlos starten →