Metraipolis › archive › dc_e143d211f0ab

Parameters vs. Context: TRACE Fine-Tuning for Robust Retrieval-Augmented Generation — arXiv:2609.30337

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.30337
License (as archived)
CC0-1.0
Archived
2026-10-08T02:10:18.363Z
SHA-256 of the archived text
904bcbc8d8afb970bdcb6ba83e6885a2f5967d448496592dec8ece50b7cd3270
Stable URL of this page
https://metraipolis.com/documento/dc_e143d211f0ab

How to cite

Parameters vs. Context: TRACE Fine-Tuning for Robust Retrieval-Augmented Generation — arXiv:2609.30337 — arXiv (en), https://arxiv.org/abs/2609.30337 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e143d211f0ab (Metraipolis archive dc_e143d211f0ab, sha256 904bcbc8d8afb970bdcb6ba83e6885a2f5967d448496592dec8ece50b7cd3270)
BibTeX
@misc{dce143d211f0ab,
  title = {Parameters vs. Context: TRACE Fine-Tuning for Robust Retrieval-Augmented Generation — arXiv:2609.30337},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.30337},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e143d211f0ab; sha256 904bcbc8d8afb970bdcb6ba83e6885a2f5967d448496592dec8ece50b7cd3270},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.30337. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Retrieval-Augmented Generation (RAG) mitigates knowledge obsolescence and factual hallucination in large language models by introducing external context. However, when retrieved knowledge conflicts with the model's internal parametric knowledge, the model may either blindly follow misleading context or incorrectly rely on parametric knowledge, leading to unreliable responses. To address this issue, this paper proposes TRACE (Debate-TRace and Answer-Completeness rEgularized fine-tuning), a robust fine-tuning framework for RAG under knowledge conflicts. First, we propose a fine-tuning method that leverages multi-agent debate traces to extract correct candidates, incorrect candidates, and answer-shift patterns, providing fine-grained supervision for reliable knowledge-source selection. In addition, we design an answer completeness regularization mechanism to alleviate empty, overly short, and prematurely terminated responses via answer-tail token reinforcement and premature termination suppression. The fine-tuning objective combines correct-answer supervision, incorrect-candidate suppression, answer-tail token reinforcement, and premature termination suppression, enabling the model to use reliable external context, resist misleading or irrelevant retrieved content, and fall back to parametric knowledge when retrieved evidence is unreliable. Experiments across multiple knowledge-conflict scenarios and datasets show that TRACE improves robustness against misleading retrieved knowledge and reduces incomplete answers. These results demonstrate that multi-agent debate traces and answer completeness regularization jointly enhance knowledge-source selection, conflict robustness, and answer quality in RAG models. Our code is available at this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-28 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.30337 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-08T01:43:06.652Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-28?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).