Metraipolis › archive › dc_71f97f316fcc

Adversarial Robustness of Retrieval-Based Fact-Checking: Detecting and Mitigating Evidence Contamination — arXiv:2611.02139

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.02139
License (as archived)
CC0-1.0
Archived
2026-11-03T11:25:17.288Z
SHA-256 of the archived text
dfec7021945168159babb761112463c31bdddee14ca19f34e591548d0480189b
Stable URL of this page
https://metraipolis.com/documento/dc_71f97f316fcc

How to cite

Adversarial Robustness of Retrieval-Based Fact-Checking: Detecting and Mitigating Evidence Contamination — arXiv:2611.02139 — arXiv (en), https://arxiv.org/abs/2611.02139 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_71f97f316fcc (Metraipolis archive dc_71f97f316fcc, sha256 dfec7021945168159babb761112463c31bdddee14ca19f34e591548d0480189b)
BibTeX
@misc{dc71f97f316fcc,
  title = {Adversarial Robustness of Retrieval-Based Fact-Checking: Detecting and Mitigating Evidence Contamination — arXiv:2611.02139},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.02139},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_71f97f316fcc; sha256 dfec7021945168159babb761112463c31bdddee14ca19f34e591548d0480189b},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.02139. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Retrieval-based fact-checking systems ground verification in large document collections, creating a critical vulnerability: adversarial evidence contamination. We demonstrate that factually inconsistent but contextually coherent evidence generated to satisfy both retrieval-similarity and linguistic-fluency constraints can systematically mislead verification models. Contaminating just 3.1-7.4% of training corpora with five label-specific strategies causes adversarial content to penetrate up to 35.2% of top-k retrievals, degrading verification accuracy by up to 46.9% relative F1. Existing retrieval-stage defenses, validated against gradient-optimized or syntactically anomalous attacks, transfer poorly to this threat model: detection rates collapse and F1 recovery exhibits high variance across seeds. We propose GLAD (Geometric-Lexical Adversarial Defense), which fuses dense embedding peer-coherence with lexical distributional isolation. Across two benchmarks, two embedding models, and five random seeds, GLAD achieves the highest detection rates (70-86%) and the most consistent F1 recovery, reclaiming 73-80% of contamination-induced degradation on LIAR-RAW while approaching clean baselines on RAWFC. Crucially, unlike defenses that trade clean performance for robustness, GLAD imposes negligible cost on uncontaminated corpora.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Information Retrieval (cs.IR) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2611.02139 — 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 2026-11-03T11:02:37.853Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).