🛰️ Daily AI Frontier
‹ back to 2026-08-16

RAGSieve: Self-Referenced Local Contrast for Knowledge-Poison Detection in Retrieval-Augmented Generation

Research LLM Agents

Ranking

Overall 77
Content 95
Popularity 35

Observed public metrics from 1 member.

Merged summary

TL;DR - RAGSieve detects poisoned knowledge in retrieval-augmented generation using local comparisons within the inspected system, requiring neither trusted corpora nor poison labels. Its query-time and corpus-ingestion defenses jointly cut attack success from 67.4% to 14.0%.

  • RSQ contrasts the top five retrieval results with ranks 6–20 to identify suspicious answer-anchor concentration and carrier transitions.
  • RSG compares documents with semantically similar but lexically distinct neighbors to detect coordinated poisoning before query time.
  • RSQ and RSG achieve 95.2% and 93.3% AUROC, respectively, outperforming the cited GMTP and CleanBase baselines.
  • The combined defense retains 41.3% F1 on unpoisoned retrieval.

Sources (1)

RAGSieve: Self-Referenced Local Contrast for Knowledge-Poison Detection in Retrieval-Augmented Generation

arXiv cs.CL Xinlong Xu, Yoshua Y. Li 2026-08-13 arXiv:2608.13010
Public signals Semantic Scholar citations 0 · Semantic Scholar influential citations 0
Providers: Hugging Face · N/A OpenAlex · N/A Publisher · N/A Semantic Scholar · Citations 0 · Influential citations 0 X · N/A Fetched 2026-09-12 14:26:10.127688 UTC

TL;DR - RAGSieve detects poisoned knowledge in retrieval-augmented generation using local comparisons within the inspected system, requiring neither trusted corpora nor poison labels. Its query-time and corpus-ingestion defenses jointly cut attack success from 67.4% to 14.0%.

  • RSQ contrasts the top five retrieval results with ranks 6–20 to identify suspicious answer-anchor concentration and carrier transitions.
  • RSG compares documents with semantically similar but lexically distinct neighbors to detect coordinated poisoning before query time.
  • RSQ and RSG achieve 95.2% and 93.3% AUROC, respectively, outperforming the cited GMTP and CleanBase baselines.
  • The combined defense retains 41.3% F1 on unpoisoned retrieval.
item →