arXiv：cs.AI· Vishisht Choudhary·· 16 小时前
PolyGraph：为 Agent 构建的知识图谱做可信清理，摘要层不可信可降至 0.4% 污染查询
Keep Your Receipts: Trustworthy Cleanup of Agent-Built Knowledge Graphs
AI 导读
针对 Agent 用知识图谱做记忆时源撤回难以清理的问题，PolyGraph 要求构建器在生成每个产物时附带“收据”，即多组各自充分的替代证据，清理时只做集合运算、无需调用模型。
正文
Abstract:Agents build knowledge graphs as memory, and building them has become easy. Maintaining them has not. When a source is retracted, overruled, or erased, everything built on top of it (entity merges, entailed edges, community summaries) must be re-examined, and the maintainer must choose between deleting too much and too little. Deployed pipelines delete only the directly extracted facts: on our corpus that reduces contaminated queries from 32.2% to 28.0% under a real retraction and changes nothing under predicate revocation. Rebuilding is no repair for a constructor that samples: two builds of the same 300-document corpus differ in 29,475 artifacts, five times the 6,248 that removing the 61 retracted documents changes. PolyGraph instead asks the constructor for receipts: for every artifact, alternative bundles of source evidence, each claimed sufficient on its own, emitted in the call that creates it. Cleanup is then set arithmetic over the receipts, with no model calls. Because receipts are the model's own self-report, a re-derivation audit checks them by deleting the cited evidence and testing whether the constructor still produces the artifact. The audit finds that a receipt is trustworthy when it is shaped like the claim it backs. Merge receipts, each backing one narrow claim, agree with re-derivation 68% and 85% of the time. Summary receipts, elicited per sentence but stored as one flat list for the whole text, are wrong about two thirds of the time, almost always by keeping dead content. One receipt per sentence instead is 100% and 88% faithful and lets cleanup keep 33 of 35 touched summaries. The result holds under a second oracle model, a claim-blind prompt, ten further deletions, a second domain and a second constructor. Distrusting the summary layer cuts contaminated queries to 0.4% and 0.8% at 99.8 to 100% of the recall a perfect maintainer would return.
| Comments: | 40th Conference on Neural Information Processing Systems (NeurIPS 2026). Workshop: Who Verifies the Agents? Toward Reliable Agent Development |
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.17719 [cs.AI] |
| (or arXiv:2610.17719v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.17719 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Vishisht Choudhary [view email]
[v1]
Tue, 13 Oct 2026 04:51:28 UTC (36 KB)
来源：arXiv：cs.AI · arxiv.org