Newest papers

Governance-State Corruption in Long-Term Memory for LLM Agents: A Cross-Model Stress Test

Bowen Liu, Youting Wang, Bonan Shen, Dingyan Shang · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.30698 · cited by 0 works (OpenAlex) · open-access PDF

German

Ein LLM-Agent kann ein Gedächtnis korrekt abrufen, aber es nach den Änderungen seines Gültigkeitszustands falsch verwenden. Wir untersuchen Governance-State-Korruption: Fehler, die gehaltene Aufzeichnungen in den Modell-sichtbaren Kontext einführen, während gespeicherte Inhalte und Aufgaben fest bleiben. Über sechs Anbieter produziert 731 benchmark-basierte Basen 2.080 übereinstimmende Fehlerpaare. In Offline-Control-Plane-Kontrollen ändert Korruption die operative Projektion bei 96.25% der Paare, während die oracle-unterstützte Durchsetzung sie bei 100% wiederherstellt. Für die nachgelagerte Bewertung löst ein deterministisch ausgewähltes, nicht-wahrscheinliches LongEval/Memora-Panel

Machine translation (Meta M2M100); the English original follows.

Abstract

An LLM agent can retrieve a memory correctly yet use it incorrectly after its validity state changes. We study governance-state corruption: faults that alter which retained records enter model-visible context while stored content and tasks remain fixed. Across six providers, 731 benchmark-grounded bases yield 2,080 matched fault pairs. In offline control-plane checks, corruption changes the operative projection in 96.25% of pairs, while oracle-backed enforcement restores it in 100%. For downstream evaluation, a deterministically selected, non-probability LongMemEval/Memora panel resolves 298 of 300 matched triplets (238 distinct bases). Clean-minus-corrupted contrasts are 40.54 percentage points in LongMemEval binary correctness (95% base-cluster interval [30.32, 50.71]) and 30.60 points in rescaled Memora FAMA ([23.31, 37.69]); both sources also show recovery. OAKS Novel native option parity holds for all 6,714 resolved outputs. Full-cohort frozen-scorer corruption/recovery effects (-27.55/+27.51 points) remain descriptive proxies, not validated native semantic population effects. These bounded findings motivate evaluating memory control-state integrity alongside retrieval accuracy.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Correctness, Population, Computer science, Context (archaeology), Econometrics, Psychology

Is this paper real? This page was built from OpenAlex and arXiv records, not generated. Software can verify any reference in one call: GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.30698, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txt

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange