analiza aiqbench
This research addresses a critical failure mode in autonomous agents where inherited memory constraints become stale due to superseded records, leading to erroneous decision-making under limited verification budgets. The study demonstrates that strategically reallocating a single verification slot to the critical provenance path significantly reduces these errors across multiple models and domains without increasing overall resource consumption. This work highlights the necessity of distinct freshness or supersession signals in memory architectures to maintain agent reliability in dynamic environments.
When Stale Constraints Go Unchecked: Budgeted Verification Failures in Inherited Agent Memory
arXiv:2608.25553v1 Announce Type: cross Abstract: An agent that inherits a consolidated memory may inherit a constraint that was true when written and has since been withdrawn by a newer authoritative record. Under a scarce verification budget, does the agent recover the withdrawal, and if not, is the error avoidable without spending more? We model supersession explicitly -- historical provenance is immutable; what changes is which record is current -- and assign by design the memory's form, the world's state (source current or superseded), and the verification policy at a fixed budget of two records: the agent's own allocation, or the same budget with one slot re-assigned to the critical provenance path or to a random record. With a constraint stated, agents inspected its provenance path in about one episode in five; when that constraint had been superseded, native allocation produced stale-consistent decisions in 77.3%, 74.7% and 74.7% of episodes across a primary run, a fresh-wording replication and a held-out domain. Re-assigning one slot to the critical path raised current-record-consistent decisions by +74.0, +72.7 and +61.3 points, positive in six of six models in each of those runs, and changed nothing when the record agreed with the memory. The held-out scenario was later found to contain a temporal inconsistency; a robustness replication with one sentence corrected, deposited externally before execution, gave +73.3 points and is reported alongside the original. The intervention uses knowledge of the critical path and is not a scheduler; it identifies that the share of stale-memory error attributable to verification allocation is close to its structural ceiling. Memory systems may need freshness or supersession signals separate from relevance.
Abstract:An agent that inherits a consolidated memory may inherit a constraint that was true when written and has since been withdrawn by a newer authoritative record. Under a scarce verification budget, does the agent recover the withdrawal, and if not, is the error avoidable without spending more? We model supersession explicitly -- historical provenance is immutable; what changes is which record is current -- and assign by design the memory's form, the world's state (source current or superseded), and the verification policy at a fixed budget of two records: the agent's own allocation, or the same budget with one slot re-assigned to the critical provenance path or to a random record. With a constraint stated, agents inspected its provenance path in about one episode in five; when that constraint had been superseded, native allocation produced stale-consistent decisions in 77.3%, 74.7% and 74.7% of episodes across a primary run, a fresh-wording replication and a held-out domain. Re-assigning one slot to the critical path raised current-record-consistent decisions by +74.0, +72.7 and +61.3 points, positive in six of six models in each of those runs, and changed nothing when the record agreed with the memory. The held-out scenario was later found to contain a temporal inconsistency; a robustness replication with one sentence corrected, deposited externally before execution, gave +73.3 points and is reported alongside the original. The intervention uses knowledge of the critical path and is not a scheduler; it identifies that the share of stale-memory error attributable to verification allocation is close to its structural ceiling. Memory systems may need freshness or supersession signals separate from relevance.
18 pages, 3 figures, 7 tables. Manuscript, LaTeX source, all 5,400 episode files, frozen specifications, timestamp proofs, analysis and generator scripts are archived at Zenodo: doi:https://doi.org/10.5281/zenodo.22108558
Information Retrieval (cs.IR); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
arXiv:2608.25553 [cs.IR]
(or arXiv:2608.25553v1 [cs.IR] for this version)
https://doi.org/10.48550/arXiv.2608.25553 arXiv-issued DOI via DataCite (pending registration)
Text integral sub licența deschisă a publicației · 316 cuvinte
Subiecte conexe
Hegseth AutoWarCom: who builds it, what the money buys, and why its hardest problem after Minab is the record, not the model
livetradingnews.com · 2 oct. 2026
[AINews] not much happened today
Latent Space · 18 sept. 2026
Lattice-Based Cryptography Explained
Quantum Zeitgeist · 15 aug. 2026
The Expensive-Loop Problem
LessWrong · 12 oct. 2026