The paper introduces the concept of State Integrity Collapse, where large language models remember a constraint but fail to apply it in later dialogue turns. It presents StateTrack‑Bench, a suite of controlled conversations testing numerical, format, role/persona, factual‑memory, and logical‑rule constraints across multiple model snapshots and dialogue depths. The study finds that while strict joint retention can be high, failures are largely due to behavioral drift rather than simple forgetting, and that reliability must be assessed over long‑horizon dialogues rather than single‑turn instruction following.
arXiv:2607. 12893v1 Announce Type: new Abstract: Long-term memory has become a foundational capability for LLM-based agents that accompany users across extended, multi-session interactions.
arXiv:2606. 25449v1 Announce Type: cross Abstract: A language model's memory can be worse than having no memory at all.
StateWise is a method that trains a single large language model to maintain and update its own constraint state during multi‑turn dialogue. By fine‑tuning the model to summarize conversations into a typed state with tombstones for revoked constraints and then self‑training on trajectories that pass exact round‑trip checks, it reduces relapse and improves conversation success. The approach outperforms a compiled‑ledger baseline on relapse without requiring inference‑time scaffolding, and it is evaluated on two backbones with held‑out constraint families and unseen phrasings.
A language model's memory can be worse than having no memory at all. Give a model a memory that kept a wrong conclusion but dropped the work behind it, and it emits that stale value as a confident answer; give the same model an empty memory and it abstains.
arXiv:2609. 04875v1 Announce Type: cross Abstract: Long-running LLM agents are stateful: beyond the transcript they accrete compressed summaries, plaintext memory, pending tool plans, and, under every serving API, a KV cache.
arXiv:2605.27186v2 Announce Type: replace Abstract: Large language models often solve tasks from a fully specified prompt but degrade when the same requirements unfold over multiple turns, known as t...
arXiv:2607. 01935v1 Announce Type: new Abstract: Long term memory lets LLM agents act as persistent assistants, but user facts change.
arXiv:2603. 23530v2 Announce Type: replace-cross Abstract: Large language models often fail to satisfy formatting instructions when they must simultaneously perform demanding tasks.
arXiv:2602. 01146v2 Announce Type: replace Abstract: Conversational assistants are increasingly integrating long-term memory with large language models (LLMs).
arXiv:2609.16073v1 Announce Type: cross Abstract: Retrieval-Augmented Generation (RAG) serves as the primary memory architecture for long-horizon autonomous agents. However, treating shared memory as...
MemCut introduces a benchmark to test user-requested forgetting in large language model agents, featuring 456 hand-designed secrets embedded in long-horizon conversations with fully annotated evidence trails. The benchmark defines two recovery metrics—native and residual—to assess how well secrets can be retrieved after deletion. Experiments across five memory designs and four LLMs show that deletion reduces native recovery more than residual recovery, highlighting candidate coverage as a key bottleneck.
TWIST is a new benchmark suite designed to evaluate the quality of interventions in conversational memory systems, focusing on how well these systems act correctly at belief change points. It includes four tracks—tension detection, draft vetting, belief-consistent answering, and sensitive recall governance—each paired with hard-negative controls to prevent gaming. The benchmark is rigorously validated through double annotation, adjudication, and a separability audit, revealing that current models face trade-offs between contradiction recall and false-positive rates.