LLM agents struggle to detect and act on implicit memory conflicts, with top models scoring 55.2% on the new STALE benchmark of 400 scenarios; CUPMem prototype strengthens state-aware revision.
Evaluating very long-term conversational memory of LLM agents
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
STALE: Can LLM Agents Know When Their Memories Are No Longer Valid?
LLM agents struggle to detect and act on implicit memory conflicts, with top models scoring 55.2% on the new STALE benchmark of 400 scenarios; CUPMem prototype strengthens state-aware revision.