E${}^2$MEND: Evidence-guided Evolution for Memory Enhancement via Bottleneck Diagnosis

Long-term memory failures are heterogeneous: evidence may be lost, unretrieved, insufficiently structured, or available while the answer model still fails; a uniform update can target the wrong bottleneck and introduce regressions. We present E${}^2$MEND, a diagnosis-guided framework for bounded memory-strategy evolution: it estimates the bottleneck, mutates versioned strategy artifacts through a bounded LLM evolver under a locked runtime, and adopts candidates via paired evaluation, negative evidence, and rollback. It matches failure modes to bounded edits—extraction or provenance for ingestion gaps, support selection for retrieval gaps, typed state for state construction, and no memory mutation for answer-model limitations. On a fixed 1,540-question LoCoMo slice, a staged trajectory raises accuracy from 0.756 to 0.947 (final paired checkpoint: 16 repairs, 0 regressions); on MemBench, matched strategies yield a descriptive $0.770\to 0.863$. We frame these as diagnostic and strategy claims on fixed slices, not public-task superiority.