Issues

• eval(veracium): longitudinal validity-interval + provenance/injection eval

  verygoodplugins/automem-evals#62 · Filter by author autohub-codex opened on Oct 9, 2026

• eval(atma): track LTP ghost-memory benchmark (blocked on public artifacts)

  verygoodplugins/automem-evals#57 · Filter by author autohub-codex opened on Oct 4, 2026

• track: LongMemEval-V2 web-agent experience benchmark

  verygoodplugins/automem-evals#54 · Filter by author autohub-codex opened on Oct 2, 2026

• eval(merit): action-producing, cost-aware AutoMem adapter

  verygoodplugins/automem-evals#53 · Filter by author autohub-codex opened on Oct 2, 2026

• eval(memtrace): evidence-use diagnostic adapter (blocked on public artifacts)

  verygoodplugins/automem-evals#51 · Filter by author autohub-codex opened on Oct 2, 2026

• StateMemBench: current vs superseded state eval

  verygoodplugins/automem-evals#50 · Filter by author autohub-codex opened on Oct 1, 2026
  3

• eval(protocol): execute MemDelta-controlled AutoMem baselines on LongMemEval-S and Mem2ActBench

  verygoodplugins/automem-evals#48 · Filter by author jack-arturo opened on Sep 30, 2026

• GAM hierarchical graph memory: architecture comparison + LoCoMo-50 eval

  verygoodplugins/automem-evals#33 · Filter by author jack-arturo opened on Sep 18, 2026

• MemoryAgentBench (ICLR 2026): 4-competency incremental multi-turn eval

  verygoodplugins/automem-evals#32 · Filter by author jack-arturo opened on Sep 18, 2026

• ForgetEval / ForgetEval-Adv: control-plane forgetting (supersede/release/purge) eval

  verygoodplugins/automem-evals#31 · Filter by author jack-arturo opened on Sep 18, 2026
  3

• MemoryCD (ICLR 2026 workshop): cross-domain personalization eval for AutoMem relation linking

  verygoodplugins/automem-evals#30 · Filter by author jack-arturo opened on Sep 18, 2026

• AML Cycle 2: AutoMem adapter readiness (opens 2026-09-20)

  verygoodplugins/automem-evals#28 · Filter by author jack-arturo opened on Sep 18, 2026
  4

• AutoMem: AMA-Bench eval adapter + ICML 2026 leaderboard submission

  verygoodplugins/automem-evals#27 · Filter by author jack-arturo opened on Jun 29, 2026
  2

• AutoMem vs PlugMem: knowledge-centric graph architecture comparison + eval run

  verygoodplugins/automem-evals#26 · Filter by author jack-arturo opened on Jun 29, 2026

• feat(automem): Evaluate AutoMem against PersonaMem-v2 — DCPM gap analysis

  verygoodplugins/automem-evals#25 · Filter by author jack-arturo opened on Jun 29, 2026

• feat(automem): Evaluate against Mem2ActBench — proactive tool-calling memory (arXiv:2601.19935)

  verygoodplugins/automem-evals#24 · Filter by author jack-arturo opened on Jun 29, 2026
  2

• feat(automem): FAMA evaluation harness — measure forgetting-aware accuracy against Memora benchmark (ACL 2026)

  verygoodplugins/automem-evals#22 · Filter by author jack-arturo opened on Jun 27, 2026
  1

• Cold vs matured benchmark methodology (maturation-v1): simulate consolidation/decay on the narrative clock

  verygoodplugins/automem-evals#18 · Filter by author jack-arturo opened on Jun 16, 2026

• Wire the BEAM retrieval proxy into nightly CI as the recall regression gate

  verygoodplugins/automem-evals#17 · Filter by author jack-arturo opened on Jun 16, 2026

• BEAM summarization: top-k recall can't cover 'complete summary' questions

  verygoodplugins/automem-evals#16 · Filter by author jack-arturo opened on Jun 16, 2026

• BEAM abstention: answerer hallucinates instead of refusing (19/40 fail)

  verygoodplugins/automem-evals#15 · Filter by author jack-arturo opened on Jun 16, 2026

• BEAM knowledge_update: exercise AutoMem's update/invalidation machinery (stale-value misses)

  verygoodplugins/automem-evals#14 · Filter by author jack-arturo opened on Jun 16, 2026