ICLR 2026 MemAgents Workshop

https://sites.google.com/view/memagent-iclr26/

Full-day hybrid workshop held April 27, 2026, Rio de Janeiro. Scope: episodic and semantic memory, working memory, parametric knowledge, KGs, vector DBs, retrieval pipelines, context management, long-context utilization, temporal credit assignment. Bridges RL, neuroscience, and cognitive psychology alongside the engineering track.

At a glance

Type
Workshop venue — RL + cognitive psychology + LLM memory
Tier
T2
Created
2026-04-27 (workshop held April 27 2026 at ICLR 2026 in Rio de Janeiro)
Latest release
not applicable — not OSS
License
not applicable — not OSS
GitHub
not applicable — no GitHub repo
Pricing
not applicable — not commercial
Funding
not applicable — not commercial

Taxonomy

storage
n/a
retrieval
n/a
persistence
n/a
update
n/a
unit
n/a
governance
n/a
conflict
n/a

When to use

Optimised for: not applicable - eval dataset, not a system

Anti-fit: not applicable - eval dataset, not a system

Pros & cons

Pros

Only dedicated memory-for-agents venue at a top-tier ML conference in 2026; neuroscience/RL bridge is a differentiator from purely engineering-oriented workshops.

Cons

Workshop proceedings not yet indexed in major databases; paper acceptance criteria and review quality vary across workshop tracks.

Claims & capabilities

Accepted papers include MemoryAgentBench (ICLR 2026 main track), CraniMem, and MemAgent oral. Full / short / tiny paper tracks.

Technical surface

API surface
not applicable — eval dataset, not a system
Backend storage
not applicable — eval dataset, not a system
Deployment
not applicable — not a deployable product
Embedding model
not applicable — eval dataset, not a system
Multi-tenancy
not applicable — eval dataset, not a system
MCP
not applicable — benchmark / evaluation harness
A2A
not applicable — benchmark / evaluation harness
OpenTelemetry
not applicable — benchmark / evaluation harness

Similar systems

Other memory benchmarks & evaluation in the catalog, ranked by inbound references.

  • LoCoMo T3

    Conversations spanning up to 35 sessions / 300 turns / ~9K tokens, built from LLM-agent personas grounded on temporal event graphs and verified by human annotators. Tests QA (single-hop, multi-hop, temporal, commonsense, adversarial), event-graph summarisation, multimodal dialogue.

  • LongBench T3

    Tsinghua THUDM. 21 datasets across 6 categories (single-doc QA, multi-doc QA, summarisation, few-shot ICL, synthetic, code completion) in English + Chinese. ~6.7K words English / 13.4K chars Chinese average. v2 at ACL 2025 with harder, longer instances.

  • LongMemEval T3

    5 core memory abilities — info extraction, multi-session reasoning, temporal reasoning, knowledge updates, abstention — via 500 curated questions in synthetic chat histories. Two scales: LongMemEval-S (115K tokens), LongMemEval-M (up to 1.5M).

  • RULER T3

    NVIDIA. Extends NIAH with multi-needle retrieval, multi-hop entity tracing, aggregation, QA — configurable for arbitrary length / difficulty. Evaluated 17 long-context models on 13 tasks.

  • InfiniteBench (∞Bench) T3

    Tsinghua / OpenBMB. First benchmark with average input >100K tokens. 12 tasks across English + Chinese: long novel QA, code debugging, math in long context, summarisation, retrieval. Realistic documents (novels, code) rather than synthetic filler.

  • Atari 100k T3

    Sample-efficiency benchmark protocol on Arcade Learning Environment (ALE) — agents limited to 100k environment steps (~2 hours of game-play) before evaluation. Introduced by SimPLe (Kaiser et al. 2019) and adopted by EfficientZero, IRIS, BBF, DreamerV3, Storm. Tests world-model / model-based methods where memory-as-imagination matters more than huge replay buffers.

Row last verified 2026-05-14. Catalog data is CC-BY-4.0 — see how to read this.