Catalog Systems Recent method papers — theorized, no distinct product MemReasoner

MemReasoner

https://openreview.net/forum?id=ODcMy97cVZ

IBM Research. Memory-augmented LLM architecture for multi-hop reasoning. Iteratively re-reads a temporal latent memory with bidirectional GRUs until readout stabilises. NeurIPS 2024.

Active · Verified 2026-05-14 · fresh as of 2026-05-14 · 19 cited facts, 0 with a third-party source · Suggest a correction

At a glance

Type
Iterative bidirectional-GRU re-read vendor fix
Tier
T3
Section
Recent method papers — theorized, no distinct product
Created
2024-10-04 (submitted to NeurIPS 2024 OpenReview Oct 4 2024; published Dec 10 2024) vendor fix
Latest release
not applicable — not OSS fix
License
not applicable — not OSS fix
GitHub
not applicable — no GitHub repo fix
Pricing
not applicable — not commercial fix
Funding
not applicable — not commercial fix
Freshness
2026-05-14 fix
Layer
agent-infra repo fix

Memory taxonomy

storage
kv-cache
retrieval
attention
persistence
session
update
overwrite
unit
kv-token
governance
n/a
conflict
n/a

When to use

Optimised for: not applicable - research paper

Anti-fit: not applicable - research paper

Pros & cons

Pros

Reasoning over memory store as a discrete inference step.

Cons

Research-stage; reasoning quality bounded by base model.

Claims & capabilities

+18% on multi-hop QA. Generalises better than off-the-shelf LLMs and existing recurrent models on tests where required facts are scattered up to 128k tokens.

Technical surface

API surface
not applicable — research paper
Backend storage
not applicable — research paper
Deployment
not applicable — not a deployable product
Embedding model
not applicable — research paper
Multi-tenancy
not applicable — research paper
MCP
not applicable — research paper, no deployed product
A2A
not applicable — research paper, no deployed product
OpenTelemetry
not applicable — research paper, no deployed product

Similar systems

Other recent method papers — theorized, no distinct product in the catalog, ranked by inbound references.

  • Compressive Transformer T3

    Maintains recent states in full resolution while compressing older memories with learned compression functions. DeepMind.

  • MemGPT v2 / agent-tools T3

    Already in catalog as the foundational MemGPT paper. Note: Letta is the productionised successor (cross-listed).

  • Transformer-XL T3

    Extends context through segment-level recurrence + caching of hidden states from prior segments. Foundational long-context architecture.

  • Generative Agents T3

    Park et al. — landmark agent-simulation paper. Reflection + memory stream + retrieval enable believable agent behavior.

  • MemoryBank T3

    Enhances LLMs with long-term memory. Early influential paper.

  • Reflexion T3

    Language agents with verbal reinforcement learning.

Sources (18 cited facts, 0 third-party)

Each fact links to the page it came from; the badge says who published it (how to read sources).