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

Memformer

External key-value stores with similarity-based cache management; trained with Memory Replay Backpropagation.

At a glance

Type: External KV + Memory Replay Backprop
Tier: T3
Section: Recent method papers — theorized, no distinct product
Created: 2022-12
Latest release: no-release no-release
License: MIT
GitHub: 9★ Python
Pricing: not applicable — research paper
Funding: not applicable — not commercial

Taxonomy

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

When to use

Optimised for: not applicable - research paper

Anti-fit: not applicable - research paper

Pros & cons

Pros

External KV stores with similarity-based cache management, trained with Memory Replay Backprop.

Cons

2020 paper — superseded in production by retrieval-augmented approaches.

Claims & capabilities

External dynamic memory transformer for sequence modeling; memory replay back-propagation (MRBP) for long-range BPTT. Headline: 8.1x less memory and 3.2x faster inference vs. standard Transformers with comparable performance; baseline: standard Transformers; primary dataset: not specified in abstract (long-sequence language modeling).

Technical surface

API surface: not applicable — research paper
Backend storage: not applicable — research paper
Deployment: not applicable — research paper
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.