MemEvolve
https://arxiv.org/abs/2512.18746
Meta-evolution of agent memory systems. Includes EvolveLab — a unified self-evolving memory codebase distilling 12 representative memory systems.
Active · Verified 2026-10-28 · fresh as of 2026-10-28 · 19 cited facts, 0 with a third-party source · Suggest a correction
At a glance
- Type
- Meta-evolution of memory systems vendor fix
- Tier
- T4
- 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
- 2026-10-28 fix
- Layer
- agent-infra repo fix
Memory taxonomy
- storage
- file
- retrieval
- agentic
- persistence
- cross-session
- update
- consolidation
- unit
- episode
- governance
- inspectable
- conflict
- llm-arbitrate
When to use
Optimised for: not applicable - research paper
Anti-fit: not applicable - research paper
Pros & cons
Pros
Meta-evolution of memory systems via EvolveLab — distills 12 memory designs.
Cons
Preprint; meta-evolution is computationally expensive.
Claims & capabilities
Meta-evolution of agent memory architecture itself, not just experiential knowledge; EvolveLab self-evolving memory codebase with encode/store/retrieve/manage modules. Headline: SmolAgent and Flash-Searcher gains up to 17.06%; baselines: SmolAgent and Flash-Searcher frameworks; primary datasets: 4 challenging agentic benchmarks.
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).
- What it is Meta-evolution of memory systems vendor
- What it does & what's distinct Meta-evolution of agent memory systems. Includes EvolveLab — a unified self-evolving mem… vendor
- Claims / benefits Meta-evolution of agent memory architecture itself, not just experiential knowledge; Evo… vendor
- Modalities text vendor
- Created 2025-12 vendor
- Mindshare 25 cites · 5.6/mo · 212* / 25 forks · HuggingFace papers feature; X/Twitter coverage 202… paper
- Citations 25 cites 25/yr paper
- Founders / pedigree no data — searched arxiv.org/abs/2512.18746; author affiliations not found in preview vendor
- Memory volume / scale no scale claim — research (12 systems distilled) vendor
- Performance +17.06% SmolAgent / Flash-Searcher gains (4 agentic benchmarks) vendor
- Reproducibility Too recent — published 2025-12-21 (4.5 mo ago); no reproductions yet expected vendor
- Code/weights release Code public + weights closed — https://github.com/bingreeky/MemEvolve · 212* / 25 forks … repo
- Contradiction handling LLM resolves and overwrites vendor
- Pros Meta-evolution of memory systems via EvolveLab — distills 12 memory designs. vendor
- Cons Preprint; meta-evolution is computationally expensive. vendor
- Project & sources arxiv 2512.18746 vendor
- citation-trajectory [{"year":2026,"cum":1,"infl":0}] paper
- taxonomy-three-layer-rollup agent-infra repo
Related systems
References (3)
- Alita cites — S2 isInfluential citation
- ExpeL cites — S2 isInfluential citation
- SkillWeaver cites — S2 isInfluential citation