MMAG
https://arxiv.org/abs/2512.01710
Mixed memory-augmented generation for LLM applications.
At a glance
- Type
- Mixed memory-augmented generation
- Tier
- T4
- Created
- 2025-12
- 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
- hybrid
- retrieval
- similarity
- persistence
- long-term
- update
- consolidation
- unit
- chunk
- governance
- user-controllable
- conflict
- llm-arbitrate
When to use
Optimised for: not applicable - research paper
Anti-fit: not applicable - research paper
Pros & cons
Pros
Multi-modal memory agent grounding.
Cons
Preprint; multi-modal eval limited.
Claims & capabilities
Mixed Memory-Augmented Generation framework organizing LLM-agent memory into five interacting layers (conversational, long-term user, episodic and event-linked, sensory and context-aware, short-term working memory); implementation in Heero conversational agent shows encrypted long-term bios and conversational history improve engagement and retention; abstract provides no quantitative metrics
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.