WISE
Rethinks knowledge memory for lifelong model editing. NeurIPS 2024.
At a glance
- Type
- Knowledge memory for lifelong editing
- Tier
- T3
- Created
- 2024-05
- 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
- parametric
- retrieval
- parametric-recall
- persistence
- parametric-permanent
- update
- parametric-edit
- unit
- fact
- governance
- inspectable
- conflict
- llm-arbitrate
When to use
Optimised for: not applicable - research paper
Anti-fit: not applicable - research paper
Pros & cons
Pros
Rethinks knowledge memory specifically for lifelong model editing; NeurIPS 2024.
Cons
Editing-quality vs general-task-performance tradeoff not fully characterized.
Claims & capabilities
Dual parametric memory scheme for lifelong model editing — main memory for pretrained knowledge plus side memory for edited knowledge with router and knowledge-sharding mechanism distributing edits across distinct subspaces before merging; outperforms previous editing methods on QA, hallucination correction, and OOD across GPT, LLaMA, Mistral; NeurIPS 2024
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.
Related systems
References (1)
- EWC (Elastic Weight Consolidation) cites — S2 isInfluential citation