Generalized Neural Memory (NeurIPS 2026)
Collection
Models, benchmarks, and eval records for 'Tell Me What To Learn' (NeurIPS 2026). • 9 items • Updated
Generalized Neural Memory — RAG-FT (CounterFACT)
The fine-tuned retrieval-augmented baseline for the CounterFACT benchmark: trained on the same data and objective as GNM, retrieving document-instruction pairs from a vector store at query time.
From "Tell Me What To Learn: Generalizing Neural Memory to be Controllable in Natural Language" (NeurIPS 2026).
Usage
Use through the code repo:
python scripts/download_models.py --counterfact # places this model at models/rag_ft_counterfact/model.pth
Registry key in the repo: llama3_8b_instruct_rag. See the repo README for the full CounterFACT
evaluation commands (Table 1) and training configs.
- Downloads last month
- 325
Model tree for maxbennett/generalized-neural-memory-ragft-counterfact
Base model
meta-llama/Meta-Llama-3-8B-Instruct