maxbennett/generalized-neural-memory-personamem
GNM PersonaMem Benchmark The PersonaMem preference-selectivity benchmark from "Tell Me What To Learn: Generalizing Neural Memory to be Controllable in Natural Language" (NeurIPS 2026) — a multiple-choice variant of PersonaMem-v2 where a memory model must learn user preferences for instructed topics while ignoring preferences for non-instructed topics. Download with python scripts/download_data.py --personamem from the code repo; everything lands under data/personamem/. Split… See the full description on the dataset page: https://huggingface.co/datasets/maxbennett/generalized-neural-memory-pers
| Author / organization | maxbennett |
|---|---|
| License (as declared) | mit |
| Downloads | 332 (as reported by Hugging Face) |
| Likes | 0 |
| Created | 14 October 2026 |
| Last updated at source | 15 October 2026 |
Popularity, last 3 days
332 now (+7 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.
Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 18 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.