Loading...
Loading...
Contributors:
Dataset URI:
10.5281/zenodo.23371368CreatorMemBench v1.0 contains 20 synthetic creator timelines and 160 checkpoints for evaluating mutable long-term memory in creator agents. The benchmark covers stable identity, scoped preferences, multi-session composition, transient noise, updates, forgetting, retention, and resurrection. The archive includes separated inputs and gold criteria, a strict semantic scoring protocol, an offline aggregation script, a prediction schema, a license, and SHA-256 checksums. It contains no production conversations, account identifiers, credentials, proprietary service code, endpoint identifiers, or raw internal traces.
Keyword(s):
agent evaluation; benchmark; creator assistants; LLM agents; long-term memory.
MeSH Term(s):
Grants and Funding:
No Grants or Funding Data Available
Associated Publication(s):
No Associated Publication Data Available
Supplementary Data:
No Supplementary Data Available
Related Dataset(s):
No Related Dataset Data Available
The National Library of Medicine (NLM) provides information about biomedical datasets. Inclusion on an NLM Web page does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: NLM Web Policies (nih.gov)