mimirs-eval 0.1.5

EvolMem-inspired memory evaluation benchmark for MimirsWell
Documentation

mimirs-eval: EvolMem-inspired memory evaluation benchmark for MimirsWell

Implements systematic evaluation of memory quality across 7 cognitive dimensions from the EvolMem framework (Shen et al., 2026, arXiv:2601.03543):

Declarative Memory:

  1. Retrieval — accuracy of recalling relevant information
  2. Summarization — quality of memory consolidation/abstraction
  3. Isolation — prevention of cross-source memory interference
  4. Inference — reasoning from stored facts
  5. Reproduction — faithful reconstruction of past interactions

Non-Declarative Memory: 6. Learning — acquisition of operational rules from experience 7. Habituation — stability of automatic memory patterns across sessions