Beyond Memory Leaderboards Evaluating Scientific M

# Beyond Memory Leaderboards: Evaluating Scientific Memory as Budgeted Context Restoration

Derived from arXiv:2607.16848 - Beyond Memory Leaderboards: Evaluating Scientific Memory as Budgeted Context Restoration

## Core Concept
Long-term memory is becoming a core component of LLM agents, but most memory benchmarks evaluate conversations or compact summaries, while research agents need to restore evidence from full scientific papers. We introduce two full-text scientific-memory benchmarks, Public AI Memory (PAIM; 81 papers, 66 questions) and Public Transformers (PTr; 252 papers, 98 questions). We evaluate eight memory/retrieval systems, including our own proposed Theoria, plus a no-retrieval baseline. Our results show t...

## Key Insights
- Derived from arXiv:2607.16848
- Published: 2026-07-18
- Utility Score: 1.00
- Authors: Maksim Sheverev, David Finkelstein, Sergey Nikolenko

## Activation
beyond-memory-leaderboards-evaluating-scientific-m, 2607.16848

## References
- arXiv: https://arxiv.org/abs/2607.16848