MAPLE - Motivation Queries: leaderboard
MAPLE (Zhejiang University and Westlake University, 2026) is a multi-aspect full-paper scientific retrieval benchmark testing whether a retriever recovers the same paper from queries targeting its motivation, method and findings, over 2,095 queries on ICLR 2026 papers. Recall@20 on motivation-aspect queries, in percent, higher is better.
Metric: Recall@20 (%; over MAPLE's 2,095 multi-aspect full-paper scientific-retrieval queries (ICLR 2026 papers); motivation-aspect queries). Source: arxiv.org. Saturation forecast: Around February 2027. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GritLM-7B | 79 | #6350 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=maple-motivation-queries · How It Works · Data refreshed daily, snapshot 2026-11-12.