ScholarGym: Benchmarking Large Language Model Capabilities in the Information-Gathering Stage of Deep Research

arXiv '26 2026 21.1 benchmark

Read-first score

Read-first score 21.1, weighted from topical fit, citation, graph, method, reproducibility, and recency signals.

Recency 11%

100

Uses a gentle age decay so recent papers surface without erasing older foundations. 2026

Reproducibility 33%

Screens links and visible text for paper, code, dataset, artifact, and repository signals. pdf=True; code=False; dataset=False; markers=none

Topical relevance 56%

Matches configured research keywords against title, abstract, tags, and analysis text. matched=0

Frontier

Stability: volatile; rank range: 39.