AI Paper Atlas

Research map / Large language models / AI and society

IDRBench: Understanding the Capability of Large Language Models on Interdisciplinary Research

Yuanhao Shen, Daniel Xavier de Sousa, Ricardo Marçal de Andrade Nascimento, Hongyu Guo, Xiaodan Zhu

ICML 2026 · Large language models › AI and society · approach: Critiques of LLM claims and judging · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add 5 few-shot examples to the promptLLMs on IPI · IDRBench IPIMacro-F1: gpt-4o-mini 0.587 -> 0.510; avg 0.556 -> 0.483Mixed
use reasoning models instead of non-reasoning modelsI3 idea integration · IDRBench I3 Subset 1Macro-F1: around 0.6-0.8 -> around 0.2-0.6Hurt
replace cross-discipline negatives with same-discipline negativesall LLMs on I3 · IDRBench I3 Subset 2 vs Subset 1Macro-F1: e.g. gpt-4o-mini 0-shot 0.750 -> 0.588Hurt
add 3 few-shot examples to the promptgpt-4o-mini on I3 Subset 2 · IDRBench I3 Subset 2Macro-F1: 0.588 -> 0.672Improved
add 3 few-shot examples to the promptGemini 2.0 on I3 Subset 1 · IDRBench I3 Subset 1Macro-F1: 0.666 -> 0.434Hurt
use full paper content instead of title and abstract (ablation)IDRBench tasks · IDRBenchperformanceNo clear effect
compare models on I2R ranking10 LLMs · IDRBench I2RMRR: approximately 0.670 MRRNo clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy