AI Paper Atlas

Research map / Large language models / Agents, code and math

Towards Multimodal Data-Driven Scientific Discovery Powered by LLM Agents

Fan Liu, Xiaozhao Zeng, Hao Liu

ICLR 2026 · Large language models › Agents, code and math · approach: Agent instruction-following benchmarks

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 task-provided domain knowledge (ablation)ReAct (o4-mini) · MoSciBench avgaccuracy: 48.4% -> 44.9%Hurt
add workflow scaffolding (ablation)ReAct (o4-mini) · MoSciBench avgaccuracy: 48.4% -> 54.1%Improved
add workflow scaffolding (ablation)ReAct (o4-mini) · MoSciBench Climate Sci.accuracy: 57.1% -> 71.4%Improved
add workflow scaffolding (ablation)ReAct (o4-mini) · MoSciBench Earth Sci.accuracy: 35.7% -> 50.0%Improved
add domain knowledge (ablation)ReAct (o4-mini) · MoSciBench Climate Sci.accuracy: 57.1% -> 50.0%Hurt
add domain knowledge (ablation)ReAct (o4-mini) · MoSciBench Cheminformaticsaccuracy: 33.3% -> 26.7%Hurt
add workflow scaffolding (ablation)ReAct (o4-mini) · MoSciBench alignment errorsaccuracy: 31.8% -> 27.3%Improved
give agent data access instead of NoDataGuesso4-mini · MoSciBench overallaccuracy: 0.105 -> 0.484Improved
use stronger base model (gpt-5-mini to o4-mini)ReAct · MoSciBench overallaccuracy: 0.174 -> 0.484Improved
add retrieval of external knowledge (RAG-ReAct vs ReAct)gpt-5-mini · MoSciBench overallaccuracy: 0.174 -> 0.128Hurt
add oracle-feedback retries (Reflexion vs ReAct)o4-mini · MoSciBench overallaccuracy: 0.484 -> 0.458Hurt
increase Best-of-N rollouts beyond N = 3ReAct (DeepSeek-V3.1) · MoSciBenchaccuracyMixed

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