AI Paper Atlas

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

TusoAI: Agentic Optimization for Scientific Methods

Alistair Turcan, Kexin Huang, Lei Li, Martin Jinye Zhang

ICLR 2026 · Large language models › Agents, code and math · approach: LLMs for scientific discovery · 5 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 domain knowledge (knowledge tree from papers) (ablation)TusoAI · ablation set averageaverage score: 0.46 -> 0.66Improved
add category structure for instructions (ablation)TusoAI · ablation set averageaverage score: 0.57 -> 0.66Improved
add Bayesian category sampling (ablation)TusoAI · ablation set averageaverage score: 0.55 -> 0.66Improved
add model diagnosis step (ablation)TusoAI · ablation set averageaverage score: 0.64 -> 0.66Improved
use GPT-5 instead of GPT-4o-mini backbone (ablation)TusoAI · ablation set averageaverage score: 0.66 -> 0.69No clear effect
use Claude-4-Sonnet instead of Claude-3.5-Haiku backbone (ablation)TusoAI · ablation set averageaverage score: 0.68 -> 0.61Hurt
use open-source GPT-oss-120b backbone (ablation)TusoAI · ablation set averageaverage score: 0.66 -> 0.54Hurt
use TusoAI instead of AIDEsingle-cell tasks · Single-cell benchmarksaverage score: 0.52 -> 0.66Improved
use TusoAI instead of DARTSNASBench-360 tasks · NASBench-360 averageaverage score: 0.71 -> 0.70Hurt
optimize scDRS scoring function with TusoAIscDRS · real-data cell type-disease associationstrue associations found: 14 -> 17Improved
optimize pgBoost distance features with TusoAIpgBoost · fine-mapped eQTL linksenrichment: +13.8% enrichment (eQTL), +7.2% (ABC)Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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