AI Paper Atlas

Research map / Learning theory and algorithms / Prediction and decision losses

Procedural Generation Of Algorithm Discovery Tasks in Machine Learning

Alexander David Goldie, Zilin Wang, Adrian Hayler, Deepak Nathani, Edan Toledo, Ken Thampiratwong, Aleksandra Kalisz, Michael Beukman, Alistair Letcher, Shashank Reddy Chirra, Clarisse Wibault, Theo Wolf and 7 more

ICML 2026 · Learning theory and algorithms › Prediction and decision losses · approach: Active learning and subset selection

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
increase tasks used for prompt optimisation from 1 to 30 (ablation)ADA prompt tuning (Deepseek-v3.2) · In-distribution domainsmeta-test Elo: 965 -> 1067Improved
increase tasks used for prompt optimisation from 1 to 30 (ablation)ADA prompt tuning (Deepseek-v3.2) · Held-out domainsmeta-test Elo: 983 -> 1024Improved
increase tasks used for prompt optimisation from 1 to 30 (ablation)ADA prompt tuning (Deepseek-v3.2) · In-distribution domainsmeta-train Elo: 955 -> 1047No clear effect
start modules from baseline implementations instead of empty (ablation)MLGym ADA (Deepseek-v3.2) · Quick-to-run domainssuccess rate: 75.4% -> 92.4%Improved
start modules from baseline implementations instead of empty (ablation)MLGym ADA (Deepseek-v3.2) · Quick-to-run domainsElo: 899 -> 1101Improved
increase number of editable modules in On-Policy RL (ablation)MLGym ADA · On-Policy RL module sweep (63 tasks)success rateHurt
edit all modules instead of a single module (ablation)MLGym ADA (GPT-OSS 120B) · DiscoBenchsuccess rate: 59.5% -> 17.8%Hurt
edit all modules instead of a single module (ablation)MLGym ADA (Deepseek-v3.2) · DiscoBenchsuccess rate: 71.3% -> 40.5%Hurt
evaluate on meta-test instead of meta-train for LLM ADAs versus fixed baselineMLGym ADA · DiscoBenchElo gap to baselineMixed
use Success@3 instead of aggregated success rateMLGym ADA · DiscoBenchsuccess rate: 10-30 percentage points higherImproved

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