AI Paper Atlas

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

Evaluating Robustness of Reasoning Models on Parameterized Logical Problems

Naïm Es-sebbani, Esteban Marquer, Yakoub Salhi, Zied Bouraoui

ICML 2026 · Large language models › Agents, code and math · approach: Inductive and abstract reasoning evaluation · 0 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
replace template verbalizer with LLM-written narratives (ablation)reasoning models on 2-SAT · 2-SAT diagnostic benchmarkdecision accuracy: drops by about 25 pointsHurt
add filler clauses at fixed core size (ablation)reasoning models · EquivalenceCore, Backbone, MonoBridgeaccuracy: filler significantly hurts (|C|=75 r=25 vs 75,0)Hurt
shuffle clause order instead of construction order (ablation)four main models · all generatorsaccuracy: p<0.003Hurt
shuffle clause order (ablation)four main models · EquivalenceCoreaccuracyImproved
shuffle clause order (ablation)four main models · ImplicationCycleaccuracy: p=0.035Mixed
vary bridge position (ablation)models · MonoBridge (75 clauses)accuracy: p=7.788e-1No clear effect
vary overall variable density (ablation)models · MonoBridgeaccuracy: p=2.073e-7Mixed
vary positive-part ratio toward balanced split (ablation)models · MonoBridge (75 clauses)accuracy: p<0.05Hurt
increase backbone fraction beta from 0 to 1 (ablation)models · Backbone (75 clauses)accuracy: drop of around 10%Hurt
increase monotone density (ablation)models · Backbone (75 clauses)accuracy: increase of up to 20% for the worst modelImproved
flip polarity pi of monotone part (ablation)models · Backboneaccuracy: all p-values > 0.9No clear effect
increase free-to-bound ratio (ablation)models · EquivalenceCore (20 clauses)accuracy: p<0.005Hurt

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