AI Paper Atlas

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

Theory-Grounded Evaluation of Human-Like Fallacy Patterns in LLM Reasoning

Andrew Keenan Richardson, Ryan Othniel Kearns, Sean Moss, Vincent Wang, Philipp Koralus

ICLR 2026 · Large language models › Agents, code and math · approach: Inductive and abstract reasoning evaluation · 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
increase model strength (higher Chatbot Arena Elo)38 LLMs · ETR fallacy problemsfallacy share of incorrect answers: r = 0.364, p = 0.0247Improved
increase model strength (Elo), exponential fit38 LLMs · ETR fallacy problemsfallacy share of incorrect answers: r = 0.407, p = 0.0113Improved
increase model strength (Elo), Spearman38 LLMs · ETR fallacy problemsfallacy share of incorrect answers: rho = 0.360, p = 0.0265Improved
increase model strength (higher Elo)38 LLMs · ETR fallacy problemslogical correctness: r = 0.004, p = 0.981No clear effect
increase model strength (higher Elo)38 LLMs · ETR fallacy problemsabsolute proportion of fallaciesNo clear effect
reverse premise orderclaude-3.5-sonnet · ETR fallacy problemsfallacies blocked: 65.08% blocked (z = 4.74, p = 2.09e-06)Improved
reverse premise ordergpt-3.5-turbo-1106 · ETR fallacy problemsfallacies blocked: 88.46% blocked (z = 4.36, p = 1.33e-05)Improved
reverse premise orderclaude-3.7-sonnet · ETR fallacy problemsfallacies blocked: 50.00% blocked (z = 4.23, p = 2.35e-05)Improved
reverse premise ordero1-mini · ETR fallacy problemsfallacies blocked: 50.00% blocked (z = 0.58, p = 5.63e-01)No clear effect
reverse premise ordergemma-3-27b-it · ETR fallacy problemsfallacies blocked: 20.39% blocked (z = 1.77, p = 7.60e-02)No clear effect
reverse premise orderllama-3.2-3b-instruct · ETR fallacy problemsfallacies blocked: 20.00% blocked (z = 0.77, p = 4.41e-01)No clear effect
reverse premise ordermistral-7b-instruct-v0.1 · ETR fallacy problemsfallacies blocked: 66.67% blocked (z = 1.42, p = 1.55e-01)No 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