AI Paper Atlas

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

Mathematical Capabilities of ChatGPT

Simon Frieder, Luca Pinchetti, Alexis Chevalier, Ryan-Rhys Griffiths, Tommaso Salvatori, Thomas Lukasiewicz, Philipp Christian Petersen, Julius Berner

NeurIPS 2023 · Large language models › Agents, code and math · approach: Reasoning benchmarks and platforms

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
Upgrade ChatGPT from the 9-January-2023 version to the 30-January-2023 versionChatGPT · GHOSTSaverage rating: 3.20 -> 3.29No clear effect
Use GPT-4 instead of ChatGPTChatGPT · miniGHOSTSaverage rating: 3.20 -> 4.15Improved
Use GPT-4 instead of ChatGPTChatGPT · Symbolic-Integrationaverage rating: 2.51 -> 3.50Improved
Use GPT-4 instead of ChatGPTChatGPT · MATHperfect-rating share: 29% -> 70%Improved
Apply prompt engineering to complex questionsChatGPT · GHOSTS complex questionsratingImproved

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