
ChatGPT, a large language model, can outperform specialized GNN models on the CLRS algorithmic reasoning benchmark by directly executing classical algorithms in Python.


coremsg

evaluating-chatgpt-s-performance-on-the-clrs-algorithmic-reasoning-benchmark


Evaluating ChatGPT's Performance on the CLRS Algorithmic Reasoning Benchmark


title_rewrite


Decomposing the reasoning process of language models into two steps - THINK to discover task-level logic expressed in pseudocode, and EXECUTE to tailor the pseudocode and simulate its execution - can significantly improve their performance on various algorithmic reasoning tasks.


language-models-as-compilers-simulating-pseudocode-execution-improves-algorithmic-reasoning-in-language-models


Language Models as Compilers: Simulating Pseudocode Execution Improves Algorithmic Reasoning in Language Models
