Chain-of-Thought Provably Enables Learning the (Otherwise) Unlearnable

Chenxiao Yang, Zhiyuan Li, David Wipf

2/12/2026

No official code found. Best implementation: chr26195/CoT-ICL (6 stars, last commit 2025-03-06).

Semantic Scholar

Code Implementations(1)

chr26195/CoT-ICLCommunity66%

This is the official code for "Chain-of-Thought Provably Enables Learning the (Otherwise) Unlearnable" in ICLR 2025

60Shell, PythonMar 1, 20251 years ago

Similar Papers

Learning High-Degree Parities: The Crucial Role of the Initialization

94% match1 repo2026

On the Humanity of Conversational AI: Evaluating the Psychological Portrayal of LLMs

92% match1 repo1412026

The Tilted Variational Autoencoder: Improving Out-of-Distribution Detection

Chain-of-Experts: When LLMs Meet Complex Operations Research Problems

91% match1 repo1222026

TAG: Text Prompt Augmentation for Zero-Shot Out-of-Distribution Detection

87% match1 repo52026

UniCoTT: A Unified Framework for Structural Chain-of-Thought Distillation

87% match1 repo92026

Cite this paper

@article{yang2026chainofthought,
  title  = {Chain-of-Thought Provably Enables Learning the (Otherwise) Unlearnable},
  author = {Chenxiao Yang and Zhiyuan Li and David Wipf},
  year   = {2026},
  url    = {https://api.semanticscholar.org/CorpusID:4978a5c2c5a61c0d55b7d4f2974297917f208a9f},
  journal = {ICLR 2025 2025}
}

Add a badge to your README

Shows whether official code is linked for this paper.

[![Code of Paper](https://codeofpaper.com/badge/4978a5c2c5a61c0d55b7d4f2974297917f208a9f.svg)](https://codeofpaper.com/paper/4978a5c2c5a61c0d55b7d4f2974297917f208a9f)

Discussion