Thinking LLMs: General Instruction Following with Thought Generation

Tianhao Wu, Janice Lan, Weizhe Yuan, Jiantao Jiao, Jason E. Weston +1 more
2/14/2026

No official code found. Best implementation: sanowl/THINKING-LLMS (7 stars, last commit 2024-12-22).

Semantic Scholar

Code Implementations(1)

sanowl/THINKING-LLMSCommunity65%

this is based on the paper THINKING LLMS: GENERAL INSTRUCTION FOLLOWING WITH THOUGHT GENERATION I might add new stuff that is not related to the paper

72Python, MakefileOct 17, 20241 years ago

Similar Papers

ChronoEpilogi: Scalable Time Series Selection with Multiple Solutions

100% match1 repo2026

FlowDiffuser: Advancing Optical Flow Estimation with Diffusion Models

100% match1 repo1042026

MARGE: Improving Math Reasoning with Guided Exploration

100% match1 repo2026

SketchKnitter: Vectorized Sketch Generation with Diffusion Models

100% match1 repo682026

XES3G5M: A Knowledge Tracing Benchmark Dataset with Auxiliary Information

100% match2026

LD2: Scalable Heterophilous Graph Neural Network with Decoupled Embeddings

100% match1 repo62026

Cite this paper

@article{wu2026thinking,
  title  = {Thinking LLMs: General Instruction Following with Thought Generation},
  author = {Tianhao Wu and Janice Lan and Weizhe Yuan and Jiantao Jiao and Jason E. Weston and Sainbayar Sukhbaatar},
  year   = {2026},
  url    = {https://api.semanticscholar.org/CorpusID:3c203ace8b469dea56b2c759c699e685eb98b626},
  journal = {ICML 2025 2025}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/3c203ace8b469dea56b2c759c699e685eb98b626.svg)](https://codeofpaper.com/paper/3c203ace8b469dea56b2c759c699e685eb98b626)

Discussion