Data

Tools

Indexes

Applications

Experiments

Awesome

An open API service indexing awesome lists of open source software.

Awesome-Latent-CoT

This repository contains a regularly updated paper list for LLMs-reasoning-in-latent-space.
https://github.com/EIT-NLP/Awesome-Latent-CoT

Last synced: 7 days ago
JSON representation

  • Keywords Convention

  • Papers

    • Token-wise Strategies

      • [pdf - orange) ![](https://img.shields.io/badge/pause_tokens-blue) ![](https://img.shields.io/badge/Pretraining-lightgray)
      • [pdf - orange) ![](https://img.shields.io/badge/Quiet--STaR-blue) ![](https://img.shields.io/badge/Pretraining-lightgray)
      • [pdf - orange) ![](https://img.shields.io/badge/thinking_tokens-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/CPT-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/LPC-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/PHD--Transformer-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/SoftCoT++-blue)
      • [pdf - ai-lab/Soft-Thinking)], 2025.05. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/Soft_Thinking-blue) ![](https://img.shields.io/badge/Training_free-lightgray)
      • [pdf - latent-reasoning.github.io/)], 2025.05. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/CoLaR-blue) ![](https://img.shields.io/badge/RL-lightgray)
      • [pdf - orange) ![](https://img.shields.io/badge/HRPO-blue) ![](https://img.shields.io/badge/RL-lightgray)
      • [pdf - orange) ![](https://img.shields.io/badge/LATENTSEEK-blue) ![](https://img.shields.io/badge/RL-lightgray)
      • [pdf - orange)
      • [pdf - orange) ![](https://img.shields.io/badge/System--1.5_Reasoning-blue)
      • [pdf - Reasoning-5C1D/README.md)], 2025.06. ![](https://img.shields.io/badge/Arxiv-orange)
      • [pdf - orange) ![](https://img.shields.io/badge/CoCoMix-blue) ![](https://img.shields.io/badge/Pretraining-lightgray)
      • [pdf - orange) ![](https://img.shields.io/badge/BoLT-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/planning_tokens-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/filler_tokens-blue)
      • [pdf - Memory-and-Reasoning)], 2024.11. ![](https://img.shields.io/badge/Arxiv-orange)
      • [pdf - orange) ![](https://img.shields.io/badge/VQ--VAE-lightgray)
      • [pdf - orange) ![](https://img.shields.io/badge/HCoT-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/Coconut-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/CCoT-blue) ![](https://img.shields.io/badge/Compressed_representations-lightgray)
      • [pdf - orange) ![](https://img.shields.io/badge/SoftCoT-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/LightThinker-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/CODI-blue)

    • Analysis and Interpretability

      • [pdf - orange)
      • [pdf - orange)
      • [pdf - orange)
      • [pdf - orange)
      • [pdf - -COLING2024-orange)
      • [pdf - orange)
      • [pdf - orange)
      • [pdf - orange)
      • [pdf - orange) ![](https://img.shields.io/badge/CREME-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/CoE-blue)
      • [pdf - Chain-of-Thought)], 2025.05. ![](https://img.shields.io/badge/Arxiv-orange)
      • [pdf - orange)
      • [pdf - orange)
      • [pdf - orange) ![](https://img.shields.io/badge/CoT2-blue)
      • [pdf - orange)
      • [pdf - orange)
      • [pdf - orange)
      • [pdf - orange)
      • [pdf - orange)
      • [pdf - orange)
      • [pdf - orange)

    • Applications and Future Directions

      • [pdf - Group/SEAL)], 2025.04. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/Training_free-lightgray)
      • [pdf - orange)![](https://img.shields.io/badge/XS--CoTs-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/DCoLT-blue)
      • [pdf - cs.github.io/SSR/)], 2025.05. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/SSR-blue) ![](https://img.shields.io/badge/Training_free-lightgray)
      • [pdf - orange) ![](https://img.shields.io/badge/benchmark-lightgray)
      • [pdf - orange) ![](https://img.shields.io/badge/DoT-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/LatentLM-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/Heima-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/DEBATER-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/ReaRec-blue)

    • Internal Mechanisms

      • [pdf - rg/recurrent-pretraining)], [[model](https://huggingface.co/tomg-group-umd/huginn-0125)], 2025.02. ![](https://img.shields.io/badge/Arxiv-orange) ![](https://img.shields.io/badge/Huggin-blue) ![](https://img.shields.io/badge/Depth_Recurrence-lightgray)
      • [pdf - orange) ![](https://img.shields.io/badge/Pondering%20LM-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/CoTFormer-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/ITT-blue)
      • [pdf - orange)![](https://img.shields.io/badge/ICoT-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/Stepwise_Internalization-lightgray)
      • [pdf - orange) ![](https://img.shields.io/badge/System_2_Distillation-lightgray)
      • [pdf - orange) ![](https://img.shields.io/badge/RELAY-blue)
      • [pdf - orange) ![](https://img.shields.io/badge/Looped_Transformers-blue)

    • Pre-training and Continue Pre-training

      • [pdf - orange) ![](https://img.shields.io/badge/LTM-blue)

    • Supervised-Finetuning

      • [pdf - orange) ![](https://img.shields.io/badge/TwT-blue)

    • Applications

      • [pdf - COT)], 2023.12. ![](https://img.shields.io/badge/AAAI2024-orange)
      • [pdf - orange)
      • [pdf - bonnet/lpn)], 2024.11. ![](https://img.shields.io/badge/Arxiv-orange)
      • [pdf - orange)
      • [pdf - orange)
      • [pdf - orange)

  • Updates

  • Resources

    • Applications and Future Directions

      • Awesome-Efficient-Reasoning - Efficient-Reasoning-Models](https://github.com/fscdc/Awesome-Efficient-Reasoning-Models) [[Paper](https://arxiv.org/pdf/2504.10903)].
Categories
Papers 74 Keywords Convention 3 Resources 1 Updates 1
Sub Categories
Token-wise Strategies 26 Analysis and Interpretability 21 Applications and Future Directions 11 Internal Mechanisms 9 Other related work or pending to be specified 5 Pre-training and Continue Pre-training 1 Applications 1 Supervised-Finetuning 1
Keywords
efficient-reasoning 1 chain-of-thought 1