Researchers introduced Self-Explainable Latent Reasoning (SELR), a unified framework that trains a single model to perform latent reasoning while decoding its own latent representations into human-rea…
A study by researchers testing the Coconut and CODI models found that these latent reasoning models barely use their hidden reasoning steps for logical tasks like PrOntoQA and ProsQA, with performance…
An independent developer has released DeepSeek-V4-Flash-0731-Latent-Reasoning, a 284B-parameter MoE model with a 35.7M-parameter reasoning head that performs latent reasoning in a 1024-dimensional spa…
An independent researcher seeking an arXiv endorsement for cs.LG reports that a causal study of latent reasoning in a chess-playing LLM found that reinforcement learning (GRPO) improved legal-move rat…
A developer known as flirp spent 21 GPU-hours training Qwen2.5-1.5B-Instruct to reason in latent space without decoding tokens, but the model's hidden-state reasoning collapsed into content-free attra…
A developer tested an aggressive version of latent-space reasoning on a 1.5B-parameter language model, where the model pauses during generation to run parallel hidden-state rollouts without decoding t…
Researchers at the University of Oxford found that observable patterns in latent reasoning models, such as BFS-like frontiers and decodable arithmetic, also appear in control models lacking the propos…