arXiv Machine Learning
Unrolled Flow Models for Reasoning
Read the original on arXiv Machine Learning →The Flow has not summarised this story yet — read it at arXiv Machine Learning.
Related stories
Hugging Face Trending Papers
Oct 7
Unrolled Flow Models for Reasoning
Flow matching enables language generation in few steps, but whether additional integration steps improve reasoning remains unclear. We prove that a flow parameterized by a two-layer Transformer can so...
arXiv Machine Learning
Sep 14
Sampling via Decision-Flow: Training-Free Extraction of Improved Latent Reasoning Paths in Large Language Models
The paper introduces Decision-Flow Sampling (DF‑Sample), a training‑free, data‑free inference framework that builds a hierarchical reasoning tree, evaluates entire trajectories, and back‑propagates utilities to guide branching decisions. Unlike local step‑wise sampling, DF‑Sample explicitly assesses global paths, enabling it to recover high‑quality, low‑probability reasoning chains that standard decoding misses. On the GPQA benchmark, DF‑Sample attains 45.6% accuracy, outperforming power sampling (38.9%) and GRPO (39.9%) and consistently surpassing baselines across multiple models and benchmarks, demonstrating significant latent reasoning potential in pretrained LLMs.
Hugging Face Trending Papers
Oct 7
Efficient Reasoning with Flow Language Models
Flow Language Models (FLMs) have emerged as a continuous-state alternative to discrete diffusion language models, yet the role of their continuous representations in reasoning remains unclear. We inve...
arXiv AI
Jul 29
Penelope: Localized Latent Recurrence for Efficient Structured Reasoning
arXiv:2607. 25915v1 Announce Type: new Abstract: Complex structured reasoning tasks often require additional computation, yet current language models obtain it mainly by increasing parameter scale or by serializing intermediate steps as chain-of-thought (CoT) tokens.
arXiv AI
Jun 15
Fractured Chain-of-Thought Reasoning
arXiv:2505. 12992v4 Announce Type: replace-cross Abstract: Inference-time scaling techniques have significantly bolstered the reasoning capabilities of large language models (LLMs) by harnessing additional computational effort at inference without retraining.
arXiv AI
Jul 2
DiscoLoop: Looping Discrete Embeddings and Continuous Hidden States for Multi-hop Reasoning
arXiv:2607. 00341v1 Announce Type: cross Abstract: Large language models achieve strong performance on many reasoning tasks when allowed to externalize intermediate steps as Chain-of-Thought (CoT).