Decision-Flow Sampling (DF-Sample), a training-free, data-free inference-time framework that constructs a hierarchical reasoning tree, scores terminal nodes for quality, and back-propagates utilities to inform each intermediate branching decision, indicating substantial latent reasoning potential in pretrained base models.
Abstract
A central question in LLM reasoning is whether reinforcement learning (RL) instills genuinely new capabilities or merely reshapes how existing knowledge is expressed during inference. Building on the distribution-sharpening hypothesis, which holds that RL reallocates probability mass toward high-reward trajectories already latent in base models, we ask: can we unlock those latent paths without costly RL fine-tuning? We present Decision-Flow Sampling (DF-Sample), a training-free, data-free inference-time framework that constructs a hierarchical reasoning tree, scores terminal nodes for quality, and back-propagates utilities to inform each intermediate branching decision. Unlike conventional sampling strategies that make purely local step-wise choices, DF-Sample performs explicit global trajectory evaluation before committing to a path, recovering high-quality but low-probability reasoning chains that standard decoding overlooks. On GPQA, DF-Sample achieves 45.6% accuracy, surpassing power sampling (38.9%) and GRPO (39.9%), showing that a training-free method can outperform a trained one. Across three models and four benchmarks, DF-Sample consistently outperforms baselines, indicating substantial latent reasoning potential in pretrained base models.
Results support a qualified internalized-search reading: under the recipe the authors test, much of the measured RL gain corresponds to a change in sampling efficiency toward operating points the base model can already reach under search.
Wen-He Sun, Cun-Xiang Wang, Zi-Jun Yao et al.· 0 citations
SPS is a training-free latent steering framework that constructs a state-conditioned Direction Bank containing multiple progress-guided steering vectors for different prefix-state regions and applies it at high-uncertainty transitions to guide the next reasoning step toward meaningful progress.
Hengyuan Zhang, Chenming Shang, Zun-Hai Su et al.· 0 citations
This work presents Token-Disentangled Latent Test-Time Scaling, an inference-time framework that makes latent refinement token-role-aware and outperforms strong output-space test-time scaling baselines under matched decoded-candidate budgets.
Hao-Xuan Ma, Yi-Hao Liu, Yu-Tao Sun et al.· Conference on Empirical Meth...· 0 citations
SOLID is proposed, a novel framework for self-improving OR language models without verified answers or external evaluators that improves solution accuracy for both general-purpose and OR-tuned models over outcome-only group-relative training.
Rui-Chen Zhu, Ming-Long Cao, Chen-Yu Zhou et al.· 0 citations
EvoResearcher is presented, a training-free, inference-time protocol that adds cost-bounded self-reflection to a single frozen LLM backbone and establishes cross-domain behavior on GSM8K and MATH on the same frozen backbone.
A*-Thought-V2 is presented, a geometric dynamics of LLM guided framework that models CoT as a hidden-state trajectory and replaces hard deletion with an explicit-implicit interleaved latent architecture that reflects broader soft targets that encourage richer step-level feature learning.
Xiao-An Xu, Si-Yuan Liu, Shuo Wang et al.· 1 citation
Related blog posts
Microsoft Research Blog· microsoft.comNov 16, 2026
AI coding agents are increasingly help developers fix bugs and build features largely on their own, but every interaction with their models consumes tokens. In each session, an agent sends instructions, tool descriptions, and session history to its model. Most efforts to reduce this cost focus on individual tasks. Yet developers return to the same […] The post TokenEconomizer: Learning from Experience to Make AI Coding Agents More Efficient appeared first on Microsoft Research.
MIT News · Artificial Intelligence· news.mit.eduNov 13, 2026