Efficient LLM Reasoning via Variational Posterior Guidance with Efficiency Awareness

• URL: http://arxiv.org/abs/2605.11019v1
• Date: Sun, 10 May 2026 15:18:24 GMT
• Title: Efficient LLM Reasoning via Variational Posterior Guidance with Efficiency Awareness
• Abstract summary: Existing reinforcement learning methods compress reasoning chains by designing elaborate reward functions.<n>We introduce an efficiency-aware evidence lower bound as the theoretical foundation.<n> Experiments on DeepSeek-R1-Distill-Qwen-1.5B and 7B scales demonstrate that VPG-EA improves the comprehensive efficiency metric epsilon cubed by 8.73% and 12.37% over the strongest baselines on each model size.
• Score: 2.83639496314338
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Although large language models rely on chain-of-thought for complex reasoning, the overthinking phenomenon severely degrades inference efficiency. Existing reinforcement learning methods compress reasoning chains by designing elaborate reward functions, which renders high-quality samples extremely sparse in the exploration space and creates a sampling bottleneck for the prior policy. Inspired by cognitive science, we theoretically prove that a posterior distribution guided by reference answers achieves higher expected utility than the prior distribution, thus capable of breaking through the sampling bottleneck of high-quality samples. However, the posterior distribution is unavailable during inference. To this end, we formalize efficient reasoning as a variational inference problem and introduce an efficiency-aware evidence lower bound as the theoretical foundation. Based on this, we propose the VPG-EA framework. It adopts a parameter-shared dual-stream architecture to instantiate both the posterior distribution and the prior policy; after filtering out pseudo-efficient paths via cross-view evaluation, it unidirectionally transfers the posterior's efficient patterns to the prior policy through variational distillation. Experiments on DeepSeek-R1-Distill-Qwen-1.5B and 7B scales demonstrate that VPG-EA improves the comprehensive efficiency metric epsilon cubed by 8.73% and 12.37% over the strongest baselines on each model size, respectively.

Related papers

• Too Correct to Learn: Reinforcement Learning on Saturated Reasoning Data [55.84428098924793]
  We propose Constrained Uniform Top-K Sampling (CUTS), a parameter-free decoding strategy enforcing structure-preserving exploration.<n>We integrate this into Mixed-CUTS, a training framework synergizing exploitative and exploratory rollouts to amplify intra-group advantage variance.<n> Notably, Mixed-CUTS improves Pass@1 accuracy on the challenging AIME25 benchmark by up to 15.1% over standard GRPO.
  arXiv  Detail & Related papers  (2026-04-20T16:43:28Z)
• Not All Tokens See Equally: Perception-Grounded Policy Optimization for Large Vision-Language Models [38.47027398567909]
  Perception-Grounded Policy Optimization (PGPO) is a novel fine-grained credit assignment framework that dynamically reshapes advantages at the token level.<n>We show that PGPO actively amplifies learning signals for visually-dependent tokens while suppressing gradient noise from linguistic priors.<n>Both theoretical and empirical analyses confirm that PGPO effectively reduces gradient variance, prevents training collapse, and acts as a potent regularizer for robust, perception-grounded multimodal reasoning.
  arXiv  Detail & Related papers  (2026-04-02T09:53:20Z)
• ERPO: Token-Level Entropy-Regulated Policy Optimization for Large Reasoning Models [3.463914032107119]
  We show that Group Relative Policy Optimization (GRPO) typically assigns a uniform, sequence-level advantage to all tokens.<n>We propose Entropy-Regulated Policy Optimization (ERPO), which transitions the optimization focus from coarse sequences to fine-grained token dynamics.
  arXiv  Detail & Related papers  (2026-03-30T09:20:25Z)
• P^2O: Joint Policy and Prompt Optimization [74.45293488495592]
  Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as a powerful paradigm for enhancing the reasoning capabilities of Large Language Models (LLMs)<n>We propose P2O, a novel framework that synergizes Prompt Optimization with Policy Optimization.
  arXiv  Detail & Related papers  (2026-03-23T12:08:47Z)
• Back to Basics: Revisiting Exploration in Reinforcement Learning for LLM Reasoning via Generative Probabilities [10.235183326885794]
  Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as an indispensable paradigm for enhancing reasoning in Large Language Models (LLMs)<n>We analyze this issue from the perspective of sampling probability dynamics, identifying that the standard objective disproportionately reinforces the highest-likelihood paths.<n>We propose a novel Advantage Re-weighting Mechanism (ARM) designed to equilibrate the confidence levels across all correct responses.
  arXiv  Detail & Related papers  (2026-02-05T04:06:55Z)
• Amortized Inference of Multi-Modal Posteriors using Likelihood-Weighted Normalizing Flows [0.0]
  We present a novel technique for amortized posterior estimation using Normalizing Flows trained with likelihood-weighted importance sampling.<n>We implement the method on multi-modal benchmark tasks in 2D and 3D to check for the efficacy.
  arXiv  Detail & Related papers  (2025-12-04T16:22:53Z)
• Efficient Thought Space Exploration through Strategic Intervention [54.35208611253168]
  We propose a novel Hint-Practice Reasoning (HPR) framework that operationalizes this insight through two synergistic components.<n>The framework's core innovation lies in Distributional Inconsistency Reduction (DIR), which dynamically identifies intervention points.<n> Experiments across arithmetic and commonsense reasoning benchmarks demonstrate HPR's state-of-the-art efficiency-accuracy tradeoffs.
  arXiv  Detail & Related papers  (2025-11-13T07:26:01Z)
• DeepPrune: Parallel Scaling without Inter-trace Redundancy [53.62015294143274]
  Over 80% of parallel reasoning traces yield identical final answers, representing substantial wasted computation.<n>We propose DeepPrune, a novel framework that enables efficient parallel scaling through dynamic pruning.<n>Our work establishes a new standard for efficient parallel reasoning, making high-performance reasoning more efficient.
  arXiv  Detail & Related papers  (2025-10-09T17:24:54Z)
• SPREAD: Sampling-based Pareto front Refinement via Efficient Adaptive Diffusion [0.8594140167290097]
  SPREAD is a generative framework based on Denoising Diffusion Probabilistic Models (DDPMs)<n>It learns a conditional diffusion process over points sampled from the decision space.<n>It refines candidates via a sampling scheme that uses an adaptive multiple gradient descent-inspired update for fast convergence.
  arXiv  Detail & Related papers  (2025-09-25T12:09:37Z)
• EconProver: Towards More Economical Test-Time Scaling for Automated Theorem Proving [64.15371139980802]
  Large Language Models (LLMs) have recently advanced the field of Automated Theorem Proving (ATP)<n>We show that different test-time scaling strategies for ATP models introduce significant computational overhead for inference.<n>We propose two complementary methods that can be integrated into a unified EconRL pipeline for amplified benefits.
  arXiv  Detail & Related papers  (2025-09-16T03:00:13Z)
• VAPO: Efficient and Reliable Reinforcement Learning for Advanced Reasoning Tasks [49.0793012627959]
  We present VAPO, a novel framework tailored for reasoning models within the value-based paradigm.<n>VAPO attains a state-of-the-art score of $mathbf60.4$.<n>In direct comparison under identical experimental settings, VAPO outperforms the previously reported results of DeepSeek-R1-Zero-Qwen-32B and DAPO by more than 10 points.
  arXiv  Detail & Related papers  (2025-04-07T14:21:11Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.