arXiv Computation and Language September 14, 2026 By Minoo Ahmadi, Seyedarmin Azizi, Erfan Baghaei Potraghloo, Mehdi Kamal, Massoud Pedram

Chopthin-Consensus Power Sampling: A Diversity-Preserving Approach to LLM Decoding

Read the original on arXiv Computation and Language →

Chopthin-Consensus Power Sampling (CCPS) is a new inference-time decoding method for large language models that uses the Chopthin resampler to preserve diversity among particle trajectories. By enforcing an upper bound on weight ratios instead of equal-weight resampling, CCPS maintains a richer set of distinct reasoning paths and guarantees a lower bound on effective sample size. Coupled with a semantic-majority selection mechanism, CCPS achieves higher oracle coverage and matches or surpasses baseline accuracy on multiple reasoning benchmarks.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computation and Language.

arXiv Machine Learning

Oct 5

Power-SMC: Low-Latency Sequence-Level Power Sampling for Training-Free LLM Reasoning

Power‑SMC is a training‑free sampling method that targets the sequence‑level power distribution used to sharpen large language model outputs for reasoning tasks. It maintains multiple candidate sequences, assigns importance weights based on how well each matches the power distribution, and periodically prunes low‑scoring candidates to keep inference latency close to standard decoding. Experiments on MATH500, GSM8K, GPQA, and HumanEval show that Power‑SMC matches or surpasses Metropolis–Hastings sampling in accuracy, preserves output diversity, and can accelerate inference speed by up to 17.6×.

By Seyedarmin Azizi, Erfan Baghaei Potraghloo, Minoo Ahmadi, Souvik Kundu, Massoud Pedram

llms reinforcement-learning efficiency benchmarks

arXiv Machine Learning

Aug 17

More Correct Mass, Worse Answers: Why Power Sampling Can Fail and How to Fix It

arXiv:2608. 14420v1 Announce Type: new Abstract: Power Sampling sharpens a language model's distribution over complete generation trajectories, offering a verifier-free way to improve reasoning at inference time.

By Haohui Yang, Jiaxing Sun, Xiujun Ma

llms benchmarks

arXiv Machine Learning

Sep 30

Explore Broadly, Reason Sharply: Push Small Models toward the Frontier via Sampling

arXiv:2609.38104v1 Announce Type: new Abstract: Power-sharpened sampling is an inference-time alternative to reinforcement-learning (RL) post-training for enhancing reasoning in large language models...

By Panagiotis Theodoropoulos, Nan Jiang, Xintong Duan, Ali Hasan, Yuriy Nevmyvaka, Evangelos A. Theodorou, Wei Deng

llms reinforcement-learning safety

arXiv Machine Learning

Oct 15

Decoupled Marginal Sharpening for Training-Free Inference-Time Scaling

arXiv:2610.20483v1 Announce Type: new Abstract: We study how to improve LLM reasoning at inference time without updating model weights or using external reward models. Recent reward-free power-sampli...

By Weixin Wang, Wei Deng, Anderson Schneider, Yuriy Nevmyvaka, Pan Xu, Andrew Bennett

llms benchmarks

arXiv AI

Sep 30

Reasoning with Sampling: Cutting at Decision Points

arXiv:2605.30327v2 Announce Type: replace-cross Abstract: Frontier reasoning models are produced by post-training base language models with reinforcement learning. Recent work has challenged this by...

By Felix Zhou, Anay Mehrotra, Quanquan C. Liu

llms reinforcement-learning benchmarks

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.

By Baohao Liao, Hanze Dong, Yuhui Xu, Doyen Sahoo, Christof Monz, Junnan Li, Caiming Xiong

llms benchmarks

One email a morning, machine-written

Related stories

Power-SMC: Low-Latency Sequence-Level Power Sampling for Training-Free LLM Reasoning

More Correct Mass, Worse Answers: Why Power Sampling Can Fail and How to Fix It

Explore Broadly, Reason Sharply: Push Small Models toward the Frontier via Sampling

Decoupled Marginal Sharpening for Training-Free Inference-Time Scaling

Reasoning with Sampling: Cutting at Decision Points

Fractured Chain-of-Thought Reasoning