mlsanityData to 7 Oct 2026

Recipes for Steering and Scaling LLMs via Sampling

arXiv 2608.26120 ↗First posted 19 Jun 2026cs.CL cs.LGcited by 1

Not on the map. Why?

Abstract (arXiv)

Large Language Models (LLMs) are probabilistic models, typically defined by an autoregressive factorization. While recent work has begun to study richer target distributions beyond the base model, the sampling strategies remain highly inefficient. In this paper, we present a flexible and theoretically grounded framework for steering and scaling autoregressive LLMs with sampling. Within this framework, we describe two algorithms -- one based on Sequential Monte Carlo (SMC) and one based on Replica Exchange (RE) -- that steer generation toward powering, product or tilting of the base model distribution. We illustrate this framework through scaling the generation quality of LLMs without external supervision or reward models. Experimental results demonstrate our methods scale more favorably than Best-of-N and standard MCMC baselines. Overall, this paper offers a systematic recipe for probabilistic inference with LLMs via sampling.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 36 of its 39 references that are in the map. First posted on arXiv 19 Jun 2026.

Builds on

36 of 39 references.

References · 36

  1. 2023GPT-4 Technical Reportcited by 18,132
  2. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
  3. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,185
  4. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 7,916
  5. 2022Classifier-Free Diffusion Guidancecited by 5,717
  6. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,601
  7. 2022Large Language Models are Zero-Shot Reasonerscited by 4,323
  8. 2023Let's Verify Step by Stepcited by 3,598
  9. 2023Self-Refine: Iterative Refinement with Self-Feedbackcited by 3,317
  10. 2023GPQA: A Graduate-Level Google-Proof Q&A Benchmarkcited by 2,748
  11. 2022Fast Inference from Transformers via Speculative Decodingcited by 1,615
  12. 2022Diffusion Posterior Sampling for General Noisy Inverse Problemscited by 1,375
24 more
  1. 2023Sparse Autoencoders Find Highly Interpretable Features in Language Modelscited by 1,359
  2. 2025Understanding R1-Zero-Like Training: A Critical Perspectivecited by 1,276
  3. 2023Inference-Time Intervention: Eliciting Truthful Answers from a Language Modelcited by 1,259
  4. 2023Reasoning with Language Model is Planning with World Modelcited by 846
  5. 2023Large Language Models as Optimizerscited by 830
  6. 2025AlphaEvolve: A coding agent for scientific and algorithmic discoverycited by 763
  7. 2023Practical and Asymptotically Exact Conditional Sampling in Diffusion Modelscited by 218
  8. 2023Reduce, Reuse, Recycle: Compositional Generation with Energy-Based Diffusion Models and MCMCcited by 216
  9. 2025A General Framework for Inference-time Scaling and Steering of Diffusion Modelscited by 201
  10. 2025Deep Think with Confidencecited by 199
  11. 2023Controlled Decoding from Language Modelscited by 133
  12. 2024Self-Improvement in Language Models: The Sharpening Mechanismcited by 102
  13. 2025The Invisible Leash: Why RLVR May or May Not Escape Its Origincited by 92
  14. 2024Probabilistic Inference in Language Models via Twisted Sequential Monte Carlocited by 85
  15. 2025Feynman-Kac Correctors in Diffusion: Annealing, Guidance, and Product of Expertscited by 85
  16. 2025Reasoning with Sampling: Your Base Model is Smarter Than You Thinkcited by 84
  17. 2024Aligning Large Language Models with Representation Editing: A Control Perspectivecited by 62
  18. 2024Efficient Evolutionary Search Over Chemical Space with Large Language Modelscited by 59
  19. 2024Diffusion Models as Constrained Samplers for Optimization with Unknown Constraintscited by 33
  20. 2025The Coverage Principle: How Pre-Training Enables Post-Trainingcited by 25
  21. 2025RNE: plug-and-play diffusion inference-time control and energy-based trainingcited by 16
  22. 2025Accelerating Scientific Discovery with Autonomous Goal-evolving Agentscited by 16
  23. 2025Accelerated Parallel Tempering via Neural Transportscited by 13
  24. 2025CREPE: Controlling Diffusion with Replica Exchangecited by 9

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Explore Broadly, Reason Sharply: Push Small Models toward the Frontier via Samplingcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025When Life Gives You Samples: The Benefits of Scaling up Inference Compute for Multilingual LLMs
  2. 2024Beyond Autoregression: Fast LLMs via Self-Distillation Through Time
  3. 2026The Illusion of Stochasticity in LLMs
  4. 2026Ladders in Chaos: When, How, (and Perhaps Why) Does Test-Time Scaling Improve LLM Machine Translation
  5. 2026The Probabilistic Structure of Large Language Models
  6. 2024A Probabilistic Perspective on Unlearning and Alignment for Large Language Models
  7. 2026Scalable Power Sampling: Unlocking Efficient, Training-Free Reasoning for LLMs via Distribution Sharpening
  8. 2025Large Language Diffusion Models