Inference-Time Guidance

Ask about this topic

Overview

Inference-time guidance studies how to steer a model while it generates, often adapting a frozen diffusion or flow model, language model, or robot policy to a new reward, constraint, or user preference. Common approaches include classifier-free or gradient-based guidance, reward and verifier scoring with resampling, and direct interventions on logits, latent states, or action probabilities. For reasoning, methods use self-consistency, process verification, and adaptive search; for generation and control, they guide denoising trajectories or select among candidate plans, with evaluations spanning reasoning benchmarks, image and molecular generation, and simulated or real robots. A central challenge is making guidance reliable and calibrated without excessive compute or damage to sample quality: noisy rewards, conflicting objectives, and task-specific gains can make stronger intervention counterproductive.

Latest papers

CardsList

104 papers29 papers in the last four weeks, up 383% on the four weeks before. 0.2% of all new papers.

  1. Prompt-Guided Image Editing with Masked Logit Nudging in Visual Autoregressive Models

    Apr 16, 2026Amir El-Ghoussani, Marc Hölle, Gustavo Carneiro +1Text-Guided Image EditingVisual Autoregressive Models

    Friedrich-Alexander-Universität Erlangen-Nürnberg, Germany · University of Surrey, United Kingdom

  2. Dissecting Failure Dynamics in Large Language Model Reasoning

    Apr 16, 2026Wei Zhu, Jian Zhang, Lixing Yu +2Inference-Time GuidanceLLM Reasoning

    School of Information Science and Engineering, Yunnan University, Kunming, China · Yunnan Key Laboratory of Intelligent Systems and Computing, Kunming, China

  3. Co-Evolution of Policy and Internal Reward for Language Agents

    Apr 3, 2026Xinyu Wang, Hanwei Wu, Jingwei Song +8RL for LLM AgentsIntrinsic Reward Methods

    McGill University · McMaster University · The University of Hong Kong +6

  4. Enhancing reasoning accuracy in large language models during inference time

    Mar 22, 2026Vinay Sharma, Manish JainLLM InferenceLLM Reasoning

  5. C2^2FG: Control Classifier-Free Guidance via Score Discrepancy Analysis

    Mar 9, 2026Jiayang Gao, Tianyi Zheng, Jiayang Zou +9Classifier-Free GuidanceInference-Time Guidance

    Shanghai Jiao Tong University, Shanghai, China · vivo BlueImage Lab, vivo Mobile Communication Co., Ltd., China

  6. Reject, Resample, Repeat: Understanding Parallel Reasoning in Language Model Inference

    Mar 9, 2026Noah Golowich, Fan Chen, Dhruv Rohatgi +4Parallel Reasoning in Language ModelsLLM Inference

    Microsoft Research · MIT · NYU

  7. PRISM: Pushing the Frontier of Deep Think via Process Reward Model-Guided Inference

    Mar 3, 2026Rituraj Sharma, Weiyuan Chen, Noah Provenzano +1LLM ReasoningInference-Time Guidance

    Virginia Tech

  8. Strategy Executability in Mathematical Reasoning: Leveraging Human-Model Differences for Effective Guidance

    Feb 26, 2026Weida Liang, Yiyou Sun, Shuyuan Nan +3LLM Mathematical ReasoningInference-Time Guidance

    National University of Singapore · University of California, Berkeley

  9. Calibrated Test-Time Guidance for Bayesian Inference

    Feb 25, 2026Daniel Geyfman, Felix Draxler, Jan Groeneveld +3Bayesian Inverse ProblemsBayesian Inference

    Department of Computer Science, University of California, Irvine, CA, USA · Seoul National University, Seoul, South Korea · Chan-Zuckerberg Initiative, Redwood City, CA, USA

  10. Momentum Guidance: Plug-and-Play Guidance for Flow Models

    Feb 23, 2026Runlong Liao, Jian Yu, Baiyu Su +3Momentum MethodsDiffusion Guidance

    The University of Texas at Austin, Austin, TX 78712, USA

  11. Steering diffusion models with quadratic rewards: a fine-grained analysis

    Feb 18, 2026Ankur Moitra, Andrej Risteski, Dhruv RohatgiDiffusion SamplingDiffusion Models

  12. What If We Allocate Test-Time Compute Adaptively?

    Feb 1, 2026Ahsan Bilal, Ahmed Mohsin, Muhammad Umer +4Inference-Time SearchTest-Time Scaling

    University of Oklahoma, OK, USA · Stanford University, CA, USA · Universitat Pompeu Fabra, Barcelona, Spain +2

  13. TexTailor: Inference-Time Textual Guidance Tailoring for Multimodal Diffusion Transformers

    Jan 5, 2026Binglei Li, Mengping Yang, Zhiyu Tan +2Diffusion TransformerCompositional T2I Generation

    Fudan University, Shanghai 200433, China · Shanghai Academy of AI for Science, Shanghai 200030, China · Shanghai Innovation Institute, Shanghai 200030, China

  14. Let Language Constrain Geometry: Vision-Language Models as Semantic and Spatial Critics for 3D Generation

    Nov 18, 2025Weimin Bai, Yubo Li, Weijian Luo +43D Spatial ReasoningText-to-3D Generation

    Peking University, Beijing · hi-lab, Xiaohongshu Inc. · MMLab, CUHK +1