Lune

ICLR2025Top-tier venue

Chain-of-Focus Prompting: Leveraging Sequential Visual Cues to Prompt Large Autoregressive Vision Models

Jiyang Zheng, Jialiang Shen, Yu Yao, Min Wang, Yang Yang, Dadong Wang, Tongliang Liu

Abstract

In-context learning (ICL) has revolutionized natural language processing by enabling models to adapt to diverse tasks with only a few illustrative examples. However, the exploration of ICL within the field of computer vision remains limited. Inspired by Chain-of-Thought (CoT) prompting in the language domain, we propose Chain-of-Focus (CoF) Prompting, which enhances vision models by enabling step-by-step visual comprehension. CoF Prompting addresses the challenges of absent logical structure in visual data by generating intermediate reasoning steps through visual saliency. Moreover, it provides a solution for creating tailored prompts from visual inputs by selecting contextually informative prompts based on query similarity and target richness. The significance of CoF prompting is demonstrated by the recent introduction of Large Autoregressive Vision Models (LAVMs), which predict downstream targets via in-context learning with pure visual input. By integrating intermediate reasoning steps into visual prompts and effectively selecting the informative ones, the LAVMs are capable of generating significantly better inferences. Extensive experiments on downstream visual understanding tasks validate the effectiveness of our proposed method for visual in-context learning. 1

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

7 top-tier papers have cited it since. Know what they found before your reviewers do.

Questions to start from

Lune can draw on

the full paper · the top-tier papers that cite it · related work

Get the answer

Free to start. No credit card required.

Lineage

  • Muse: Text-To-Image Generation via Masked Generative Transformers cites Language Models are Few-Shot Learners
  • Large Language Models are Zero-Shot Reasoners cites Language Models are Few-Shot Learners
  • Large Language Models are Zero-Shot Reasoners cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
  • FORCE: Transferable Visual Jailbreaking Attacks via Feature Over-Reliance CorrEction cites Aligning What Matters: Masked Latent Adaptation for Text-to-Audio-Video Generation
  • Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answering cites Language Models are Few-Shot Learners
  • Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answering cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
  • Aligning What Matters: Masked Latent Adaptation for Text-to-Audio-Video Generation cites MLLM-For3D: Adapting Multimodal Large Language Model for 3D Reasoning Segmentation
  • Rethinking Model Selection in VLM Through the Lens of Gromov-Wasserstein Distance cites Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answering
Builds onCited byLeft to right by year; the more cited, the higher.

Cited by top-tier papers7

  1. MLLM-For3D: Adapting Multimodal Large Language Model for 3D Reasoning SegmentationJiaxin Huang, Runnan Chen, Ziwen Li, Zhengqing Gao et al.NeurIPS 2025 · 18 citations
  2. FORCE: Transferable Visual Jailbreaking Attacks via Feature Over-Reliance CorrEctionRunqi Lin, Alasdair Paren, Suqin Yuan, Muyang Li et al.CVPR 2026 · 13 citations
  3. Aligning What Matters: Masked Latent Adaptation for Text-to-Audio-Video GenerationJiyang Zheng, Siqi Pan, Yu Yao, Zhaoqing Wang et al.NeurIPS 2025 · 6 citations
  4. OraPO: Oracle-educated Reinforcement Learning for Data-efficient and Factual Radiology Report GenerationZhuoxiao Chen, Hongyang Yu, Ying Xu, Yadan Luo et al.CVPR 2026 · 3 citations
  5. Rethinking Model Selection in VLM Through the Lens of Gromov-Wasserstein DistanceMuyang Li, Yucheng Liu, Jianbo Ma, Elliot Osborne et al.CVPR 2026 · 2 citations

Builds on26

  1. Language Models are Few-Shot LearnersTom B. Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah et al.NeurIPS 2020 · 64,255 citations
  2. Chain-of-Thought Prompting Elicits Reasoning in Large Language ModelsJason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma et al.NeurIPS 2022 · 22,562 citations
  3. Large Language Models are Zero-Shot ReasonersTakeshi Kojima, Shixiang Shane Gu, Machel Reid, Yutaka Matsuo et al.NeurIPS 2022 · 8,168 citations
  4. Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question AnsweringPan Lu, Swaroop Mishra, Tanglin Xia, Liang Qiu et al.NeurIPS 2022 · 2,727 citations
  5. Muse: Text-To-Image Generation via Masked Generative TransformersHuiwen Chang, Han Zhang, Jarred Barber, Aaron Maschinot et al.ICML 2023 · 751 citations

Related papers

  1. Interleaved-Modal Chain-of-ThoughtJun Gao, Yongqi Li, Ziqiang Cao, Wenjie LiCVPR 2025
  2. LaRS: Latent Reasoning Skills for Chain-of-Thought ReasoningZifan Xu, Haozhu Wang, Dmitriy Bespalov, Xian Wu et al.EMNLP Findings 2024 · 4 citations
  3. CoFFT: Chain of Foresight-Focus Thought for Visual Language ModelsXinyu Zhang, Yuxuan Dong, Lingling Zhang, Chengyou Jia et al.NeurIPS 2025 · 7 citations
  4. Diffusion Guided Chain-of-Vision for Large Autoregressive Vision ModelsXinyang Wang, Kecheng Zheng, Minfeng Zhu, Wei Wu et al.CVPR 2026
  5. What Makes Good Examples for Visual In-Context Learning?Yuanhan Zhang, Kaiyang Zhou, Ziwei LiuNeurIPS 2023 · 219 citations
Dusk over the sea between two cliffs drawn in fine vertical lines