pith. sign in

arxiv: 2606.08231 · v1 · pith:HGY6XTLBnew · submitted 2026-06-06 · 💻 cs.CV

Test-Time Scaling in Multimodal Foundation Models: A Comprehensive Survey of Generation and Reasoning

Cong Wan , Ying He , Zhongzhan Huang , Hefeng Wu This is my paper

Pith reviewed 2026-06-27 19:35 UTC · model grok-4.3

classification 💻 cs.CV
keywords test-time scalingmultimodal foundation modelssurveysampling-basedfeedback-basedsearch-basedgenerationreasoning
0
0 comments X

The pith

Test-time scaling methods for multimodal foundation models fall into sampling-based, feedback-based, and search-based strategies.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper establishes a unified taxonomic framework that organizes existing test-time scaling techniques for multimodal foundation models into three categories. It reviews representative applications and benchmarks used in generation and reasoning tasks. A sympathetic reader would care because the framework supplies a systematic map of a fast-moving area where extra computation at inference time improves performance without retraining the model.

Core claim

We present the first comprehensive review of TTS research for MFMs, proposing a unified taxonomic framework that categorizes existing methodologies into three distinct strategies: sampling-based, feedback-based, and search-based approaches. We further summarize representative applications and benchmarks commonly utilized to evaluate multimodal TTS capabilities in generation and reasoning tasks. Finally, this survey discusses open challenges and outlines future research directions, providing a systematic roadmap for subsequent studies in this rapidly evolving field.

What carries the argument

The unified taxonomic framework that partitions TTS methodologies for MFMs into sampling-based, feedback-based, and search-based approaches.

Load-bearing premise

The body of existing test-time scaling methods for multimodal foundation models can be partitioned into these three categories without major omissions or overlaps that would make the grouping less useful.

What would settle it

Discovery of a test-time scaling method for multimodal models that cannot be placed in any of the three categories, or evidence of substantial overlap between categories that collapses their distinctiveness.

Figures

Figures reproduced from arXiv: 2606.08231 by Cong Wan, Hefeng Wu, Ying He, Zhongzhan Huang.

Figure 1
Figure 1. Figure 1: Recent trends in multimodal test-time scaling regarding historical evolution, publication growth, and the [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Taxonomy of multimodal test-time scaling methods. [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: , prevalent strategies primarily include Best￾of-N (BoN) and Majority Voting. A detailed tax￾onomy and summary of all surveyed methods are provided in [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Illustration of feedback-based methods. ORM: [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Illustration of search-based methods. Please [PITH_FULL_IMAGE:figures/full_fig_p007_5.png] view at source ↗
read the original abstract

Test-time Scaling (TTS) has emerged as a pivotal research direction for enhancing model performance by dynamically allocating computational resources during inference. Recent advancements have adapted this paradigm to Multimodal Foundation Models (MFMs), unlocking their potential in multimodal reasoning and generation. Despite rapid progress, the field lacks a systematic survey and unified theoretical framework to delineate the developmental landscape of multimodal TTS. To bridge this gap, we present the first comprehensive review of TTS research for MFMs, proposing a unified taxonomic framework that categorizes existing methodologies into three distinct strategies: sampling-based, feedback-based, and search-based approaches. We further summarize representative applications and benchmarks commonly utilized to evaluate multimodal TTS capabilities in generation and reasoning tasks. Finally, this survey discusses open challenges and outlines future research directions, providing a systematic roadmap for subsequent studies in this rapidly evolving field.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 0 minor

Summary. The manuscript claims to be the first comprehensive survey of Test-Time Scaling (TTS) for Multimodal Foundation Models (MFMs). It proposes a unified taxonomic framework partitioning existing methods into three distinct strategies—sampling-based, feedback-based, and search-based—while also summarizing representative applications, benchmarks for generation and reasoning tasks, open challenges, and future research directions.

Significance. A well-justified taxonomy and complete coverage would organize a fast-moving area and provide a useful roadmap; the survey format itself supplies no machine-checked proofs or parameter-free derivations but could still deliver organizing value if the partition is shown to be exhaustive and non-overlapping.

major comments (1)
  1. [Abstract] Abstract: the central claim that the three strategies constitute a 'unified taxonomic framework' with 'distinct' categories is load-bearing yet unsupported by any stated categorization criteria, decision rules, or explicit mapping of cited works to categories; without this, overlaps, omissions, or alternative groupings (e.g., optimization-based or model-internal scaling) cannot be ruled out.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their detailed review and constructive feedback on our survey. The single major comment concerns the justification of the proposed taxonomy in the abstract. We respond point-by-point below and indicate planned revisions.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the central claim that the three strategies constitute a 'unified taxonomic framework' with 'distinct' categories is load-bearing yet unsupported by any stated categorization criteria, decision rules, or explicit mapping of cited works to categories; without this, overlaps, omissions, or alternative groupings (e.g., optimization-based or model-internal scaling) cannot be ruled out.

    Authors: We agree the abstract is concise and does not spell out the decision rules. The full manuscript (Section 3) defines the categories by the primary test-time compute mechanism: sampling-based methods draw multiple independent generations and aggregate (e.g., best-of-N); feedback-based methods iteratively refine via external verifiers or critics; search-based methods explore structured spaces with algorithms such as beam search or MCTS. These criteria are applied consistently to the surveyed literature. We will revise the abstract to include a one-sentence statement of these criteria and add an explicit mapping table (new Table 1) listing representative works under each category. Regarding alternatives, optimization-based methods are covered under search-based when they allocate test-time compute via search; model-internal scaling (e.g., extended CoT) is classified under sampling or feedback depending on whether external signals are used. We believe the taxonomy is exhaustive for multimodal TTS methods that explicitly scale inference compute, and the revision will make the mapping transparent. revision: yes

Circularity Check

0 steps flagged

No circularity: literature survey with no derivations or self-referential reductions

full rationale

The paper is a survey that proposes a taxonomic framework for categorizing existing TTS methods in MFMs into sampling-based, feedback-based, and search-based approaches. No equations, predictions, fitted parameters, or derivation chains are present. The taxonomy is offered as an organizing lens drawn from the reviewed literature rather than derived from or reduced to any internal inputs or self-citations. The central claim of providing the first comprehensive review does not reduce to a self-definition or fitted input; it is a descriptive synthesis. This matches the default expectation for non-circular survey work.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

This is a survey paper; it introduces no free parameters, mathematical axioms, or invented entities.

pith-pipeline@v0.9.1-grok · 5673 in / 976 out tokens · 16736 ms · 2026-06-27T19:35:29.115805+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

121 extracted references · 78 canonical work pages · 27 internal anchors

  1. [1]

    Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, et al. 2023. Gpt-4 technical report. arXiv preprint arXiv:2303.08774

    work page internal anchor Pith review Pith/arXiv arXiv 2023

  2. [2]

    On the Opportunities and Risks of Foundation Models

    Rishi Bommasani, Drew A. Hudson, Ehsan Adeli, et al. 2022. On the opportunities and risks of foundation models. arXiv preprint arXiv:2108.07258

    work page internal anchor Pith review Pith/arXiv arXiv 2022

  3. [3]

    Tom Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared D Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, et al. 2020. Language models are few-shot learners. Advances in neural information processing systems, 33:1877--1901

    2020

  4. [4]

    Ji Young Byun, Young-Jin Park, Navid Azizan, and Rama Chellappa. 2025. Test-time-scaling for zero-shot diagnosis with visual-language reasoning. arXiv preprint arXiv:2506.11166

    work page arXiv 2025

  5. [5]

    Guo Chen, Yicheng Liu, Yifei Huang, Yuping He, Baoqi Pei, Jilan Xu, Yali Wang, Tong Lu, and Limin Wang. 2024. Cg-bench: Clue-grounded question answering benchmark for long video understanding. arXiv preprint arXiv:2412.12075

    work page arXiv 2024

  6. [6]

    Yuhang Chen, Zhen Tan, and Tianlong Chen. 2025 a . Eqa-rm: A generative embodied reward model with test-time scaling. arXiv preprint arXiv:2506.10389

    work page arXiv 2025

  7. [7]

    Zhekai Chen, Ruihang Chu, Yukang Chen, Shiwei Zhang, Yujie Wei, Yingya Zhang, and Xihui Liu. 2025 b . Tts-var: A test-time scaling framework for visual auto-regressive generation. arXiv preprint arXiv:2507.18537

    work page arXiv 2025

  8. [8]

    An-Chieh Cheng, Hongxu Yin, Yang Fu, Qiushan Guo, Ruihan Yang, Jan Kautz, Xiaolong Wang, and Sifei Liu. 2024 a . Spatialrgpt: Grounded spatial reasoning in vision-language models. Advances in Neural Information Processing Systems, 37:135062--135093

    2024

  9. [9]

    Kanzhi Cheng, Qiushi Sun, Yougang Chu, Fangzhi Xu, Yantao Li, Jianbing Zhang, and Zhiyong Wu. 2024 b . Seeclick: Harnessing gui grounding for advanced visual gui agents. arXiv preprint arXiv:2401.10935

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  10. [10]

    Yinlam Chow, Guy Tennenholtz, Izzeddin Gur, Vincent Zhuang, Bo Dai, Sridhar Thiagarajan, Craig Boutilier, Rishabh Agarwal, Aviral Kumar, and Aleksandra Faust. 2024. Inference-aware fine-tuning for best-of-n sampling in large language models. arXiv preprint arXiv:2412.15287

    work page arXiv 2024

  11. [11]

    Karl Cobbe, Vineet Kosaraju, Mohammad Bavarian, Mark Chen, Heewoo Jun, Lukasz Kaiser, Matthias Plappert, Jerry Tworek, Jacob Hilton, Reiichiro Nakano, et al. 2021. Training verifiers to solve math word problems. arXiv preprint arXiv:2110.14168

    work page internal anchor Pith review Pith/arXiv arXiv 2021

  12. [12]

    Wenyan Cong, Hanqing Zhu, Peihao Wang, Bangya Liu, Dejia Xu, Kevin Wang, David Z Pan, Yan Wang, Zhiwen Fan, and Zhangyang Wang. 2025. Can test-time scaling improve world foundation model? arXiv preprint arXiv:2503.24320

    work page arXiv 2025

  13. [13]

    Mingtong Dai, Lingbo Liu, Yongjie Bai, Yang Liu, Zhouxia Wang, Rui Su, Chunjie Chen, Liang Lin, and Xinyu Wu. 2025. Rover: Robot reward model as test-time verifier for vision-language-action model. arXiv preprint arXiv:2510.10975

    work page arXiv 2025

  14. [14]

    Karan Dalal, Daniel Koceja, Jiarui Xu, Yue Zhao, Shihao Han, Ka Chun Cheung, Jan Kautz, Yejin Choi, Yu Sun, and Xiaolong Wang. 2025. One-minute video generation with test-time training. In Proceedings of the Computer Vision and Pattern Recognition Conference, pages 17702--17711

    2025

  15. [15]

    Fernando Diaz and Michael Madaio. 2024. Scaling laws do not scale. In Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society, volume 7, pages 341--357

    2024

  16. [16]

    Guanting Dong, Chenghao Zhang, Mengjie Deng, Yutao Zhu, Zhicheng Dou, and Ji-Rong Wen. 2025. Progressive multimodal reasoning via active retrieval. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 3579--3602

    2025

  17. [17]

    Sunqi Fan, Meng-Hao Guo, and Shuojin Yang. 2025. Agentic keyframe search for video question answering. arXiv preprint arXiv:2503.16032

    work page arXiv 2025

  18. [18]

    Xinyu Fang, Kangrui Mao, Haodong Duan, Xiangyu Zhao, Yining Li, Dahua Lin, and Kai Chen. 2024. Mmbench-video: A long-form multi-shot benchmark for holistic video understanding. Advances in Neural Information Processing Systems, 37:89098--89124

    2024

  19. [19]

    Chaoyou Fu, Yuhan Dai, Yongdong Luo, Lei Li, Shuhuai Ren, Renrui Zhang, Zihan Wang, Chenyu Zhou, Yunhang Shen, Mengdan Zhang, et al. 2025. Video-mme: The first-ever comprehensive evaluation benchmark of multi-modal llms in video analysis. In Proceedings of the Computer Vision and Pattern Recognition Conference, pages 24108--24118

    2025

  20. [20]

    Zeyu Gan, Yun Liao, and Yong Liu. 2025. Rethinking external slow-thinking: From snowball errors to probability of correct reasoning. arXiv preprint arXiv:2501.15602

    work page arXiv 2025

  21. [21]

    Bingjie Gao, Qianli Ma, Xiaoxue Wu, Shuai Yang, Guanzhou Lan, Haonan Zhao, Jiaxuan Chen, Qingyang Liu, Yu Qiao, Xinyuan Chen, et al. 2025. Rapo++: Cross-stage prompt optimization for text-to-video generation via data alignment and test-time scaling. arXiv preprint arXiv:2510.20206

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  22. [22]

    Dhruba Ghosh, Hannaneh Hajishirzi, and Ludwig Schmidt. 2023. Geneval: An object-focused framework for evaluating text-to-image alignment. Advances in Neural Information Processing Systems, 36:52132--52152

    2023

  23. [23]

    Jiayuan Gu, Fanbo Xiang, Xuanlin Li, Zhan Ling, Xiqiang Liu, Tongzhou Mu, Yihe Tang, Stone Tao, Xinyue Wei, Yunchao Yao, et al. 2023. Maniskill2: A unified benchmark for generalizable manipulation skills. arXiv preprint arXiv:2302.04659

    work page arXiv 2023

  24. [24]

    Wenkai Guo, Guanxing Lu, Haoyuan Deng, Zhenyu Wu, Yansong Tang, and Ziwei Wang. 2025 a . Vla-reasoner: Empowering vision-language-action models with reasoning via online monte carlo tree search. arXiv preprint arXiv:2509.22643

    work page arXiv 2025

  25. [25]

    Ziyu Guo, Renrui Zhang, Chengzhuo Tong, Zhizheng Zhao, Rui Huang, Haoquan Zhang, Manyuan Zhang, Jiaming Liu, Shanghang Zhang, Peng Gao, et al. 2025 b . Can we generate images with cot? let's verify and reinforce image generation step by step. arXiv preprint arXiv:2501.13926

    work page arXiv 2025

  26. [26]

    Haoran He, Jiajun Liang, Xintao Wang, Pengfei Wan, Di Zhang, Kun Gai, and Ling Pan. 2025. Scaling image and video generation via test-time evolutionary search. arXiv preprint arXiv:2505.17618

    work page arXiv 2025

  27. [27]

    Jonathan Ho and Tim Salimans. 2022. Classifier-free diffusion guidance. arXiv preprint arXiv:2207.12598

    work page internal anchor Pith review Pith/arXiv arXiv 2022

  28. [28]

    Arian Hosseini, Xingdi Yuan, Nikolay Malkin, Aaron Courville, Alessandro Sordoni, and Rishabh Agarwal. 2024. V-star: Training verifiers for self-taught reasoners. arXiv preprint arXiv:2402.06457

    work page arXiv 2024

  29. [29]

    Kairui Hu, Penghao Wu, Fanyi Pu, Wang Xiao, Yuanhan Zhang, Xiang Yue, Bo Li, and Ziwei Liu. 2025. Video-mmmu: Evaluating knowledge acquisition from multi-discipline professional videos. arXiv preprint arXiv:2501.13826

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  30. [30]

    Xiwei Hu, Rui Wang, Yixiao Fang, Bin Fu, Pei Cheng, and Gang Yu. 2024 a . Ella: Equip diffusion models with llm for enhanced semantic alignment. arXiv preprint arXiv:2403.05135

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  31. [31]

    Yutao Hu, Tianbin Li, Quanfeng Lu, Wenqi Shao, Junjun He, Yu Qiao, and Ping Luo. 2024 b . Omnimedvqa: A new large-scale comprehensive evaluation benchmark for medical lvlm. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 22170--22183

    2024

  32. [32]

    Kaiyi Huang, Kaiyue Sun, Enze Xie, Zhenguo Li, and Xihui Liu. 2023. T2i-compbench: A comprehensive benchmark for open-world compositional text-to-image generation. Advances in Neural Information Processing Systems, 36:78723--78747

    2023

  33. [33]

    Qidong Huang, Xiaoyi Dong, Pan Zhang, Bin Wang, Conghui He, Jiaqi Wang, Dahua Lin, Weiming Zhang, and Nenghai Yu. 2024 a . Opera: Alleviating hallucination in multi-modal large language models via over-trust penalty and retrospection-allocation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 13418--13427

    2024

  34. [34]

    Zhongzhan Huang, Guoming Ling, Shanshan Zhong, Hefeng Wu, and Liang Lin. 2025. Minilongbench: The low-cost long context understanding benchmark for large language models. In Annual Meeting of the Association for Computational Linguistics, pages 11442--11460

    2025

  35. [35]

    Ziqi Huang, Yinan He, Jiashuo Yu, Fan Zhang, Chenyang Si, Yuming Jiang, Yuanhan Zhang, Tianxing Wu, Qingyang Jin, Nattapol Chanpaisit, et al. 2024 b . Vbench: Comprehensive benchmark suite for video generative models. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 21807--21818

    2024

  36. [36]

    Purvish Jajal, Nick John Eliopoulos, Benjamin Shiue-Hal Chou, George K Thiruvathukal, James C Davis, and Yung-Hsiang Lu. 2025. Inference-time alignment of diffusion models with evolutionary algorithms. arXiv preprint arXiv:2506.00299

    work page arXiv 2025

  37. [37]

    Suhyeok Jang, Dongyoung Kim, Changyeon Kim, Youngsuk Kim, and Jinwoo Shin. 2025. Verifier-free test-time sampling for vision language action models. arXiv preprint arXiv:2510.05681

    work page arXiv 2025

  38. [38]

    Yixin Ji, Juntao Li, Yang Xiang, Hai Ye, Kaixin Wu, Kai Yao, Jia Xu, Linjian Mo, and Min Zhang. 2025. A survey of test-time compute: From intuitive inference to deliberate reasoning. arXiv preprint arXiv:2501.02497

    work page arXiv 2025

  39. [39]

    Hongbo Jin, Ruyang Liu, Wenhao Zhang, Guibo Luo, and Ge Li. 2025. Cot-vid: Dynamic chain-of-thought routing with self verification for training-free video reasoning. arXiv preprint arXiv:2505.11830

    work page arXiv 2025

  40. [40]

    Jared Kaplan, Sam McCandlish, Tom Henighan, Tom B Brown, Benjamin Chess, Rewon Child, Scott Gray, Alec Radford, Jeffrey Wu, and Dario Amodei. 2020. Scaling laws for neural language models. arXiv preprint arXiv:2001.08361

    work page internal anchor Pith review Pith/arXiv arXiv 2020

  41. [41]

    Kangsan Kim, Geon Park, Youngwan Lee, Woongyeong Yeo, and Sung Ju Hwang. 2025. Videoicl: Confidence-based iterative in-context learning for out-of-distribution video understanding. In Proceedings of the Computer Vision and Pattern Recognition Conference, pages 3295--3305

    2025

  42. [42]

    Jacky Kwok, Christopher Agia, Rohan Sinha, Matt Foutter, Shulu Li, Ion Stoica, Azalia Mirhoseini, and Marco Pavone. 2025. Robomonkey: Scaling test-time sampling and verification for vision-language-action models. arXiv preprint arXiv:2506.17811

    work page arXiv 2025

  43. [43]

    Gyubin Lee, Bao N Nguyen Truong, Jaesik Yoon, Dongwoo Lee, Minsu Kim, Yoshua Bengio, and Sungjin Ahn. 2025. Adaptive inference-time scaling via cyclic diffusion search. In The Thirty-ninth Annual Conference on Neural Information Processing Systems

    2025

  44. [44]

    Bingxuan Li, Yiwei Wang, Jiuxiang Gu, Kai-Wei Chang, and Nanyun Peng. 2025 a . Metal: A multi-agent framework for chart generation with test-time scaling. arXiv preprint arXiv:2502.17651

    work page arXiv 2025

  45. [45]

    Chunyuan Li, Zhe Gan, Zhengyuan Yang, Jianwei Yang, Linjie Li, Lijuan Wang, Jianfeng Gao, et al. 2024 a . Multimodal foundation models: From specialists to general-purpose assistants. Foundations and Trends in Computer Graphics and Vision , 16(1-2):1--214

    2024

  46. [46]

    Junnan Li, Dongxu Li, Silvio Savarese, and Steven Hoi. 2023. Blip-2: Bootstrapping language-image pre-training with frozen image encoders and large language models. In International conference on machine learning, pages 19730--19742. PMLR

    2023

  47. [47]

    Kaixin Li, Ziyang Meng, Hongzhan Lin, Ziyang Luo, Yuchen Tian, Jing Ma, Zhiyong Huang, and Tat-Seng Chua. 2025 b . Screenspot-pro: Gui grounding for professional high-resolution computer use. arXiv preprint arXiv:2504.07981

    work page arXiv 2025

  48. [48]

    Kunchang Li, Yali Wang, Yinan He, Yizhuo Li, Yi Wang, Yi Liu, Zun Wang, Jilan Xu, Guo Chen, Ping Luo, et al. 2024 b . Mvbench: A comprehensive multi-modal video understanding benchmark. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 22195--22206

    2024

  49. [49]

    Shufan Li, Konstantinos Kallidromitis, Akash Gokul, Arsh Koneru, Yusuke Kato, Kazuki Kozuka, and Aditya Grover. 2025 c . Reflect-dit: Inference-time scaling for text-to-image diffusion transformers via in-context reflection. arXiv preprint arXiv:2503.12271

    work page arXiv 2025

  50. [50]

    Xiner Li, Masatoshi Uehara, Xingyu Su, Gabriele Scalia, Tommaso Biancalani, Aviv Regev, Sergey Levine, and Shuiwang Ji. 2025 d . Dynamic search for inference-time alignment in diffusion models. arXiv preprint arXiv:2503.02039

    work page arXiv 2025

  51. [51]

    Xuanlin Li, Kyle Hsu, Jiayuan Gu, Karl Pertsch, Oier Mees, Homer Rich Walke, Chuyuan Fu, Ishikaa Lunawat, Isabel Sieh, Sean Kirmani, et al. 2024 c . Evaluating real-world robot manipulation policies in simulation. arXiv preprint arXiv:2405.05941

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  52. [52]

    Jian Liang, Ran He, and Tieniu Tan. 2025. A comprehensive survey on test-time adaptation under distribution shifts. International Journal of Computer Vision, 133(1):31--64

    2025

  53. [53]

    Tsung-Yi Lin, Michael Maire, Serge Belongie, James Hays, Pietro Perona, Deva Ramanan, Piotr Doll \'a r, and C Lawrence Zitnick. 2014. Microsoft coco: Common objects in context. In European conference on computer vision, pages 740--755. Springer

    2014

  54. [54]

    Bo Liu, Yifeng Zhu, Chongkai Gao, Yihao Feng, Qiang Liu, Yuke Zhu, and Peter Stone. 2023 a . Libero: Benchmarking knowledge transfer for lifelong robot learning. Advances in Neural Information Processing Systems, 36:44776--44791

    2023

  55. [55]

    Fangfu Liu, Hanyang Wang, Yimo Cai, Kaiyan Zhang, Xiaohang Zhan, and Yueqi Duan. 2025. Video-t1: Test-time scaling for video generation. arXiv preprint arXiv:2503.18942

    work page arXiv 2025

  56. [56]

    Haotian Liu, Chunyuan Li, Qingyang Wu, and Yong Jae Lee. 2023 b . Visual instruction tuning. Advances in neural information processing systems, 36:34892--34916

    2023

  57. [57]

    Pan Lu, Hritik Bansal, Tony Xia, Jiacheng Liu, Chunyuan Li, Hannaneh Hajishirzi, Hao Cheng, Kai-Wei Chang, Michel Galley, and Jianfeng Gao. 2023. Mathvista: Evaluating mathematical reasoning of foundation models in visual contexts. arXiv preprint arXiv:2310.02255

    work page internal anchor Pith review Pith/arXiv arXiv 2023

  58. [58]

    Liangchen Luo, Yinxiao Liu, Rosanne Liu, Samrat Phatale, Meiqi Guo, Harsh Lara, Yunxuan Li, Lei Shu, Yun Zhu, Lei Meng, et al. 2024. Improve mathematical reasoning in language models by automated process supervision. arXiv preprint arXiv:2406.06592

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  59. [59]

    Nanye Ma, Shangyuan Tong, Haolin Jia, Hexiang Hu, Yu-Chuan Su, Mingda Zhang, Xuan Yang, Yandong Li, Tommi Jaakkola, Xuhui Jia, et al. 2025. Inference-time scaling for diffusion models beyond scaling denoising steps. arXiv preprint arXiv:2501.09732

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  60. [60]

    Oier Mees, Lukas Hermann, Erick Rosete-Beas, and Wolfram Burgard. 2022. Calvin: A benchmark for language-conditioned policy learning for long-horizon robot manipulation tasks. IEEE Robotics and Automation Letters, 7(3):7327--7334

    2022

  61. [61]

    Jiahao Meng, Shuyang Sun, Yue Tan, Lu Qi, Yunhai Tong, Xiangtai Li, and Longyin Wen. 2025. Cyberv: Cybernetics for test-time scaling in video understanding. arXiv preprint arXiv:2506.07971

    work page arXiv 2025

  62. [62]

    Yuta Oshima, Masahiro Suzuki, Yutaka Matsuo, and Hiroki Furuta. 2025. Inference-time text-to-video alignment with diffusion latent beam search. arXiv preprint arXiv:2501.19252

    work page arXiv 2025

  63. [63]

    Adam Polyak, Amit Zohar, Andrew Brown, Andros Tjandra, Animesh Sinha, Ann Lee, Apoorv Vyas, Bowen Shi, Chih-Yao Ma, Ching-Yao Chuang, et al. 2024. Movie gen: A cast of media foundation models. arXiv preprint arXiv:2410.13720

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  64. [64]

    Runqi Qiao, Qiuna Tan, Guanting Dong, Minhui Wu, Chong Sun, Xiaoshuai Song, Zhuoma GongQue, Shanglin Lei, Zhe Wei, Miaoxuan Zhang, et al. 2024. We-math: Does your large multimodal model achieve human-like mathematical reasoning? arXiv preprint arXiv:2407.01284

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  65. [65]

    Yuming Qiao, Yuechen Wang, Xudong Zhang, and Dan Meng. 2025. Ttgen: Incorporating test-time scaling to diffusion models. In Proceedings of the Computer Vision and Pattern Recognition Conference, pages 3362--3366

    2025

  66. [66]

    Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, et al. 2021. Learning transferable visual models from natural language supervision. In International conference on machine learning, pages 8748--8763

    2021

  67. [67]

    Vignav Ramesh and Morteza Mardani. 2025. Test-time scaling of diffusion models via noise trajectory search. arXiv preprint arXiv:2506.03164

    work page arXiv 2025

  68. [68]

    Arijit Ray, Jiafei Duan, Ellis Brown, Reuben Tan, Dina Bashkirova, Rose Hendrix, Kiana Ehsani, Aniruddha Kembhavi, Bryan A Plummer, Ranjay Krishna, et al. 2024. Sat: Dynamic spatial aptitude training for multimodal language models. arXiv preprint arXiv:2412.07755

    work page arXiv 2024

  69. [69]

    Chitwan Saharia, William Chan, Saurabh Saxena, Lala Li, Jay Whang, Emily L Denton, Kamyar Ghasemipour, Raphael Gontijo Lopes, Burcu Karagol Ayan, Tim Salimans, et al. 2022. Photorealistic text-to-image diffusion models with deep language understanding. Advances in neural information processing systems, 35:36479--36494

    2022

  70. [70]

    Haozhan Shen, Kangjia Zhao, Tiancheng Zhao, Ruochen Xu, Zilun Zhang, Mingwei Zhu, and Jianwei Yin. 2025. Zoomeye: Enhancing multimodal llms with human-like zooming capabilities through tree-based image exploration. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 6613--6629

    2025

  71. [71]

    Anuj Singh, Sayak Mukherjee, Ahmad Beirami, and Hadi Jamali-Rad. 2025. Code: Blockwise control for denoising diffusion models. arXiv preprint arXiv:2502.00968

    work page arXiv 2025

  72. [72]

    Charlie Snell, Jaehoon Lee, Kelvin Xu, and Aviral Kumar. 2024. Scaling llm test-time compute optimally can be more effective than scaling model parameters. arXiv preprint arXiv:2408.03314

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  73. [73]

    Lingran Song, Yucheng Zhou, and Jianbing Shen. 2025. Sim4seg: Boosting multimodal multi-disease medical diagnosis segmentation with region-aware vision-language similarity masks. arXiv preprint arXiv:2511.06665

    work page arXiv 2025

  74. [74]

    Yang Song, Jascha Sohl-Dickstein, Diederik P Kingma, Abhishek Kumar, Stefano Ermon, and Ben Poole. 2020. Score-based generative modeling through stochastic differential equations. arXiv preprint arXiv:2011.13456

    work page internal anchor Pith review Pith/arXiv arXiv 2020

  75. [75]

    Mirac Suzgun, Mert Yuksekgonul, Federico Bianchi, Dan Jurafsky, and James Zou. 2026. Dynamic cheatsheet: Test-time learning with adaptive memory. In Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers), pages 7080--7106

    2026

  76. [76]

    Gemini Team, Rohan Anil, Sebastian Borgeaud, Jean-Baptiste Alayrac, Jiahui Yu, Radu Soricut, Johan Schalkwyk, Andrew M Dai, Anja Hauth, Katie Millican, et al. 2023. Gemini: a family of highly capable multimodal models. arXiv preprint arXiv:2312.11805

    work page internal anchor Pith review Pith/arXiv arXiv 2023

  77. [77]

    Rui Tian, Mingfei Gao, Mingze Xu, Jiaming Hu, Jiasen Lu, Zuxuan Wu, Yinfei Yang, and Afshin Dehghan. 2025. Unigen: Enhanced training & test-time strategies for unified multimodal understanding and generation. arXiv preprint arXiv:2505.14682

    work page arXiv 2025

  78. [78]

    Kaishen Wang, Ruibo Chen, Tong Zheng, and Heng Huang. 2025 a . Imagent: A unified multimodal agent framework for test-time scalable image generation. arXiv preprint arXiv:2511.11483

    work page arXiv 2025

  79. [79]

    Ke Wang, Junting Pan, Weikang Shi, Zimu Lu, Houxing Ren, Aojun Zhou, Mingjie Zhan, and Hongsheng Li. 2024. Measuring multimodal mathematical reasoning with math-vision dataset. Advances in Neural Information Processing Systems, 37:95095--95169

    2024

  80. [80]

    Qiuchen Wang, Ruixue Ding, Zehui Chen, Weiqi Wu, Shihang Wang, Pengjun Xie, and Feng Zhao. 2025 b . Vidorag: Visual document retrieval-augmented generation via dynamic iterative reasoning agents. arXiv preprint arXiv:2502.18017

    work page arXiv 2025

Showing first 80 references.