Pith. sign in

REVIEW 3 major objections 3 minor 69 references

V.I.P. : Iterative Online Preference Distillation for Efficient Video Diffusion Models

T0 review · 3 major / 3 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper claims that pairing direct preference optimization with supervised fine-tuning cuts text-to-video diffusion model parameters by 36.2% and 67.5% on VideoCrafter2 and AnimateDiff while matching or exceeding full-model generation…

desk verdict A plausible and potentially important distillation idea for video diffusion, but the review packet supplies the wrong full text, so the central claims rest entirely on an abstract. read the letter →

arxiv 2508.03254 v1 pith:2RCLJABU submitted 2025-08-05 cs.CV cs.AI

Jisoo Kim , Wooseok Seo , Junwan Kim , Seungho Park , Sooyeon Park , Youngjae Yu This is my paper · ORCID
classification cs.CVcs.AI
keywords text-to-videogenerationvideodiffusionmodelsmodelcompressionknowledgedistillationdirectpreferenceoptimizationdatacurationefficientparameterreduction
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper claims that compressed text-to-video (T2V) diffusion models recover full-model quality when distillation is treated as a preference-learning problem instead of an imitation problem. The authors' ReDPO method combines direct preference optimization (DPO), which teaches the smaller student to move toward outputs that a quality signal ranks higher, with supervised fine-tuning (SFT) that upholds overall performance; this is meant to avoid the quality collapse they attribute to pure imitation training of reduced-capacity models. To feed that objective, the paper introduces V.I.P., a framework that filters and curates high-quality preferred-versus-rejected pair datasets and trains the student step-by-step online so the preference signal stays calibrated as the student improves. The reported validation on VideoCrafter2 and AnimateDiff shows parameter reductions of 36.2% and 67.5% with quality maintained or slightly improved relative to the full models. If correct, the recipe gives practitioners a concrete way to run capable video generators on a fraction of the original compute.

What carries the argument

The load-bearing mechanism is a two-part training signal. Supervised fine-tuning (SFT) gives the student a general baseline of generation quality, while direct preference optimization (DPO) supplies a targeted signal: from curated pairs of outputs labeled preferred and rejected, the student learns to increase the probability of the preferred generation, recovering specific properties without having to imitate the teacher in full. Around that objective sits V.I.P. (the paper's named framework), which filters and curates the high-quality pair datasets and imposes a step-by-step online procedure the paper calls calibrated training, so the preference orderings remain meaningful to the student as it changes during distillation. The named testbeds are VideoCrafter2 and AnimateDiff, two text-to-video diffusion models.

What would settle it

Rebuild the pipeline with corrupted preference orderings (shuffled or reversed pair ranks) while keeping the SFT term and the online schedule unchanged: if the students still match full-model quality on the same benchmarks, the preference signal is not what carries the result, and the paper's explanation of the gains would need revision.

Watch

Extended reading notes

Core claim

The central claim is that a pruned video diffusion model can be brought back to parity with its full-size teacher by teaching it which of two outputs is better, rather than by forcing it to copy the teacher's outputs. The proposed ReDPO objective layers a DPO term onto SFT: the student sees curated pairs of generations and is pushed toward the higher-ranked one, so its limited capacity goes to targeted, recoverable properties instead of being spread thin trying to match the teacher everywhere. Because the objective is only as good as its training pairs, the paper also proposes V.I.P. to assemble high-quality pair datasets and a step-by-step online update scheme for calibrated training. Validation is reported on two leading text-to-video models, VideoCrafter2 at 36.2% parameter reduction and AnimateDiff at 67.5%, with the students maintaining or surpassing the full models' performance.

Load-bearing premise

The argument depends on the curated preference pairs carrying reliable, correctly ordered quality information for the smaller student: if the ranking signal is noisy or reflects what the teacher can do rather than what the student can recover, the preference-training step would amplify that noise and the reported parity with the full models would not transfer beyond the tested benchmarks.

Editorial extensions

If this is right

  • Students trained with ReDPO and V.I.P. reach full-model quality at 36.2% (VideoCrafter2) and 67.5% (AnimateDiff) parameter reductions on the reported benchmarks.
  • Pure-SFT distillation is the failure the paper sets out to fix, so the combined objective implies that generative-model distillation should pair imitation with preference signals rather than relying on either alone.
  • Because the recipe is built on curated pair data rather than architecture-specific modules, the same ReDPO plus V.I.P. pipeline transfers to other text-to-video backbones beyond the two validated models.
  • Deploying video generation in resource-constrained settings becomes practical at roughly one-third to two-thirds fewer parameters, without the quality drop the paper attributes to naive distillation.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A testable next step the paper does not run: weaken V.I.P.'s curation (random pairs or noisier ranks) and watch student quality; the paper's account predicts a steep drop, while a flat curve would shift credit to the SFT term or the online schedule.
  • The step-by-step online preference updates resemble alignment recipes from language modeling, in which preference data is regenerated from the current policy; if that analogy holds, iterative data refresh and periodic recalibration could be imported into video-model distillation.
  • The authors leave implicit that the preference target need not be generic quality: the same machinery could steer a compressed student toward temporal consistency, prompt adherence, or safety while shrinking it, making model compression a channel for targeted behavior control.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 3 minor

Summary. The manuscript arXiv:2508.03254, as submitted, consists of an abstract for a paper titled "V.I.P.: Iterative Online Preference Distillation for Efficient Video Diffusion Models" and a full text that belongs to a different paper, "Beyond Isolated Words: Diffusion Brush for Handwritten Text-Line Generation" (arXiv:2508.03256). The abstract claims a new distillation method, ReDPO, combining DPO and SFT, plus a preference-pair filtering framework, V.I.P., applied to VideoCrafter2 and AnimateDiff, with parameter reductions of 36.2% and 67.5% while maintaining or surpassing full-model performance. No equations, evaluation protocol, ablations, or implementation details for ReDPO or V.I.P. are present in the submitted material.

Significance. If the abstract's claims are correct, the work would provide a practical recipe for compressing text-to-video diffusion models with only modest quality loss, which is a timely contribution given the cost of running such models. However, because the submitted full text is a completely different paper, the central claims are currently unverifiable. The significance is therefore conditional on the authors supplying the actual manuscript with the claimed methods and experimental details. No machine-checked proofs, reproducible code, or parameter-free derivations are present in the submitted material to offset the missing evidence.

major comments (3)
  1. [Full text (entire submission)] The submitted full text is the paper "Beyond Isolated Words: Diffusion Brush for Handwritten Text-Line Generation" (arXiv:2508.03256), not the claimed video-diffusion distillation manuscript. None of the methods ReDPO or V.I.P., nor any equations, ablations, or experimental protocols supporting the abstract's claims, are present. This is a load-bearing omission: the central claims of 36.2% and 67.5% parameter reduction with maintained or surpassed performance cannot be checked on the basis of the submitted material.
  2. [Abstract, experimental reporting] The abstract reports parameter reduction and performance parity or improvement without specifying any evaluation protocol: no prompt set, video resolution, sampling steps, guidance scale, seeds, evaluation metrics, baseline models, ablations, or error bars are given. This makes the headline claim unverifiable and leaves open the possibility that the student model is evaluated on metrics that are correlated with the V.I.P. curation filter rather than on independent human-quality judgments.
  3. [Abstract, online preference-pair curation] The abstract describes V.I.P. as "filtering and curating high-quality pair datasets" with a "step-by-step online approach for calibrated training." If the same scoring function used to curate preference pairs is also used to evaluate the distilled student, the reported parity could reflect overfitting to that criterion rather than genuine quality. The authors should specify whether preference labels come from human raters or from the teacher model, and whether the evaluation metrics are independent of the curation signal.
minor comments (3)
  1. [Abstract, terminology] The term "calibrated training" is introduced without definition; please clarify what calibration means in this context and how it is measured.
  2. [Abstract, acronyms] The acronyms "ReDPO" and "V.I.P." are used without expansion; please provide full names at first mention.
  3. [General] The project page link (https://jiiiisoo.github.io/VIP.github.io/) is provided; please ensure that it is accessible and contains the promised code and videos.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity demonstrable from the abstract: the central claim is empirical and no equation, fitted parameter, or self-citation chain is exhibited.

full rationale

Only the abstract of arXiv:2508.03254 was provided; the supplied full text belongs to a different paper (DiffBrush, arXiv:2508.03256). Within the abstract, every claim is empirical: ReDPO integrates DPO and SFT, V.I.P. curates preference pairs, and the reported outcome is a parameter-reduction and performance-parity result on VideoCrafter2 and AnimateDiff. There are no equations, no fitted parameters renamed as predictions, and no uniqueness theorem or self-citation invoked as a load-bearing premise. The plausible concern that the online preference-curation signal could be self-referential if the scoring model were the teacher being distilled is not evidenced by any quoted text, so under the hard rules it cannot be counted as circularity. Missing evaluation details are a completeness and support concern, not a demonstrated circular reduction. Therefore no significant circularity is found; the abstract is not verifiable in depth, but nothing in it reduces to its own inputs by construction.

Assumptions & free parameters 0 free parameters · 2 assumptions · 0 invented entities

The ledger is based only on the abstract. No free parameters or invented entities are identifiable; two domain assumptions carry the method's plausibility. The actual manuscript was not supplied, so the ledger is necessarily incomplete.

assumptions (2)
  • domain assumption DPO can be used as a distillation objective for reduced-capacity video diffusion models without causing mode collapse.
    The abstract motivates ReDPO from the failure of SFT-only distillation and assumes preference optimization recovers targeted properties while SFT maintains overall quality. This is the load-bearing modeling bet.
  • domain assumption Preference pairs curated by V.I.P. are reliable and aligned with human perceptual quality.
    The V.I.P. framework's utility depends on the filtering signal being a valid proxy for quality. The abstract does not define the filtering criterion.

how reviews work

0 comments
Cite this review

Pith. "Pith review of V.I.P. : Iterative Online Preference Distillation for Efficient Video Diffusion Models." pith.science (2026). https://pith.science/paper/2RCLJABU

@misc{pith2026250803254,
  author       = {Pith},
  title        = {Pith review of: V.I.P. : Iterative Online Preference Distillation for Efficient Video Diffusion Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/2RCLJABU}},
  note         = {Machine review of arXiv:2508.03254}
}
read the original abstract

With growing interest in deploying text-to-video (T2V) models in resource-constrained environments, reducing their high computational cost has become crucial, leading to extensive research on pruning and knowledge distillation methods while maintaining performance. However, existing distillation methods primarily rely on supervised fine-tuning (SFT), which often leads to mode collapse as pruned models with reduced capacity fail to directly match the teacher's outputs, ultimately resulting in degraded quality. To address this challenge, we propose an effective distillation method, ReDPO, that integrates DPO and SFT. Our approach leverages DPO to guide the student model to focus on recovering only the targeted properties, rather than passively imitating the teacher, while also utilizing SFT to enhance overall performance. We additionally propose V.I.P., a novel framework for filtering and curating high-quality pair datasets, along with a step-by-step online approach for calibrated training. We validate our method on two leading T2V models, VideoCrafter2 and AnimateDiff, achieving parameter reduction of 36.2% and 67.5% each, while maintaining or even surpassing the performance of full models. Further experiments demonstrate the effectiveness of both ReDPO and V.I.P. framework in enabling efficient and high-quality video generation. Our code and videos are available at https://jiiiisoo.github.io/VIP.github.io/.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

69 extracted references · 68 canonical work pages

  1. [1]

    Adver- sarial generation of handwritten text images conditioned on sequences

    Eloi Alonso, Bastien Moysset, and Ronaldo Messina. Adver- sarial generation of handwritten text images conditioned on sequences. In International Conference on Document Anal- ysis and Recognition, pages 481–486, 2019. 1, 2

    work page 2019

  2. [2]

    Improving image generation with better captions

    James Betker, Gabriel Goh, Li Jing, Tim Brooks, Jianfeng Wang, Linjie Li, Long Ouyang, Juntang Zhuang, Joyce Lee, Yufei Guo, et al. Improving image generation with better captions. Computer Science, 2(3):8, 2023. 12

    work page 2023

  3. [3]

    Handwriting transformers

    Ankan Kumar Bhunia, Salman Khan, Hisham Cholakkal, Rao Muhammad Anwer, Fahad Shahbaz Khan, and Mubarak Shah. Handwriting transformers. In IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 1086– 1094, 2021. 1, 2

    work page 2021

  4. [4]

    Human-guided image generation for expanding small-scale training image datasets

    Changjian Chen, Fei Lv, Yalong Guan, Pengcheng Wang, Shengjie Yu, Yifan Zhang, and Zhuo Tang. Human-guided image generation for expanding small-scale training image datasets. IEEE Transactions on Visualization and Computer Graphics, 2025. 3

    work page 2025

  5. [5]

    Improved learn- ing for online handwritten chinese text recognition with con- volutional prototype network

    Yi Chen, Heng Zhang, and Cheng-Lin Liu. Improved learn- ing for online handwritten chinese text recognition with con- volutional prototype network. In International Conference on Document Analysis and Recognition, pages 38–53, 2023. 14

    work page 2023

  6. [6]

    Disentan- gling writer and character styles for handwriting generation

    Gang Dai, Yifan Zhang, Qingfeng Wang, Qing Du, Zhu- liang Yu, Zhuoman Liu, and Shuangping Huang. Disentan- gling writer and character styles for handwriting generation. In IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 5977–5986, 2023. 2

    work page 2023

  7. [7]

    One-shot diffusion mimicker for hand- written text generation

    Gang Dai, Yifan Zhang, Quhui Ke, Qiangya Guo, and Shuangping Huang. One-shot diffusion mimicker for hand- written text generation. In European Conference on Com- puter Vision, 2024. 1, 2, 3, 5, 6, 8, 12, 13, 14, 15, 25, 26

    work page 2024

  8. [8]

    Hospedales, Tao Xi- ang, and Yi-Zhe Song

    Ayan Das, Yongxin Yang, Timothy M. Hospedales, Tao Xi- ang, and Yi-Zhe Song. Chirodiff: Modelling chirographic data with diffusion models. In International Conference on Learning Representations, 2023. 2

    work page 2023

Show all 69 references
  1. [9]

    Davis, Bryan S

    Brian L. Davis, Bryan S. Morse, Brian L. Price, Chris Tens- meyer, Curtis Wigington, and Rajiv Jain. Text and style con- ditioned gan for the generation of offline-handwriting lines. In British Machine Vision Conference, 2020. 1, 3, 5, 6, 7

    2020

  2. [10]

    Diffusion models beat gans on image synthesis

    Prafulla Dhariwal and Alexander Nichol. Diffusion models beat gans on image synthesis. Advances in Neural Informa- tion Processing Systems, 34:8780–8794, 2021. 3, 12

    2021

  3. [11]

    Scrabblegan: Semi-supervised vary- ing length handwritten text generation

    Sharon Fogel, Hadar Averbuch-Elor, Sarel Cohen, Shai Ma- zor, and Roee Litman. Scrabblegan: Semi-supervised vary- ing length handwritten text generation. In IEEE/CVF Con- ference on Computer Vision and Pattern Recognition, pages 4324–4333, 2020. 1, 2

    2020

  4. [12]

    Hi- gan+: handwriting imitation gan with disentangled represen- tations

    Ji Gan, Weiqiang Wang, Jiaxu Leng, and Xinbo Gao. Hi- gan+: handwriting imitation gan with disentangled represen- tations. ACM Transactions on Graphics, 42(1):1–17, 2022. 1, 5, 7, 16

    2022

  5. [13]

    Monocular and generalizable gaussian talking head animation

    Shengjie Gong, Haojie Li, Jiapeng Tang, Dongming Hu, Shuangping Huang, Hao Chen, Tianshui Chen, and Zhuo- man Liu. Monocular and generalizable gaussian talking head animation. In IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 5523–5534, 2025. 14

    2025

  6. [14]

    Connectionist temporal classification: labelling unsegmented sequence data with recurrent neural networks

    Alex Graves, Santiago Fern ´andez, Faustino Gomez, and J¨urgen Schmidhuber. Connectionist temporal classification: labelling unsegmented sequence data with recurrent neural networks. In International Conference on Machine Learn- ing, pages 369–376, 2006. 8, 13, 14, 15

    2006

  7. [15]

    Gans trained by a two time-scale update rule converge to a local nash equilib- rium

    Martin Heusel, Hubert Ramsauer, Thomas Unterthiner, Bernhard Nessler, and Sepp Hochreiter. Gans trained by a two time-scale update rule converge to a local nash equilib- rium. Advances in Neural Information Processing Systems , 30, 2017. 6, 14

    2017

  8. [16]

    Classifier-free diffusion guidance

    Jonathan Ho and Tim Salimans. Classifier-free diffusion guidance. arXiv, 2022. 6, 12

    2022

  9. [17]

    Denoising diffu- sion probabilistic models

    Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denoising diffu- sion probabilistic models. Advances in Neural Information Processing Systems, 33:6840–6851, 2020. 3, 12

    2020

  10. [18]

    Enhancing table structure recognition via bounding box guidance

    Lei Hu and Shuangping Huang. Enhancing table structure recognition via bounding box guidance. In International Conference on Pattern Recognition , pages 209–225, 2024. 12

    2024

  11. [19]

    Replaycad: Generative diffusion replay for continual anomaly detection

    Lei Hu, Zhiyong Gan, Ling Deng, Jinglin Liang, Lingyu Liang, Shuangping Huang, and Tianshui Chen. Replaycad: Generative diffusion replay for continual anomaly detection. arXiv, 2025. 3

    2025

  12. [20]

    Agtgan: Unpaired image translation for photographic ancient character generation

    Hongxiang Huang, Daihui Yang, Gang Dai, Zhen Han, Yuyi Wang, Kin-Man Lam, Fan Yang, Shuangping Huang, Yongge Liu, and Mengchao He. Agtgan: Unpaired image translation for photographic ancient character generation. In ACM International Conference on Multimedia, pages 5456– 5467, 2022. 2

    2022

  13. [21]

    Context-aware se- lective label smoothing for calibrating sequence recognition model

    Shuangping Huang, Yu Luo, Zhenzhou Zhuang, Jin-Gang Yu, Mengchao He, and Yongpan Wang. Context-aware se- lective label smoothing for calibrating sequence recognition model. In ACM International Conference on Multimedia , pages 4591–4599, 2021. 13

    2021

  14. [22]

    Ganwriting: content- conditioned generation of styled handwritten word images

    Lei Kang, Pau Riba, Yaxing Wang, Marc ¸al Rusinol, Ali- cia Forn ´es, and Mauricio Villegas. Ganwriting: content- conditioned generation of styled handwritten word images. In European Conference on Computer Vision , pages 273– 289, 2020. 2

    2020

  15. [23]

    Content and style aware generation of text-line images for handwriting recognition

    Lei Kang, Pau Riba, Marcal Rusinol, Alicia Fornes, and Mauricio Villegas. Content and style aware generation of text-line images for handwriting recognition. IEEE Transac- tions on Pattern Analysis and Machine Intelligence, 44(12): 8846–8860, 2021. 1, 3, 5, 6

    2021

  16. [24]

    Geometry score: A method for comparing generative adversarial networks

    Valentin Khrulkov and Ivan Oseledets. Geometry score: A method for comparing generative adversarial networks. In International conference on machine learning, pages 2621– 2629, 2018. 6

    2018

  17. [25]

    Proxy anchor loss for deep metric learning

    Sungyeon Kim, Dongwon Kim, Minsu Cho, and Suha Kwak. Proxy anchor loss for deep metric learning. In IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 3238–3247, 2020. 2, 4

    2020

  18. [26]

    Cvl-database: An off-line database for writer re- trieval, writer identification and word spotting

    Florian Kleber, Stefan Fiel, Markus Diem, and Robert Sab- latnig. Cvl-database: An off-line database for writer re- trieval, writer identification and word spotting. In Interna- tional Conference on Document Analysis and Recognition , pages 560–564, 2013. 6

    2013

  19. [27]

    Face anti-spoofing with cross-stage relation enhance- ment and spoof material perception

    Daiyuan Li, Guo Chen, Xixian Wu, Zitong Yu, and Mingkui Tan. Face anti-spoofing with cross-stage relation enhance- ment and spoof material perception. Neural Networks, 175: 106275, 2024. 12

    2024

  20. [28]

    Diffusion-driven data replay: A novel approach to combat forgetting in federated class continual learning

    Jinglin Liang, Jin Zhong, Hanlin Gu, Zhongqi Lu, Xingxing Tang, Gang Dai, Shuangping Huang, Lixin Fan, and Qiang Yang. Diffusion-driven data replay: A novel approach to combat forgetting in federated class continual learning. In European Conference on Computer Vision, pages 303–319,

  21. [29]

    Pptser: A plug-and-play tag-guided method for few-shot semantic entity recognition on visually- rich documents

    Wenhui Liao, Jiapeng Wang, Zening Lin, Longfei Xiong, and Lianwen Jin. Pptser: A plug-and-play tag-guided method for few-shot semantic entity recognition on visually- rich documents. In Findings of the Association for Compu- tational Linguistics, pages 10522–10539, 2024. 12

    2024

  22. [30]

    Doclayllm: An efficient multi- modal extension of large language models for text-rich docu- ment understanding

    Wenhui Liao, Jiapeng Wang, Hongliang Li, Chengyu Wang, Jun Huang, and Lianwen Jin. Doclayllm: An efficient multi- modal extension of large language models for text-rich docu- ment understanding. In IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 4038–4049, 2025

    2025

  23. [31]

    Contrastive representation enhancement and learning for handwritten mathematical ex- pression recognition

    Zihao Lin, Jinrong Li, Gang Dai, Tianshui Chen, Shuang- ping Huang, and Jianmin Lin. Contrastive representation enhancement and learning for handwritten mathematical ex- pression recognition. Pattern Recognition Letters, 186:14– 20, 2024. 12

    2024

  24. [32]

    Casia online and offline chinese handwriting databases

    Cheng-Lin Liu, Fei Yin, Da-Han Wang, and Qiu-Feng Wang. Casia online and offline chinese handwriting databases. In International Conference on Document Analysis and Recog- nition, pages 37–41, 2011. 6, 14

    2011

  25. [33]

    Deep view synthesis via self-consistent generative network

    Zhuoman Liu, Wei Jia, Ming Yang, Peiyao Luo, Yong Guo, and Mingkui Tan. Deep view synthesis via self-consistent generative network. IEEE Transactions on Multimedia, 24: 451–465, 2021. 2

    2021

  26. [34]

    Diffusion models for hand- writing generation

    Troy Luhman and Eric Luhman. Diffusion models for hand- writing generation. arXiv, 2020. 2

    2020

  27. [35]

    Slogan: handwriting style synthesis for arbitrary- length and out-of-vocabulary text

    Canjie Luo, Yuanzhi Zhu, Lianwen Jin, Zhe Li, and Dezhi Peng. Slogan: handwriting style synthesis for arbitrary- length and out-of-vocabulary text. IEEE Transactions on Neural Networks and Learning Systems, 34(11):8503–8515,

  28. [36]

    The iam-database: an english sentence database for offline handwriting recognition

    U-V Marti and Horst Bunke. The iam-database: an english sentence database for offline handwriting recognition. Inter- national journal on document analysis and recognition , 5: 39–46, 2002. 2, 5, 6

    2002

  29. [37]

    Pointer sentinel mixture models

    Stephen Merity, Caiming Xiong, James Bradbury, and Richard Socher. Pointer sentinel mixture models. arXiv,

  30. [38]

    Le- ung, Sergey Ioffe, and Saurabh Singh

    Yair Movshovitz-Attias, Alexander Toshev, Thomas K. Le- ung, Sergey Ioffe, and Saurabh Singh. No fuss distance met- ric learning using proxies. In IEEE International Conference on Computer Vision, pages 360–368, 2017. 2, 4

    2017

  31. [39]

    Wordstylist: Styled verbatim handwritten text generation with latent diffusion models

    Konstantina Nikolaidou, George Retsinas, Vincent Christlein, Mathias Seuret, Giorgos Sfikas, Elisa Barney Smith, Hamam Mokayed, and Marcus Liwicki. Wordstylist: Styled verbatim handwritten text generation with latent diffusion models. In International Conference on Document An...

    2023

  32. [40]

    Diffusionpen: Towards controlling the style of handwritten text generation

    Konstantina Nikolaidou, George Retsinas, Giorgos Sfikas, and Marcus Liwicki. Diffusionpen: Towards controlling the style of handwritten text generation. In European Confer- ence on Computer Vision, 2024. 1, 3, 6, 8, 13, 15

    2024

  33. [41]

    Rethinking htg evaluation: Bridg- ing generation and recognition

    Konstantina Nikolaidou, George Retsinas, Giorgos Sfikas, and Marcus Liwicki. Rethinking htg evaluation: Bridg- ing generation and recognition. In European Conference on Computer Vision workshop, 2024. 6, 13

    2024

  34. [42]

    A threshold selection method from gray-level histograms

    Nobuyuki Otsu et al. A threshold selection method from gray-level histograms. Automatica, 11(285-296):23–27,

  35. [43]

    Globally correlation- aware hard negative generation

    Wenjie Peng, Hongxiang Huang, Tianshui Chen, Quhui Ke, Gang Dai, and Shuangping Huang. Globally correlation- aware hard negative generation. International Journal of Computer Vision, pages 1–22, 2024. 12

    2024

  36. [44]

    Hand- written text generation from visual archetypes

    Vittorio Pippi, Silvia Cascianelli, and Rita Cucchiara. Hand- written text generation from visual archetypes. InIEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 22458–22467, 2023. 1, 2, 5, 6, 12, 14, 15

    2023

  37. [45]

    HWD: A novel evaluation score for styled hand- written text generation

    Vittorio Pippi, Fabio Quattrini, Silvia Cascianelli, and Rita Cucchiara. HWD: A novel evaluation score for styled hand- written text generation. In British Machine Vision Confer- ence, pages 7–9, 2023. 6, 16

    2023

  38. [46]

    Learn- ing transferable visual models from natural language super- vision

    Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, et al. Learn- ing transferable visual models from natural language super- vision. In International Conference on Machine Learning...

    2021

  39. [47]

    Diff-writer: A diffusion model-based styl- ized online handwritten chinese character generator

    Min-Si Ren, Yan-Ming Zhang, Qiu-Feng Wang, Fei Yin, and Cheng-Lin Liu. Diff-writer: A diffusion model-based styl- ized online handwritten chinese character generator. In In- ternational Conference on Neural Information Processing , pages 86–100, 2023. 2

    2023

  40. [48]

    Best practices for a handwritten text recognition system

    George Retsinas, Giorgos Sfikas, Basilis Gatos, and Christophoros Nikou. Best practices for a handwritten text recognition system. In International Conference on Docu- ment Analysis and Recognition workshop , pages 247–259,

  41. [49]

    High-resolution image syn- thesis with latent diffusion models

    Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Bj¨orn Ommer. High-resolution image syn- thesis with latent diffusion models. InIEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 10684– 10695, 2022. 3, 12

    2022

  42. [50]

    U-net: Convolutional networks for biomedical image segmentation

    Olaf Ronneberger, Philipp Fischer, and Thomas Brox. U-net: Convolutional networks for biomedical image segmentation. In International Conference on Medical Image Computing and Computer Assisted Intervention , pages 234–241, 2015. 3

    2015

  43. [51]

    Improved techniques for training gans

    Tim Salimans, Ian Goodfellow, Wojciech Zaremba, Vicki Cheung, Alec Radford, and Xi Chen. Improved techniques for training gans. Advances in Neural Information Process- ing Systems, 29, 2016. 6

    2016

  44. [52]

    Aster: An attentional scene text recognizer with flexible rectification

    Baoguang Shi, Mingkun Yang, Xinggang Wang, Pengyuan Lyu, Cong Yao, and Xiang Bai. Aster: An attentional scene text recognizer with flexible rectification. IEEE transactions on pattern analysis and machine intelligence , 41(9):2035– 2048, 2018. 5

    2018

  45. [53]

    Denois- ing diffusion implicit models

    Jiaming Song, Chenlin Meng, and Stefano Ermon. Denois- ing diffusion implicit models. In International Conference on Learning Representations, 2021. 6

    2021

  46. [54]

    Off-line recognition of realistic chinese handwriting using segmentation-free strategy

    Tong-Hua Su, Tian-Wen Zhang, De-Jun Guan, and Hu-Jie Huang. Off-line recognition of realistic chinese handwriting using segmentation-free strategy. Pattern Recognition, 42 (1):167–182, 2009. 14

    2009

  47. [55]

    Learning spatiotemporal features with 3d convolutional networks

    Du Tran, Lubomir Bourdev, Rob Fergus, Lorenzo Torresani, and Manohar Paluri. Learning spatiotemporal features with 3d convolutional networks. In IEEE International Confer- ence on Computer Vision, pages 4489–4497, 2015. 2, 5, 12

    2015

  48. [56]

    Gomez, Lukasz Kaiser, and Illia Polosukhin

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszko- reit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser, and Illia Polosukhin. Attention is all you need. In Advances in Neural Information Processing Systems, pages 5998–6008, 2017. 12

    2017

  49. [57]

    Handwrit- ten chinese text recognition by integrating multiple contexts

    Qiu-Feng Wang, Fei Yin, and Cheng-Lin Liu. Handwrit- ten chinese text recognition by integrating multiple contexts. IEEE transactions on pattern analysis and machine intelli- gence, 34(8):1469–1481, 2011. 14

    2011

  50. [58]

    Diffusion-gan: Training gans with diffusion

    Zhendong Wang, Huangjie Zheng, Pengcheng He, Weizhu Chen, and Mingyuan Zhou. Diffusion-gan: Training gans with diffusion. In International Conference on Learning Representations. OpenReview.net, 2023. 3, 12

    2023

  51. [59]

    A random-binding based bio-hashing template protection method for palm vein recog- nition

    Tianming Xie and Wenxiong Kang. A random-binding based bio-hashing template protection method for palm vein recog- nition. IEEE Transactions on Information Forensics and Se- curity, 2025. 13

    2025

  52. [60]

    Ufogen: You forward once large scale text-to-image genera- tion via diffusion gans

    Yanwu Xu, Yang Zhao, Zhisheng Xiao, and Tingbo Hou. Ufogen: You forward once large scale text-to-image genera- tion via diffusion gans. In IEEE/CVF Conference on Com- puter Vision and Pattern Recognition , pages 8196–8206,

  53. [61]

    Handwritten text gener- ation with character-specific encoding for style imitation

    Jan Zdenek and Hideki Nakayama. Handwritten text gener- ation with character-specific encoding for style imitation. In International Conference on Document Analysis and Recog- nition, pages 313–329, 2023. 7, 16

    2023

  54. [62]

    Hipa: enabling one-step text- to-image diffusion models via high-frequency-promoting adaptation

    Yifan Zhang and Bryan Hooi. Hipa: enabling one-step text- to-image diffusion models via high-frequency-promoting adaptation. arXiv, 2023. 3

    2023

  55. [63]

    Expanding small-scale datasets with guided imag- ination

    Yifan Zhang, Daquan Zhou, Bryan Hooi, Kai Wang, and Ji- ashi Feng. Expanding small-scale datasets with guided imag- ination. In Advances in Neural Information Processing Sys- tems, pages 76558–76618, 2023

    2023

  56. [64]

    Matrix-game: Interactive world foundation model

    Yifan Zhang, Chunli Peng, Boyang Wang, Puyi Wang, Qingcheng Zhu, Fei Kang, Biao Jiang, Zedong Gao, Eric Li, Yang Liu, et al. Matrix-game: Interactive world foundation model. arXiv, 2025. 3

    2025

  57. [65]

    Mpdrive: Improving spatial understanding with marker-based prompt learning for autonomous driving

    Zhiyuan Zhang, Xiaofan Li, Zhihao Xu, Wenjie Peng, Zi- jian Zhou, Miaojing Shi, and Shuangping Huang. Mpdrive: Improving spatial understanding with marker-based prompt learning for autonomous driving. In IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 120...

    2025

  58. [66]

    Memo: Memory-guided diffusion for ex- pressive talking video generation

    Longtao Zheng, Yifan Zhang, Hanzhong Guo, Jiachun Pan, Zhenxiong Tan, Jiahao Lu, Chuanxin Tang, Bo An, and Shuicheng Yan. Memo: Memory-guided diffusion for ex- pressive talking video generation. arXiv, 2024. 3

    2024

  59. [67]

    Conditional text image generation with diffusion models

    Yuanzhi Zhu, Zhaohai Li, Tianwei Wang, Mengchao He, and Cong Yao. Conditional text image generation with diffusion models. In IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 14235–14245, 2023. 3

    2023

  60. [68]

    A new semi-automatic annota- tion model via semantic boundary estimation for scene text detection

    Zhenzhou Zhuang, Zonghao Liu, Kin-Man Lam, Shuang- ping Huang, and Gang Dai. A new semi-automatic annota- tion model via semantic boundary estimation for scene text detection. In International Conference on Document Analy- sis and Recognition, pages 257–273, 2021. 13 Beyond Is...

    2021

  61. [69]

    Comparing the content discriminators and CTC recog- nizer. 2) More visual ablation results of the style mod- ule and content discriminators, 3) More visual ablation results of vertical enhancing and horizontal enhancing heads, 4) More ablation results on discriminators archi- ...

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.