Ophora: A Large-Scale Data-Driven Text-Guided Ophthalmic Surgical Video Generation Model

Guoan Wang; Junjun He; Junzhi Ning; Kaijing Zhou; Lihao Liu; Lixu Gu; Ming Hu; Wei Li; Xin Guo; Zongyuan Ge

REVIEW 5 major objections 6 minor 1 cited by

Ophora: A Large-Scale Data-Driven Text-Guided Ophthalmic Surgical Video Generation Model

T0 review · 5 major / 6 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read Ophora generates realistic ophthalmic surgical videos from natural-language instructions, and using them as training data lifts phase-recognition accuracy on the OphNet benchmark from 37.92% to 42.24%.

desk verdict A genuinely useful dataset and a plausible tuning recipe, but the headline numbers are not airtight until the authors rule out video-level leakage between Ophora and OphNet. read the letter →

arxiv 2505.07449 v7 pith:KWXGQFLZ submitted 2025-05-12 eess.IV cs.CV

Wei Li , Ming Hu , Guoan Wang , Lihao Liu , Kaijing Zhou , Junzhi Ning , Xin Guo , Zongyuan Ge

show 2 more authors

Lixu Gu Junjun He

This is my paper · ORCID

classification eess.IVcs.CV

keywords ophthalmicsurgerytext-guidedvideogenerationdiffusionmodeldatacurationinstructiontuningprivacypreservationsurgicalworkflowunderstandingaugmentation

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

The paper presents Ophora, a text-guided video generation model that synthesizes ophthalmic surgical videos from natural-language instructions. To train it, the authors built Ophora-160K, a dataset of over 160K video-instruction pairs, by refining captions from narrated surgical videos with a large language model and filtering clips on visual dynamics and resolution. They then progressively tuned a pre-trained video diffusion model, first on the full dataset and then on a privacy-cleaned subset, to transfer natural-video knowledge to the surgical domain. The paper's central claim is that the resulting videos are realistic and instruction-following, as judged by automated metrics and ophthalmologist ratings, and that they are useful as training data: adding them to a real surgical benchmark raises phase-level Top-1 accuracy from 37.92% to 42.24% for an MViTv2 classifier on the OphNet test set.

What carries the argument

The argument turns on two coordinated mechanisms. The Comprehensive Data Curation pipeline turns narrative ophthalmic videos into a large instruction-following dataset: Qwen2.5-72B, prompted with ten hand-written examples, rewrites verbose narrative captions into concise generation instructions; PySceneDetect keyframe counts are used to discard clips with extreme dynamics (fewer than 2 or more than 100 keyframes); and clips below 720×480 are removed. The Progressive Video-Instruction Tuning scheme then adapts CogVideoX-2b, a latent diffusion model pre-trained on natural video-text pairs, to surgery: transfer pre-training trains only the denoising network on all of Ophora-160K while the T5 text encoder and 3D VAE stay frozen, and privacy-preserving fine-tuning continues on Ophora-28K, a subset cleared by Qwen2.5-VL-72B of frames carrying subtitles or watermarks. The training objective is the standard diffusion denoising loss of Eq. (1).

What would settle it

Take 100 held-out instruction prompts from the Ophora test split, generate videos with Ophora, and have two independent ophthalmologists rate each video on a fine-grained checklist: correct instrument used, correct tissue/anatomy affected, and correct order of sub-actions. If the mean score on any of those categories falls below 2 on the paper's own 0-3 realism scale, the claim that Ophora reliably follows surgeon instructions would be contradicted.

Watch

Extended reading notes

Core claim

Ophora, built from the CogVideoX-2b latent diffusion backbone, can generate realistic, instruction-following ophthalmic surgical videos and does so without reproducing sensitive overlays such as subtitles or watermarks. The authors establish this by constructing the Ophora-160K dataset from OphVL's narrative videos through LLM-based instruction refinement and keyframe-dynamics filtering, then applying two-stage tuning: transfer pre-training on the full dataset followed by privacy-preserving fine-tuning on a filtered 28K subset. They report that Ophora outperforms the Endora and Bora baselines on FID, FVD, and OphCLIP-based CLIPScore, and that three ophthalmologists rate its generations as realistic across seven criteria covering phase, construction, and action. Finally, they show a downstream benefit: using Ophora-generated videos as additional training data improves phase and operation recognition on the OphNet benchmark across both SlowFast and MViTv2 classifiers.

Load-bearing premise

The load-bearing assumption is that the LLM-rewritten instructions are medically accurate, aligned with each clip, and free of clinically meaningful errors, since the rewriting was taught with only ten hand-annotated examples and was never validated against a ground-truth surgical description.

Editorial extensions

If this is right

If the claims hold, surgical AI training can use Ophora-generated videos to augment scarce real recordings, with a measured phase-level Top-1 gain from 37.92% to 42.24% on the OphNet test set for MViTv2.
Because the final fine-tuning step removes subtitles and watermarks, the pipeline produces shareable video data that avoids the privacy obstacles attached to real ophthalmic recordings.
The instruction-following capability supports on-demand generation of specific procedural steps, which the authors position as a resource for training and rehearsal.
The authors state the same approach can be extended to other surgery types and longer-duration videos, a direction they list as future work.

Reading between the lines

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

A direct testable extension is applying the same curation-plus-tuning recipe to other surgical video archives; the pipeline's components are not ophthalmic-specific, though the paper does not claim this.
The downstream accuracy gain may reflect increased training set size rather than the realism of Ophora's videos; holding the number of augmented videos constant while varying their source (e.g., real vs. Ophora vs. an untuned generator) would isolate the mechanism.
Given the 10-example instruction rewrite stage, regenerating instructions with more expert-annotated examples or adding a surgical-knowledge filter would likely change both FVD/CLIPScore and downstream gains, a comparison the paper does not run.
The model's clips average 5.54 seconds, so whether multi-phase, longer-duration generation preserves cross-phase instrument and tissue dynamics is an open question that the paper's future-work statement acknowledges.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

You should know this paper before citing its numbers: Ophora is a real engineering effort—a 162K clip-instruction ophthalmic surgery dataset and a CogVideoX-2b model fine-tuned in two stages—but the evaluation as written does not rule out leakage between the synthetic-augmentation training data and the OphNet test set. That makes the headline +4.3 point Top-1 gain (37.92→42.24) plausible but not yet trustworthy.

What is genuinely new: Ophora-160K is, as far as I know, the first large-scale text-video dataset for ophthalmic surgery. The curation pipeline—LLM-based caption rewriting with a 10-example in-context prompt, dynamics-based keyframe filtering, resolution floor—is sensible. The two-stage training (transfer pre-training on the full set, then privacy-preserving fine-tuning on the 28K clean subset) is a reasonable design, and releasing code is good practice. The ophthalmologist feedback is a useful addition, and FID/FVD are standard.

The soft spots are in the evaluation, not the construction. Section 3 says the dataset was split 80/20 but never says the split is video-level. Ophora-160K has 16.5 clips per source video on average, so a clip-level split would put near-duplicates in both train and test, inflating Table 1. More importantly, Table 2's downstream gain is only interpretable if Ophora's training sources are disjoint from OphNet's test videos. Both are assembled from public surgical videos; the paper shows no such disjointness analysis. If the generated augmentation videos carry test-domain appearance, the accuracy jump could be memorization rather than useful synthesis. That is load-bearing, and the authors should be asked to address it.

Also missing: SurgEN, the most relevant text-guided surgical video baseline, is cited but never compared. The CLIPScore uses OphCLIP, from the same group and data lineage as the training captions, which is at least a suspicious choice for an alignment metric. There are no error bars or significance tests. The privacy claim is overstated: filtering subtitles and watermarks does not guarantee that the model hasn't memorized patient-specific visual features. The unvalidated LLM-refined instructions are a real concern for the dataset's semantic correctness, though it doesn't directly threaten Table 2, where prompts were written by ophthalmologists.

Overall, the core idea is sound and the dataset is a useful community asset. But the paper's strongest quantitative claim is not yet supported. It should go to peer review with a request for video-level split details, a disjointness analysis against OphNet, a SurgEN comparison, and variance estimates.

For whom: surgical AI and medical video generation researchers. Worth a reading-group slot on evaluation pitfalls in generative medical video.

Recommendation: engage, but push on the leakage issue.

Referee Report

5 major / 6 minor

Summary. The paper presents Ophora, a text-guided video generation model for ophthalmic surgery. The authors construct Ophora-160K, a dataset of over 160K video-instruction pairs, by curating OphVL narrative videos: they use Qwen2.5-72B to rewrite narrative captions into generation instructions, apply dynamics-based filtering, and remove low-resolution clips. The model is based on CogVideoX-2b and is trained in two stages: transfer pre-training on the full Ophora-160K, followed by privacy-preserving fine-tuning on Ophora-28K, which excludes clips flagged by Qwen2.5-VL-72B as containing subtitles or watermarks. The authors evaluate video quality with FID, FVD, and CLIPScore, collect ophthalmologist ratings, and test downstream value by using Ophora-generated videos as augmented training data for OphNet workflow understanding, reporting improved phase-level and operation-level Top-1/Top-5 accuracy.

Significance. If the results hold, the paper makes a useful contribution: a large curated ophthalmic surgical video-instruction dataset, a privacy-aware two-stage tuning approach for adapting a natural video T2V model to a medical domain, and evidence that synthetic surgical videos can improve downstream workflow understanding. The release of code and the dataset is a practical strength, as is the inclusion of ophthalmologist feedback. However, the central quantitative claims are weakened by the lack of source-level split guarantees, potential data overlap between Ophora-160K and OphNet, and the use of a circular text-video alignment metric. These issues leave the strongest claims (Table 1 and Table 2) not fully supported.

major comments (5)

[Section 3, Table 2 and Dataset description] The downstream augmentation result is load-bearing, but the paper does not establish that the Ophora-160K training sources are disjoint from the OphNet test set. OphVL collects public ophthalmic surgery videos and OphNet is also assembled from public videos. If Ophora-generated videos used for augmentation were trained on clips from the same source videos or same surgical cases as the OphNet test split, the reported Top-1 accuracy gain from 37.92% to 42.24% could reflect memorization of test-domain appearance rather than genuine generalization. The paper must provide a source-level or surgery-level disjointness analysis, or explicitly justify why overlap is impossible.
[Section 3, Dataset split] The sentence 'The dataset was split into 80% and 20% for training and testing' does not specify the granularity of the split. Since Ophora-160K consists of short clips cut from longer narrative videos, a clip-level split places temporally adjacent clips from the same surgery in both training and test partitions. This can inflate the FID, FVD, and CLIPScore numbers in Table 1 by measuring near-duplicate temporal content, and it can also leak training information into the test set. The split should be performed at the source-video or surgery level, or the authors should justify why clip-level splitting is sufficient.
[Section 3, Quantitative Analysis and equation for CLIPScore] The CLIPScore is computed with OphCLIP, which is developed on OphVL narrative data from which Ophora-160K is derived. Since Ophora is trained on the same narrative distribution, the text-video alignment score may be biased in favor of Ophora by construction. This makes the CS column in Table 1 a weak form of evidence for instruction following. The authors should report scores with an independent CLIP variant or a CLIP model not fine-tuned on OphVL data, and should also present error bars or significance tests over multiple generated video samples.
[Section 2.1, Narrative Information Refining] The LLM-based refinement uses only 10 hand-annotated examples to instruct Qwen2.5-72B on rewriting narrative captions into generation instructions, but no validation is provided that the rewritten instructions preserve procedure-critical details or maintain temporal alignment with video content. Since Ophora-160K is the foundation of the model's instruction-following capability, the paper should include a human evaluation of a random sample of refined instructions to confirm their accuracy, completeness, and clinical correctness.
[Section 3, Quantitative Analysis and References] The paper does not compare against SurgEN [4], a text-guided diffusion model for surgical video generation that is cited in the introduction as related work. Given the paper's claim of being a text-guided ophthalmic surgical video generation model, a direct comparison with SurgEN on quality or downstream metrics is necessary to substantiate the stated advantage over existing text-guided surgical generation approaches.

minor comments (6)

[Abstract and Section 1] The phrase 'privacy-preserved ophthalmic surgical video generation' is used in the abstract and introduction, but the privacy mechanism described later only removes subtitles and watermarks. Clarify the scope of the privacy guarantee, since removing these artifacts does not necessarily prevent re-identification from surgical content.
[Figure 1 caption] The abbreviation P2FT is used in the figure but is not defined in the figure caption or in the main text before its usage. Define it at first mention in Section 2.3.
[Table 1] The row labeled 'OphVL [12]' is ambiguous: it appears to be a dataset setting, but the table caption describes model comparisons. Clarify what is being evaluated in each row and what the column 'OphVL [12]' indicates.
[Figure 2 caption] The caption marks Endora and Bora with an asterisk indicating fine-tuning on Ophora-160K, but the main text states that Endora is an unconditional generation model. This makes the comparison somewhat less direct; please state the conditionality of each model explicitly in the caption or text.
[Section 3, Ophthalmologist Feedback] The paper states that 600 videos were generated based on Cataract-1K phase labels, but it does not specify how many videos were generated per instruction or whether the ophthalmologist-written instructions were verified before generation. Adding this detail would help reproducibility.
[References] Several references are cited in an incomplete or non-uniform style (e.g., reference [10] and [12] have publisher information; others lack it). Please ensure all references follow a consistent journal format.

Circularity Check

1 steps flagged · score 4.0 of 10

CLIPScore evaluation is computed with OphCLIP, a same-group model pretrained on OphVL, the parent dataset of Ophora-160K, making part of the text-alignment evidence self-referential; the core generation and downstream OphNet experiments retain independent content.

self citation load bearing [Section 3, Quantitative Analysis (Table 1)]
"We employ three metrics: Fréchet Inception Distance (FID) [9], Fréchet Video Distance (FVD) [24], and CLIPScore (CS) [8] to evaluate the realism and video-text consistency of the synthesized videos. The CS is calculated with the coefficient ω = 100 based on OphCLIP [12], which aligns ophthalmic clips with narrative texts."

OphCLIP [12] is prior work by overlapping authors (including M. Hu and W. Li) and is trained on OphVL, the same narrative video-caption collection from which Ophora-160K is derived: Section 2.1 states 'OphVL [12] collects large-scale ophthalmic surgical narrative videos from the Internet and converts them into video clip-caption pairs,' and Ophora-160K is built by refining those captions. Thus the CLIPScore evaluator has been optimized on the very distribution from which Ophora's training videos and instructions are drawn, so a high CS partly reflects that the generated samples match the metric's own training domain rather than independent evidence of instruction following.

full rationale

The central generation pipeline is not circular by construction: Ophora is a CogVideoX-2b backbone fine-tuned on Ophora-160K with the standard diffusion objective in Eq. 1, and its downstream workflow-understanding claim is tested on the external OphNet benchmark using SlowFast and MViTv2 classifiers. The 80/20 split ambiguity is a data-leakage risk, not a demonstration that any result equals its input by definition. The only concrete circular element is the CLIPScore metric, which uses OphCLIP, a same-author model pretrained on OphVL, the parent dataset of Ophora-160K; this makes one of the three quantitative evaluation measures partially self-referential. Because the realism and downstream claims also rest on independent FID/FVD numbers, human ophthalmologist ratings, and an external benchmark, the paper is not fundamentally circular but deserves a moderate score for relying on a self-referential evaluation instrument for its text-consistency claim.

Assumptions & free parameters 6 free parameters · 6 assumptions · 0 invented entities

The paper introduces no new physical or conceptual entities such as forces, particles, or dimensions. Ophora and Ophora-160K are engineered artifacts, not postulates requiring independent evidence. The main ledger items are the hand-chosen filtering thresholds, training hyperparameters, and domain assumptions about data quality and transfer learning.

free parameters (6)

Keyframe filtering thresholds = upper=100, lower=2
Section 2.1 sets these thresholds empirically to filter clips with extreme temporal dynamics; they directly determine which clips enter Ophora-160K.
Minimum resolution threshold = 720x480
Section 2.1 filters out clips below 720x480, affecting dataset size and visual fidelity.
Frame sampling count = 49 frames
Section 3 describes resizing and uniformly sampling or padding all clips to 49 frames, setting the temporal length of training and generation.
Training hyperparameters = LR 1e-4/5e-5, batch 128, iterations 65000/4500
Section 3 Implementation uses these hand-chosen values without ablation; they influence final model quality.
Sensitive frame sampling rate = 1 FPS
Section 2.3 uses 1 FPS sampling for Qwen2.5-VL-72B to detect sensitive content, affecting which clips remain in Ophora-28K.
LLM few-shot examples = 10 captions
Section 2.1 manually annotates 10 captions as examples for Qwen2.5-72B, influencing the style and reliability of all generated instructions.

assumptions (6)

domain assumption OphVL narrative captions are reliable descriptions of visible surgical content.
Section 2.1 builds Ophora-160K from OphVL; caption quality is inherited without re-verification.
domain assumption Qwen2.5-72B accurately converts narrative captions into clean generation instructions.
Section 2.1 uses the LLM with only 10 hand-annotated examples and does not validate whether instructions preserve clinically relevant details.
domain assumption Spatial-temporal knowledge from natural-video pretraining transfers to ophthalmic surgery via continual pretraining.
Section 2.3 relies on this transfer hypothesis; the paper only supports it indirectly through quality metrics.
ad hoc to paper Removing frames with visible subtitles or watermarks yields privacy-preserving video generation.
Section 2.3 and the abstract claim privacy preservation, but filtering visible overlays does not remove patient-identifying content in the underlying surgical scenes.
domain assumption FID, FVD, CLIPScore, and ophthalmologist scores measure clinically meaningful video realism.
Section 3 uses these proxy metrics as the sole evidence for realism; no clinical outcome or anatomical correctness check is performed.
standard math The standard diffusion training objective and latent VAE from CogVideoX-2b are valid background tools.
Equation (1) states the standard diffusion loss; this is uncontroversial background.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Ophora: A Large-Scale Data-Driven Text-Guided Ophthalmic Surgical Video Generation Model." pith.science (2026). https://pith.science/paper/KWXGQFLZ

@misc{pith2026250507449,
  author       = {Pith},
  title        = {Pith review of: Ophora: A Large-Scale Data-Driven Text-Guided Ophthalmic Surgical Video Generation Model},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KWXGQFLZ}},
  note         = {Machine review of arXiv:2505.07449}
}

read the original abstract

In ophthalmic surgery, developing an AI system capable of interpreting surgical videos and predicting subsequent operations requires numerous ophthalmic surgical videos with high-quality annotations, which are difficult to collect due to privacy concerns and labor consumption. Text-guided video generation (T2V) emerges as a promising solution to overcome this issue by generating ophthalmic surgical videos based on surgeon instructions. In this paper, we present Ophora, a pioneering model that can generate ophthalmic surgical videos following natural language instructions. To construct Ophora, we first propose a Comprehensive Data Curation pipeline to convert narrative ophthalmic surgical videos into a large-scale, high-quality dataset comprising over 160K video-instruction pairs, Ophora-160K. Then, we propose a Progressive Video-Instruction Tuning scheme to transfer rich spatial-temporal knowledge from a T2V model pre-trained on natural video-text datasets for privacy-preserved ophthalmic surgical video generation based on Ophora-160K. Experiments on video quality evaluation via quantitative analysis and ophthalmologist feedback demonstrate that Ophora can generate realistic and reliable ophthalmic surgical videos based on surgeon instructions. We also validate the capability of Ophora for empowering downstream tasks of ophthalmic surgical workflow understanding. Code is available at https://github.com/uni-medical/Ophora.

Figures

Figures reproduced from arXiv: 2505.07449 by the authors.

**Figure 1.** Illustration of proposed Ophora that can generate ophthalmic surgical videos from instructions. Specifically, we propose a Comprehensive Data Curation pipeline to construct a large-scale, high-quality video-instruction dataset, Ophora-160K, from narrative videos (Sec. 2.1). We introduce a T2V model pre-trained on natural videotext pairs (Sec. 2.2) and leverage Progressive Video-Instruction Tuning to transfer spatia… view at source ↗

**Figure 2.** Synthesized video frames from the input text prompts of different models. ‘*’ denotes that this model was fine-tuned on the proposed Ophora-160K. Construction Realism Phase Completeness Phase Matching Action Effectiveness Action Logic Action Realism Construction Stability 0 1 2 3 (a) Ophthalmologists’ scores on different criteria Hydrodissection Lens Implantation Capsulorhexis Capsule Pulishing Phacoemulsification … view at source ↗

**Figure 3.** Ophthalmologists’ scores on different criteria (a) and surgical phases (b). tual procedures; 6) Action Logic reflecting the consistency of the sequential order of surgical actions with actual orders; 7) Action Effectiveness, meaning the realism of the deformation effects on the corresponding tissues caused by surgical actions. Each criterion is scored on a scale of 0 to 3, representing different levels of realism, … view at source ↗

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

RetinaLogos: Fine-Grained Synthesis of High-Resolution Retinal Images Through Captions
eess.IV 2025-05 conditional novelty 7.0 of 10

A large captioned retinal dataset and a three-step flow-matching text-to-image model enable fine-grained, caption-controlled synthesis of realistic color fundus photographs.

Reference graph

Works this paper leans on

31 extracted references · 19 canonical work pages · cited by 1 Pith paper

[4]

arXiv preprint arXiv:2408.14028 (2024)

Cho, J., Schmidgall, S., Zakka, C., Mathur, M., Kaur, D., Shad, R., Hiesinger, W.: Surgen: Text-guided diffusion model for surgical video generation. arXiv preprint arXiv:2408.14028 (2024)

arXiv 2024
[1]

5-vl technical report

Bai, S., Chen, K., Liu, X., Wang, J., Ge, W., Song, S., Dang, K., Wang, P., Wang, S., Tang, J., et al.: Qwen2. 5-vl technical report. arXiv preprint arXiv:2502.13923 (2025)

arXiv 2025
[2]

Journal of Robotic Surgery18(1), 74 (2024)

Cepolina, F., Razzoli, R.: Review of robotic surgery platforms and end effectors. Journal of Robotic Surgery18(1), 74 (2024)

work page 2024
[3]

In: International Workshop on Trustworthy Machine Learning for Healthcare

Cheng, Y., Liu, L., Wang, S., Jin, Y., Schönlieb, C.B., Aviles-Rivero, A.I.: Why deep surgical models fail?: Revisiting surgical action triplet recognition through the lens of robustness. In: International Workshop on Trustworthy Machine Learning for Healthcare. pp. 177–189. Springer (2023)

work page 2023
[5]

In: Proceedings of the IEEE/CVF international conference on computer vision

Feichtenhofer, C., Fan, H., Malik, J., He, K.: Slowfast networks for video recog- nition. In: Proceedings of the IEEE/CVF international conference on computer vision. pp. 6202–6211 (2019)

2019
[6]

Scientific Data11(1), 373 (2024)

Ghamsarian, N., El-Shabrawi, Y., Nasirihaghighi, S., Putzgruber-Adamitsch, D., Zinkernagel, M., Wolf, S., Schoeffmann, K., Sznitman, R.: Cataract-1k dataset for deep-learning-assisted analysis of cataract surgery videos. Scientific Data11(1), 373 (2024)

work page 2024
[7]

Eye35(12), 3192–3201 (2021)

He, B., de Smet, M.D., Sodhi, M., Etminan, M., Maberley, D.: A review of robotic surgical training: establishing a curriculum and credentialing process in ophthal- mology. Eye35(12), 3192–3201 (2021)

work page 2021
[8]

Hessel, J., Holtzman, A., Forbes, M., Bras, R.L., Choi, Y.: Clipscore: A reference- freeevaluationmetricforimagecaptioning.arXivpreprintarXiv:2104.08718(2021)

arXiv 2021

Show all 31 references

[9]

In: Guyon, I., Luxburg, U.V., Bengio, S., Wallach, H., Fergus, R., Vishwanathan, S., Garnett, R

Heusel,M.,Ramsauer,H.,Unterthiner,T.,Nessler,B.,Hochreiter,S.:Ganstrained by a two time-scale update rule converge to a local nash equilibrium. In: Guyon, I., Luxburg, U.V., Bengio, S., Wallach, H., Fergus, R., Vishwanathan, S., Garnett, R. (eds.) Advances in Neural Informatio...

2017
[10]

(eds.) Advances in Neural Infor- mation Processing Systems

Ho,J.,Jain,A.,Abbeel,P.:Denoisingdiffusionprobabilisticmodels.In:Larochelle, H., Ranzato, M., Hadsell, R., Balcan, M., Lin, H. (eds.) Advances in Neural Infor- mation Processing Systems. vol. 33, pp. 6840–6851. Curran Associates, Inc. (2020)

2020
[11]

In: Leonardis, A., Ricci, E., Roth, S., Russakovsky, O., Sattler, T., Varol, G

Hu, M., Xia, P., Wang, L., Yan, S., Tang, F., Xu, Z., Luo, Y., Song, K., Leitner, J., Cheng, X., Cheng, J., Liu, C., Zhou, K., Ge, Z.: Ophnet: A large-scale video bench- mark for ophthalmic surgical workflow understanding. In: Leonardis, A., Ricci, E., Roth, S., Russakovsky, O...
[12]

arXiv preprint arXiv:2411.15421 (2024)

Hu, M., Yuan, K., Shen, Y., Tang, F., Xu, X., Zhou, L., Li, W., Chen, Y., Xu, Z., Peng, Z., et al.: Ophclip: Hierarchical retrieval-augmented learning for ophthalmic surgical video-language pretraining. arXiv preprint arXiv:2411.15421 (2024)

2024 arXiv
[13]

In: 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

Huang, Z., He, Y., Yu, J., Zhang, F., Si, C., Jiang, Y., Zhang, Y., Wu, T., Jin, Q., Chanpaisit, N., Wang, Y., Chen, X., Wang, L., Lin, D., Qiao, Y., Liu, Z.: Vbench: Comprehensive benchmark suite for video generative models. In: 2024 IEEE/CVF Conference on Computer Vision and...

2024
[14]

In: Mukhopadhyay, A., Oksuz, I., Engelhardt, S., Mehrof, D., Yuan, Y

Iliash, I., Allmendinger, S., Meissen, F., Kühl, N., Rückert, D.: Interactive genera- tion of laparoscopic videos with diffusion models. In: Mukhopadhyay, A., Oksuz, I., Engelhardt, S., Mehrof, D., Yuan, Y. (eds.) Deep Generative Models. pp. 109–118. Springer Nature Switzerlan...

2025
[15]

IEEE Transactions on Medical Imaging37(5), 1114–1126 (2018)

Jin, Y., Dou, Q., Chen, H., Yu, L., Qin, J., Fu, C.W., Heng, P.A.: Sv-rcnet: Work- flow recognition from surgical videos using recurrent convolutional network. IEEE Transactions on Medical Imaging37(5), 1114–1126 (2018)

2018
[16]

In: Globerson, A., Mackey, L., Belgrave, D., Fan, A., Paquet, U., Tomczak, J., Zhang, C

Ju, X., Gao, Y., Zhang, Z., Yuan, Z., Wang, X., ZENG, A., Xiong, Y., Xu, Q., Shan, Y.: Miradata: A large-scale video dataset with long durations and struc- tured captions. In: Globerson, A., Mackey, L., Belgrave, D., Fan, A., Paquet, U., Tomczak, J., Zhang, C. (eds.) Advances ...

2024
[17]

In: Linguraru, M.G., Dou, Q., Feragen, A., Giannarou, S., Glocker, B., Lekadir, K., Schnabel, J.A

Li, C., Liu, H., Liu, Y., Feng, B.Y., Li, W., Liu, X., Chen, Z., Shao, J., Yuan, Y.: Endora: Video generation models as endoscopy simulators. In: Linguraru, M.G., Dou, Q., Feragen, A., Giannarou, S., Glocker, B., Lekadir, K., Schnabel, J.A. (eds.) Medical Image Computing and C...

2024
[18]

In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

Li, Y., Wu, C.Y., Fan, H., Mangalam, K., Xiong, B., Malik, J., Feichtenhofer, C.: Mvitv2: Improved multiscale vision transformers for classification and detec- tion. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 4804–4814 (2022)

2022
[19]

Proceedings of the AAAI Conference on Artificial Intelligence32(1) (Apr 2018)

Li, Y., Min, M., Shen, D., Carlson, D., Carin, L.: Video generation from text. Proceedings of the AAAI Conference on Artificial Intelligence32(1) (Apr 2018)

2018
[20]

Nature medicine 25(1), 37–43 (2019)

Price, W.N., Cohen, I.G.: Privacy in the age of medical big data. Nature medicine 25(1), 37–43 (2019)

2019
[21]

Raffel, C., Shazeer, N., Roberts, A., Lee, K., Narang, S., Matena, M., Zhou, Y., Li, W., Liu, P.J.: Exploring the limits of transfer learning with a unified text-to-text transformer. J. Mach. Learn. Res.21(1) (Jan 2020)

2020
[22]

arXiv preprint arXiv:2407.08944 (2024)

Sun, W., You, X., Zheng, R., Yuan, Z., Li, X., He, L., Li, Q., Sun, L.: Bora: Biomedical generalist video generation model. arXiv preprint arXiv:2407.08944 (2024)

2024 arXiv
[23]

Survey of Ophthalmology 64(4), 570–578 (2019) Ophora 11

Thia, B.C., Wong, N.J., Sheth, S.J.: Video recording in ophthalmic surgery. Survey of Ophthalmology 64(4), 570–578 (2019) Ophora 11

2019
[24]

arXiv preprint arXiv:1812.01717 (2018)

Unterthiner, T., Van Steenkiste, S., Kurach, K., Marinier, R., Michalski, M., Gelly, S.: Towards accurate generative models of video: A new metric & challenges. arXiv preprint arXiv:1812.01717 (2018)

2018 arXiv
[25]

Nature Medicine30(5), 1257–1268 (2024)

Varghese, C., Harrison, E.M., O’Grady, G., Topol, E.J.: Artificial intelligence in surgery. Nature Medicine30(5), 1257–1268 (2024)

2024
[26]

5 technical report

Yang, A., Yang, B., Zhang, B., Hui, B., Zheng, B., Yu, B., Li, C., Liu, D., Huang, F., Wei, H., et al.: Qwen2. 5 technical report. arXiv preprint arXiv:2412.15115 (2024)

2024 arXiv
[27]

arXiv preprint arXiv:2408.06072 (2024)

Yang, Z., Teng, J., Zheng, W., Ding, M., Huang, S., Xu, J., Yang, Y., Hong, W., Zhang, X., Feng, G., et al.: Cogvideox: Text-to-video diffusion models with an expert transformer. arXiv preprint arXiv:2408.06072 (2024)

2024 arXiv
[28]

In: Linguraru, M.G., Dou, Q., Feragen, A., Giannarou, S., Glocker, B., Lekadir, K., Schnabel, J.A

Yuan, K., Srivastav, V., Navab, N., Padoy, N.: Hecvl: Hierarchical video-language pretraining for zero-shot surgical phase recognition. In: Linguraru, M.G., Dou, Q., Feragen, A., Giannarou, S., Glocker, B., Lekadir, K., Schnabel, J.A. (eds.) Medical Image Computing and Compute...

2024
[29]

Proceedings of the IEEE109(1), 43– 76 (2021)

Zhuang, F., Qi, Z., Duan, K., Xi, D., Zhu, Y., Zhu, H., Xiong, H., He, Q.: A comprehensive survey on transfer learning. Proceedings of the IEEE109(1), 43– 76 (2021)

2021
[316]

Springer Nature Switzerland, Cham (2024)

2024
[2024]

pp. 481–500. Springer Nature Switzerland, Cham (2025)

2025

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.

[1] [4]

arXiv preprint arXiv:2408.14028 (2024)

Cho, J., Schmidgall, S., Zakka, C., Mathur, M., Kaur, D., Shad, R., Hiesinger, W.: Surgen: Text-guided diffusion model for surgical video generation. arXiv preprint arXiv:2408.14028 (2024)

arXiv 2024

[2] [1]

5-vl technical report

Bai, S., Chen, K., Liu, X., Wang, J., Ge, W., Song, S., Dang, K., Wang, P., Wang, S., Tang, J., et al.: Qwen2. 5-vl technical report. arXiv preprint arXiv:2502.13923 (2025)

arXiv 2025

[3] [2]

Journal of Robotic Surgery18(1), 74 (2024)

Cepolina, F., Razzoli, R.: Review of robotic surgery platforms and end effectors. Journal of Robotic Surgery18(1), 74 (2024)

work page 2024

[4] [3]

In: International Workshop on Trustworthy Machine Learning for Healthcare

Cheng, Y., Liu, L., Wang, S., Jin, Y., Schönlieb, C.B., Aviles-Rivero, A.I.: Why deep surgical models fail?: Revisiting surgical action triplet recognition through the lens of robustness. In: International Workshop on Trustworthy Machine Learning for Healthcare. pp. 177–189. Springer (2023)

work page 2023

[5] [5]

In: Proceedings of the IEEE/CVF international conference on computer vision

Feichtenhofer, C., Fan, H., Malik, J., He, K.: Slowfast networks for video recog- nition. In: Proceedings of the IEEE/CVF international conference on computer vision. pp. 6202–6211 (2019)

2019

[6] [6]

Scientific Data11(1), 373 (2024)

Ghamsarian, N., El-Shabrawi, Y., Nasirihaghighi, S., Putzgruber-Adamitsch, D., Zinkernagel, M., Wolf, S., Schoeffmann, K., Sznitman, R.: Cataract-1k dataset for deep-learning-assisted analysis of cataract surgery videos. Scientific Data11(1), 373 (2024)

work page 2024

[7] [7]

Eye35(12), 3192–3201 (2021)

He, B., de Smet, M.D., Sodhi, M., Etminan, M., Maberley, D.: A review of robotic surgical training: establishing a curriculum and credentialing process in ophthal- mology. Eye35(12), 3192–3201 (2021)

work page 2021

[8] [8]

Hessel, J., Holtzman, A., Forbes, M., Bras, R.L., Choi, Y.: Clipscore: A reference- freeevaluationmetricforimagecaptioning.arXivpreprintarXiv:2104.08718(2021)

arXiv 2021