Pith. sign in

REVIEW 3 major objections 4 minor 49 references

VidForensics-M1: Meta-Detection Reinforcement Learning with Verifiable Temporal Grounding for AI-Generated Video Forensics

T0 review · 3 major / 4 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 AI-generated video detection improves sharply when reinforcement learning rewards not just the predicted label but the exact manipulated time interval, and this rule-based temporal evidence beats model-written text rationales as training…

desk verdict The EGRR reward-redistribution theory is clean and genuinely useful, but the paper's central empirical claim that temporal grounding beats textual rationales is undermined by a training-set confound in Section 5.1. read the letter →

arxiv 2608.11201 v1 pith:V23WHSRY submitted 2026-08-11 cs.CV

Bowei Liu , Zheng Lu , Yuhan Bian , Xinchen Zhang , Xingming Shui , Yuesheng Huang , Xuhuan Li , Zihao Liu
show 3 more authors
Yifan Yang Jun Zhou Xiu Li
This is my paper · ORCID
classification cs.CV
keywords AI-generatedvideodetectionforensicsmeta-detectionevidence-guidedrewardredistributiontemporalgroundingreinforcementlearningmultimodallargelanguagemodelsGRPO
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 argues that the main obstacle to detecting AI-generated video is the supervision signal, not model size: label-only rewards let a detector exploit shortcuts, and text explanations of artifacts inherit the biases of the model that wrote them. Its proposal, VidForensics-M1, is a "meta-detection" training loop that rewards the predicted label and the supporting evidence together, using the manipulated time interval as verifiable evidence. Training fakes are built by cutting a random segment out of a real clip, regenerating that segment from its boundary frames, and splicing it back, so the forged interval is known exactly by construction. The paper reports that training with temporal grounding lifts accuracy, recall, and F1 by roughly ten points over label-only reinforcement learning on ViF-Bench and raises recall on fully generated videos in GenBuster-Bench, indicating that evidence-based feedback carries over to unseen generators.

What carries the argument

Two coupled mechanisms carry the argument. First, the automated data construction pipeline removes a random temporal segment from each real 5-second video, regenerates the missing content from its boundary frames with video generation models, aligns duration, resolution, and frame rate, and reinserts it, producing paired real-fake videos whose manipulated interval is ground truth by construction. Second, Evidence-Guided Reward Redistribution (EGRR) modifies the reinforcement-learning reward so that among label-correct responses for a fake video, reward flows from responses whose predicted interval has below-group-average temporal IoU to those with above-average IoU, via the additive term $\ell_{ij}(e_{ij}-q_i)$ in the semantic reward, where $\ell_{ij}$ is label correctness, $e_{ij}$ is the IoU between predicted and true forged intervals, and $q_i$ is the mean evidence quality among label-correct responses in the sampled group. The zero-sum identity $\sum_j \ell_{ij}(e_{ij}-q_i)=0$ guarantees the label-reward mass is conserved exactly, so EGRR only reallocates credit within the correct-label subset; when all sampled evidence has equal quality it degenerates to plain label-level RL. A theorem shows that the expected evidence gradient is an ascent direction for evidence quality conditioned on a correct label, so the scheme provably teaches localization without disturbing label learning.

What would settle it

Evaluate the detector on fully AI-generated videos that contain no spliced real footage and hence no boundary seam or content-mismatch cue, and compare recall against the same model trained with label-only reinforcement learning; if the roughly ten-point advantage disappears, the temporal evidence was teaching construction fingerprints rather than generalizable forgery detection. A companion check: fully generated videos have no ground-truth manipulated interval, so the detector's "verifiable" localization cannot be scored by $\mathrm{IoU}$ — asking humans whether the flagged seconds contain visible artifacts, and testing whether the detector fails on segment-replacement fakes whose regenerated segment is unusually well matched to its boundaries, would settle whether temporal grounding is genuinely the source of the gain.

Watch

Extended reading notes

Core claim

The central claim is that rule-based temporal grounding is a more reliable and verifiable meta-detection feedback signal than model-based textual rationales. Detectors trained with the Evidence-Guided Reward Redistribution (EGRR) scheme, in which reward among label-correct responses is shifted toward those whose predicted forged interval overlaps the ground-truth interval (temporal $\mathrm{IoU} \geq 0.7$), outperform label-only reinforcement learning by 10.30% accuracy, 9.74% recall, and 11.21% F1 on ViF-Bench, and outperform text-rationale supervision by 6.25%, 6.38%, and 6.96%; on the fake-only subset of GenBuster-Bench, recall improves by 10.1% (OOD) and 19.2% (Wild) over label-only RL. The paper attributes the difference to objectivity: temporal intervals are recorded by the construction process as ground truth, whereas textual rationales are generated by a reference model and verified by another model, so they remain vulnerable to hallucination, semantic bias, and reward hacking. A side observation is that temporal grounding also teaches the detector to localize the forged segment, and its free-form explanations become more accurate as well.

Load-bearing premise

The load-bearing premise is that a fake video made by deleting a random segment of a real clip and regenerating that segment from its boundary frames is representative of the AI-generated videos the detector is tested on, including fully generated clips from Sora-2, Kling-V1, and others; the paper offers no analysis of how much of its reported gain depends on splice seams and construction artifacts unique to its own pipeline.

Editorial extensions

If this is right

  • Detectors trained with temporal-grounding reward generalize to generators unseen in training: the gains on ViF-Bench are spread across 18 models including Sora-2 and Kling-V1, and GenBuster-Bench fake-only recall rises 10.1% on OOD and 19.2% on Wild data over label-only RL.
  • The same evidence signal teaches temporal localization itself: interval-trained detectors output forged intervals that overlap ground truth (tIoU 0.81 in the reported example), whereas text-trained detectors report the whole clip as suspicious.
  • Evidence-guided training can be layered onto label-level RL without changing the label objective: EGRR conserves label reward mass exactly and reduces to plain label-level RL whenever sampled evidence carries no discriminative information.
  • Model-written textual rationales are the weaker evidence source on every reported metric, consistent with the claim that their generation and verification depend on external models and are prone to hallucination and semantic bias.

Reading between the lines

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

  • If the central claim is right, the cheapest lever for improving detector generalization is controlling how training forgeries are made so that ground-truth evidence is mechanically extractable, rather than scaling models or accumulating more generators; the same construct-then-verify recipe should transfer to image splicing and audio deepfakes.
  • An implicit corollary the paper does not test: the gap between temporal grounding and text rationales should widen as generators improve, because fixed artifact categories age out while interval-localization pressure keeps forcing evidence-based decisions.
  • A clean check of the causal story would train on fully generated fake videos (no real context, no splice seams) while rewarding localization through a rule-based proxy, for instance consistency between two independently sampled generations, and see whether the accuracy gain persists; the paper's construction leaves this variant untested.
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.

Referee Report

3 major / 4 minor

Summary. The paper introduces VidForensics-M1, a reinforcement-learning framework for AI-generated video detection that jointly optimizes label correctness and supporting evidence. It proposes an automated data construction pipeline that creates paired real/fake videos by removing a random temporal segment from a real video, reconstructing the missing content with boundary-frame-conditioned generative models, and reinserting the aligned segment. This process yields ground-truth manipulated temporal intervals. The authors also propose Evidence-Guided Reward Redistribution (EGRR), which redistributes the label-level reward among label-correct responses according to evidence quality while conserving total reward mass; the theoretical analysis in Appendix A proves conservation, correctness dominance, orthogonality of label and evidence signals, and an expected evidence-improving gradient. The central empirical claim is that rule-based temporal grounding provides more reliable and verifiable feedback than model-based textual rationales for meta-detection, supported by comparisons of Label-Tem-RL versus Label-Exp-RL on ViF-Bench and GenBuster-Bench.

Significance. If the central claim holds, the paper contributes a practical mechanism for incorporating verifiable, rule-based evidence into RL training of video forensics detectors, with potential to improve generalization to unseen generators. The proposed data construction pipeline is a valuable and reusable resource, and the EGRR theoretical analysis is mathematically sound and clearly presented. However, the main empirical comparison is currently confounded by unequal training data between the temporal and textual branches, and the generalization claim rests on an unexamined distribution shift between constructed segment-replacement fakes and fully generated test videos. These issues must be resolved before the paper's central claim can be accepted.

major comments (3)
  1. [§5.1, Tables 1 and 2] The comparison between Label-Tem-RL and Label-Exp-RL is confounded by the Gemini-3.1-Pro filtering step applied only to the textual branch. The paper states that Gemini-3.1-Pro is used to "filter out samples with unreliable textual explanations and remove their corresponding real videos to maintain a 1:1 ratio" (Section 5.1), so Label-Exp-RL is trained on a strict subset of the 100K balanced set available to Label-Tem-RL. The only equalization mentioned is binarizing temporal IoU at 0.7 to match the binary textual consistency score; no effort is made to equalize the training distribution. The headline gains in Table 1 (+6.25% accuracy, +6.96% F1) and the GenBuster-Bench gains in Table 2 (+6.4% OOD, +11.4% Wild recall) could therefore be attributable to more training data rather than to the superiority of temporal grounding as an evidence source. To support the central claim, the paper must train Label-Exp-RL on an equal-size random subset of the retained samples or otherwise control for data volume.
  2. [§5.2, Tables 1 and 2] All reported metrics are point estimates without error bars, confidence intervals, or significance tests. MLLM training and evaluation are known to be noisy, and the claimed improvements over Label-Exp-RL (e.g., 6.25% accuracy and 6.96% F1 on ViF-Bench) are small enough that they could fall within run-to-run variability. The paper should report results from multiple seeds with means and standard deviations, or at least provide a significance test, to substantiate the empirical claims.
  3. [§3.3 and §5.2] The training data for VidForensics-M1 consists of segment-replacement fakes created by removing a random interval from a real video and reinserting a generated segment, while the evaluation benchmarks include fully generated videos from models such as Sora-2 and Kling-V1. The paper provides no analysis of the distribution shift between these two forgery types, and no ablation to determine whether the detector relies on boundary seams, content inconsistency, or other artifacts unique to the construction process. Without such analysis, the reported GenBuster-Bench gains (Table 2: +10.1% OOD and +19.2% Wild recall for Label-Tem-RL over Label-RL) may not reflect genuine generalization to unseen generative models. The authors should include experiments that vary the forged-segment ratio, add post-processing to remove seams, or evaluate on an additional fully generated held-out set to verify that the learned signal transfers.
minor comments (4)
  1. [Figure 2] The figure caption contains typographical errors: "start tmie end tmie" should read "start time end time".
  2. [Table 1] The row label "+ Label-Tem-RL" appears after the numeric values in the row, which is confusing; the label should precede the metrics for clarity, as in the other method rows.
  3. [References] Reference [37] appears to be a duplicate of reference [11]; please consolidate or distinguish them.
  4. [Equation (2)] The notation for interval intersection and union in the IoU definition is not explicitly defined; please state that ∩ and ∪ denote interval intersection and union, respectively.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: EGRR conservation is an algebraic identity, temporal ground truth is externally constructed, and no prediction is equivalent to a fitted input.

full rationale

The central derivations do not reduce to their inputs. In EGRR, qi is defined as the empirical mean of evidence quality among label-correct responses (Eq. 3), so the zero-sum identity of Lemma 1 (Eq. 5) and reward-mass conservation of Corollary 1 are exact algebraic consequences of the definition; this is a transparent design property, not a prediction smuggled from a fitted parameter. Temporal-grounding supervision comes from an external rule-based construction process: a segment is removed, regenerated from boundary frames, aligned, and reinserted, with the manipulated interval recorded automatically. The model's predicted interval is compared against this externally recorded ground truth through IoU, so the reward depends on an independent rule, not on the model's own output. The temporal-versus-textual comparison is an experimental evaluation on independent benchmarks; although Section 5.1 indicates that Gemini filtering removes some textual-training samples while temporal training uses the full 100K set, that is a potential experimental confound about data volume, not a circular derivation. The self-citations [33-36] appear in a related-work sentence about advances in generative models and are not used to justify the meta-detection or EGRR claims. Lemma 2's uniqueness result is proven in the paper from the conservation constraint rather than imported from prior work. No fitted parameter is relabeled as a prediction, and no equation equates an output to an input by construction. Therefore the circularity score is 0.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

No new physical entities, particles, forces, or conserved quantities are postulated. 'Meta-detection' is a re-branding of evidence-aware RL, not an invented entity. The free parameters are standard training thresholds and hyperparameters; none is fitted to the test benchmarks.

free parameters (4)
  • IoU threshold 0.7 = 0.7
    Converts continuous temporal IoU into a binary evidence-correctness signal; chosen to match the binary textual evidence evaluation and affects which responses receive redistributed rewards.
  • lambda_format = not reported
    Weights the format reward versus the semantic reward in Eqs. (1) and (4); the value is not specified in the paper.
  • DAPO learning rate = 1e-6
    Optimizer hyperparameter used for one training epoch; no sensitivity analysis is provided.
  • Video preprocessing = 4 FPS sampling, 5-second clips
    Sampling rate and clip length are design choices that affect both the data construction and the detector input.
assumptions (4)
  • domain assumption GRPO/DAPO policy-gradient optimization with rule-based rewards converges to a useful optimum within one training epoch.
    Training relies on DAPO with learning rate 1e-6 for one epoch (Section 5.1); no convergence analysis or training curves are shown.
  • domain assumption The constructed segment-replacement fake videos share the same content and temporal context as the original real videos, and the ground-truth manipulated interval is a complete and unbiased evidence signal.
    Section 3.3 asserts this as an advantage of the pipeline; it underpins the entire temporal supervision signal.
  • standard math Interchanging expectation and differentiation in Theorem 3 is valid for the policy-gradient estimator, and rule-based rewards are treated as stop-gradient quantities.
    Assumed in Appendix A.3 before Eq. (46); standard in RLVR but not formally justified for the non-differentiable JSON output parsing.
  • domain assumption Textual explanations generated by GPT-5.5 and filtered by Gemini-3.1-Pro constitute a fair baseline for model-based evidence.
    The fairness of the temporal versus textual comparison depends on this filtering not materially changing the sample distribution (Section 5.1).

how reviews work

0 comments
Cite this review

Pith. "Pith review of VidForensics-M1: Meta-Detection Reinforcement Learning with Verifiable Temporal Grounding for AI-Generated Video Forensics." pith.science (2026). https://pith.science/paper/V23WHSRY

@misc{pith2026260811201,
  author       = {Pith},
  title        = {Pith review of: VidForensics-M1: Meta-Detection Reinforcement Learning with Verifiable Temporal Grounding for AI-Generated Video Forensics},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/V23WHSRY}},
  note         = {Machine review of arXiv:2608.11201}
}
read the original abstract

Recent advances in video generation models have significantly improved the realism of synthetic videos, blurring the boundary between generated and authentic content and raising concerns about misinformation. Existing MLLM-based detectors mainly rely on supervised fine-tuning or label-level reinforcement learning, where coarse supervision limits generalization to unseen scenarios and emerging video generators. To overcome these limitations, we are the first to introduce \textbf{meta-detection} into AI-generated video detection, enabling reliable forgery detection by jointly optimizing predicted labels and supporting evidence within reinforcement learning. This paradigm requires reliable evidence signals and effective mechanisms to integrate them into label-level optimization. Textual rationales provide semantic descriptions of forgery artifacts, but their generation and verification depend on external models, making supervision vulnerable to hallucinations and semantic biases. In contrast, temporal grounding provides more objective and verifiable evidence, as manipulated intervals can be precisely controlled during forgery construction. Based on this insight, we propose an automated data construction pipeline that generates paired real-fake videos by replacing temporal segments with boundary-frame-conditioned video generation models. Furthermore, we introduce \textbf{Evidence-Guided Reward Redistribution}, which performs evidence-aware credit assignment by redistributing rewards among label-correct responses according to evidence quality. This preserves reliable label supervision while encouraging detectors to acquire fine-grained and verifiable forgery localization capabilities. Extensive experiments demonstrate that \textbf{VidForensics-M1} effectively leverages verifiable temporal evidence to achieve robust and generalizable AI-generated video detection.

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

49 extracted references · 25 canonical work pages

  1. [1]

    Detecting ai-generated video: A vision–language dual-view survey

    Dylan Xinming Hou, Juntian Zhang, Xu Gu, Yichen Wu, Nils Lukas, Gus Xia, Xiuying Chen, and Yuhan Liu. Detecting ai-generated video: A vision–language dual-view survey. In Findings of the Association for Computational Linguistics: ACL 2026, pages 32221–32255, 2026

    work page 2026

  2. [2]

    Cubecomposer: Spatio-temporal autoregressive 4k 360°video generation from perspective video.arXiv e-prints, pages arXiv–2603, 2026

    Lingen Li, Guangzhi Wang, Xiaoyu Li, Zhaoyang Zhang, Qi Dou, Jinwei Gu, Tianfan Xue, and Ying Shan. Cubecomposer: Spatio-temporal autoregressive 4k 360°video generation from perspective video.arXiv e-prints, pages arXiv–2603, 2026

    work page 2026

  3. [3]

    Evatok: Adaptive length video tokenization for efficient visual autoregressive generation.arXiv preprint arXiv:2603.12267, 2026

    Tianwei Xiong, Jun Hao Liew, Zilong Huang, Zhijie Lin, Jiashi Feng, and Xihui Liu. Evatok: Adaptive length video tokenization for efficient visual autoregressive generation.arXiv preprint arXiv:2603.12267, 2026

    arXiv 2026

  4. [4]

    Smrabooth: Subject and motion representation alignment for customized video generation.arXiv preprint arXiv:2512.12193, 2025

    Xuancheng Xu, Yaning Li, Sisi You, and Bing-Kun Bao. Smrabooth: Subject and motion representation alignment for customized video generation.arXiv preprint arXiv:2512.12193, 2025

    arXiv 2025

  5. [5]

    Hunyuanportrait: Implicit condition control for enhanced portrait animation

    Zunnan Xu, Zhentao Yu, Zixiang Zhou, Jun Zhou, Xiaoyu Jin, Fa-Ting Hong, Xiaozhong Ji, Junwei Zhu, Chengfei Cai, Shiyu Tang, et al. Hunyuanportrait: Implicit condition control for enhanced portrait animation. In2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 15909–15919. IEEE, 2025

    work page 2025

  6. [6]

    Zo3t: Zero-shot 3d-aware trajectory-guided image-to-video generation via test- time training

    Ruicheng Zhang, Jun Zhou, Zunnan Xu, Zihao Liu, Jiehui Huang, Mingyang Zhang, Yu Sun, and Xiu Li. Zo3t: Zero-shot 3d-aware trajectory-guided image-to-video generation via test- time training. InProceedings of the AAAI Conference on Artificial Intelligence, volume 40, pages 12708–12716, 2026

    work page 2026

  7. [7]

    Perceptual judgments of video authenticity: An examination of viewing duration, confidence, content, and strategies.Law Review, 107:1753–1819, 2026

    Catherine E Davodi, Sarah Barrington, Hany Farid, and Emily A Cooper. Perceptual judgments of video authenticity: An examination of viewing duration, confidence, content, and strategies.Law Review, 107:1753–1819, 2026. 10

    work page 2026

  8. [8]

    Deepfakes and disinformation: Exploring the impact of synthetic political video on deception, uncertainty, and trust in news.Social media+ society, 6(1):2056305120903408, 2020

    Cristian Vaccari and Andrew Chadwick. Deepfakes and disinformation: Exploring the impact of synthetic political video on deception, uncertainty, and trust in news.Social media+ society, 6(1):2056305120903408, 2020

    2020

Show all 49 references
  1. [9]

    Reducing risks posed by synthetic content an overview of technical approaches to digital content transparency

    Bilva Chandra, Jesse Dunietz, Kathleen Roberts, Yooyoung Lee, Peter Fontana, and George Awad. Reducing risks posed by synthetic content an overview of technical approaches to digital content transparency. 2024

    2024

  2. [10]

    David-xr1: Detecting ai-generated videos with explainable reasoning.arXiv preprint arXiv:2506.14827, 2025

    Yifeng Gao, Yifan Ding, Hongyu Su, Juncheng Li, Yunhan Zhao, Lin Luo, Zixing Chen, Li Wang, Xin Wang, Yixu Wang, et al. David-xr1: Detecting ai-generated videos with explainable reasoning.arXiv preprint arXiv:2506.14827, 2025

    2025 arXiv

  3. [12]

    Skyra: Ai-generated video detection via grounded artifact reasoning.arXiv preprint arXiv:2512.15693, 2025

    Yifei Li, Wenzhao Zheng, Yanran Zhang, Runze Sun, Yu Zheng, Lei Chen, Jie Zhou, and Jiwen Lu. Skyra: Ai-generated video detection via grounded artifact reasoning.arXiv preprint arXiv:2512.15693, 2025

    2025 arXiv

  4. [13]

    Videoveritas: Ai-generated video detection via perception pretext reinforcement learning

    Hao Tan, Senyuan Shi, Zichang Tan, Zijian Yu, Huijia Zhu, Weiqiang Wang, Jun Wan, Zhen Lei, et al. Videoveritas: Ai-generated video detection via perception pretext reinforcement learning. InForty-third International Conference on Machine Learning, 2026

    2026

  5. [14]

    Seeing what matters: Generalizable ai-generated video detection with forensic-oriented augmentation.arXiv preprint arXiv:2506.16802, 2025

    Riccardo Corvi, Davide Cozzolino, Ekta Prashnani, Shalini De Mello, Koki Nagano, and Luisa Verdoliva. Seeing what matters: Generalizable ai-generated video detection with forensic-oriented augmentation.arXiv preprint arXiv:2506.16802, 2025

    2025

  6. [15]

    Chain-of-thought prompting elicits reason- ing in large language models

    Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, brian ichter, Fei Xia, Ed Chi, Quoc V Le, and Denny Zhou. Chain-of-thought prompting elicits reason- ing in large language models. In S. Koyejo, S. Mohamed, A. Agarwal, D. Bel- grave, K. Cho, and A. Oh, editors,Advances i...

    2022

  7. [16]

    Vidguard-r1: Ai-generated video detection and explanation via reasoning mllms and rl.arXiv preprint arXiv:2510.02282, 2025

    Kyoungjun Park, Yifan Yang, Juheon Yi, Shicheng Zheng, Yifei Shen, Dongqi Han, Caihua Shan, Muhammad Muaz, and Lili Qiu. Vidguard-r1: Ai-generated video detection and explanation via reasoning mllms and rl.arXiv preprint arXiv:2510.02282, 2025

  8. [17]

    Busterx++: Towards unified cross-modal ai-generated content detection and explanation with mllm.arXiv preprint arXiv:2507.14632, 2025

    Haiquan Wen, Tianxiao Li, Zhenglin Huang, Yiwei He, and Guangliang Cheng. Busterx++: Towards unified cross-modal ai-generated content detection and explanation with mllm.arXiv preprint arXiv:2507.14632, 2025

    2025 arXiv

  9. [18]

    Christiano, Jan Leike, Tom B

    Paul F. Christiano, Jan Leike, Tom B. Brown, Miljan Martic, Shane Legg, and Dario Amodei. Deep reinforcement learning from human preferences. InProceedings of the 31st International Conference on Neural Information Processing Systems, NIPS’17, page 4302–4310, Red Hook, NY, USA...

    2017

  10. [19]

    Concrete problems in ai safety, 2016

    Dario Amodei, Chris Olah, Jacob Steinhardt, Paul Christiano, John Schulman, and Dan Mané. Concrete problems in ai safety, 2016. URLhttps://arxiv.org/abs/1606.06565

    2016 arXiv

  11. [20]

    Daya Guo, Dejian Yang, Haowei Zhang, Junxiao Song, Peiyi Wang, Qihao Zhu, Runxin Xu, Ruoyu Zhang, Shirong Ma, Xiao Bi, Xiaokang Zhang, Xingkai Yu, Yu Wu, Z. F. Wu, Zhibin Gou, Zhihong Shao, Zhuoshu Li, Ziyi Gao, Aixin Liu, Bing Xue, Bingxuan Wang, Bochao Wu, 11 Bei Feng, Cheng...

    2025

  12. [21]

    Language models don’t always say what they think: Unfaithful explanations in chain-of-thought prompting.Advances in Neural Information Processing Systems, 36:74952–74965, 2023

    Miles Turpin, Julian Michael, Ethan Perez, and Samuel Bowman. Language models don’t always say what they think: Unfaithful explanations in chain-of-thought prompting.Advances in Neural Information Processing Systems, 36:74952–74965, 2023

    2023

  13. [22]

    Forgerynet: A versatile benchmark for comprehensive forgery analysis

    Yinan He, Bei Gan, Siyu Chen, Yichun Zhou, Guojun Yin, Luchuan Song, Lu Sheng, Jing Shao, and Ziwei Liu. Forgerynet: A versatile benchmark for comprehensive forgery analysis. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 4360–4369, 2021

    2021

  14. [23]

    Grounded-VideoLLM: Sharpening fine-grained temporal ground- ing in video large language models

    Haibo Wang, Zhiyang Xu, Yu Cheng, Shizhe Diao, Yufan Zhou, Yixin Cao, Qifan Wang, Weifeng Ge, and Lifu Huang. Grounded-VideoLLM: Sharpening fine-grained temporal ground- ing in video large language models. In Christos Christodoulopoulos, Tanmoy Chakraborty, Carolyn Rose, and V...

    2025 doi

  15. [24]

    Cnn- generated images are surprisingly easy to spot

    Sheng-Yu Wang, Oliver Wang, Richard Zhang, Andrew Owens, and Alexei A Efros. Cnn- generated images are surprisingly easy to spot... for now. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 8695–8704, 2020

    2020

  16. [25]

    Towards universal fake image detectors that generalize across generative models

    Utkarsh Ojha, Yuheng Li, and Yong Jae Lee. Towards universal fake image detectors that generalize across generative models. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 24480–24489, 2023. 12

    2023

  17. [26]

    Transcending forgery specificity with latent space augmentation for generalizable deepfake detection

    Zhiyuan Yan, Yuhao Luo, Siwei Lyu, Qingshan Liu, and Baoyuan Wu. Transcending forgery specificity with latent space augmentation for generalizable deepfake detection. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 8984–8994, 2024

    2024

  18. [27]

    Rethinking the up-sampling operations in cnn-based generative network for generalizable deepfake detection

    Chuangchuang Tan, Yao Zhao, Shikui Wei, Guanghua Gu, Ping Liu, and Yunchao Wei. Rethinking the up-sampling operations in cnn-based generative network for generalizable deepfake detection. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, page...

    2024

  19. [28]

    Laa-net: Localized artifact attention network for quality-agnostic and generalizable deepfake detection

    Dat Nguyen, Nesryne Mejri, Inder Pal Singh, Polina Kuleshova, Marcella Astrid, Anis Kacem, Enjie Ghorbel, and Djamila Aouada. Laa-net: Localized artifact attention network for quality-agnostic and generalizable deepfake detection. InProceedings of the IEEE/CVF Conference on Co...

    2024

  20. [29]

    Exploring unbiased deepfake detection via token-level shuffling and mixing

    Xinghe Fu, Zhiyuan Yan, Taiping Yao, Shen Chen, and Xi Li. Exploring unbiased deepfake detection via token-level shuffling and mixing. InProceedings of the AAAI Conference on Artificial Intelligence, volume 39, pages 3040–3048, 2025

    2025

  21. [30]

    A sanity check for ai-generated image detection

    Shilin Yan, Ouxiang Li, Jiayin Cai, Yanbin Hao, Xiaolong Jiang, Yao Hu, and Weidi Xie. A sanity check for ai-generated image detection. InInternational Conference on Learning Representations, volume 2025, pages 70702–70720, 2025

    2025

  22. [31]

    All patches matter, more patches better: Enhance ai-generated image detection via panoptic patch learning.arXiv preprint arXiv:2504.01396, 2025

    Zheng Yang, Ruoxin Chen, Zhiyuan Yan, Ke-Yue Zhang, Xinghe Fu, Shuang Wu, Xiujun Shu, Taiping Yao, Shouhong Ding, Zequn Qin, et al. All patches matter, more patches better: Enhance ai-generated image detection via panoptic patch learning.arXiv preprint arXiv:2504.01396, 2025

    2025

  23. [32]

    Busterx: Mllm-powered ai-generated video forgery detection and explanation.arXiv preprint arXiv:2505.12620, 2025

    Haiquan Wen, Yiwei He, Zhenglin Huang, Tianxiao Li, Zihan Yu, Xingru Huang, Lu Qi, Baoyuan Wu, Xiangtai Li, and Guangliang Cheng. Busterx: Mllm-powered ai-generated video forgery detection and explanation.arXiv preprint arXiv:2505.12620, 2025

    2025 arXiv

  24. [33]

    Omniverifier-m1: Multimodal meta-verifier with explicit structured recalibration.arXiv preprint arXiv:2605.28805, 2026

    Xinchen Zhang, Bowei Liu, Jiale Liu, Chufan Shi, Yizhen Zhang, Junhong Liu, Youliang Zhang, Zhiheng Li, Yujiu Yang, and Ling Yang. Omniverifier-m1: Multimodal meta-verifier with explicit structured recalibration.arXiv preprint arXiv:2605.28805, 2026

    2026 arXiv

  25. [34]

    Generative universal verifier as multimodal meta-reasoner

    Xinchen Zhang, Xiaoying Zhang, Youbin Wu, Yanbin Cao, Renrui Zhang, Ruihang Chu, Ling Yang, Yujiu Yang, and Guang Shi. Generative universal verifier as multimodal meta-reasoner. InInternational Conference on Learning Representations, volume 2026, pages 109211–109243, 2026

    2026

  26. [35]

    Realcompo: Balancing realism and compositionality improves text-to-image diffusion models.Advances in Neural Information Processing Systems, 37:96963–96992, 2024

    Xinchen Zhang, Ling Yang, Yaqi Cai, Zhaochen Yu, Kai-Ni Wang, Jiake Xie, Ye Tian, Minkai Xu, Yong Tang, Yujiu Yang, et al. Realcompo: Balancing realism and compositionality improves text-to-image diffusion models.Advances in Neural Information Processing Systems, 37:96963–96992, 2024

    2024

  27. [36]

    Itercomp: Iterative composition-aware feedback learning from model gallery for text-to-image generation

    Xinchen Zhang, Ling Yang, Guohao Li, Yaqi Cai, Yong Tang, Yujiu Yang, Mengdi Wang, Bin CUI, et al. Itercomp: Iterative composition-aware feedback learning from model gallery for text-to-image generation. InInternational Conference on Learning Representations, volume 2025, page...

    2025

  28. [37]

    Learning human-perceived fakeness in ai-generated videos via multimodal llms.arXiv preprint arXiv:2509.22646, 2025

    Xingyu Fu, Siyi Liu, Yinuo Xu, Pan Lu, Guangqiuse Hu, Tianbo Yang, Taran Anantasagar, Christopher Shen, Yikai Mao, Yuanzhe Liu, et al. Learning human-perceived fakeness in ai-generated videos via multimodal llms.arXiv preprint arXiv:2509.22646, 2025. 13

    2025

  29. [38]

    Ai-generated video detection via spatial- temporal anomaly learning

    Jianfa Bai, Man Lin, Gang Cao, and Zijie Lou. Ai-generated video detection via spatial- temporal anomaly learning. InChinese Conference on Pattern Recognition and Computer Vision (PRCV), pages 460–470. Springer, 2024

    2024

  30. [39]

    Demamba: Ai-generated video detection on million-scale genvideo benchmark.arXiv preprint arXiv:2405.19707, 2024

    Haoxing Chen, Yan Hong, Zizheng Huang, Zhuoer Xu, Zhangxuan Gu, Yaohui Li, Jun Lan, Huijia Zhu, Jianfu Zhang, Weiqiang Wang, et al. Demamba: Ai-generated video detection on million-scale genvideo benchmark.arXiv preprint arXiv:2405.19707, 2024

    2024 arXiv

  31. [40]

    Detecting ai-generated video via frame consistency

    Long Ma, Zhiyuan Yan, Qinglang Guo, Yong Liao, Haiyang Yu, and Pengyuan Zhou. Detecting ai-generated video via frame consistency. In2025 IEEE International Conference on Multimedia and Expo (ICME), pages 1–6. IEEE, 2025

    2025

  32. [41]

    Genvidbench: A 6-million benchmark for ai-generated video detection

    Zhenliang Ni, Qiangyu Yan, Mouxiao Huang, Tianning Yuan, Yehui Tang, Hailin Hu, Xinghao Chen, and Yunhe Wang. Genvidbench: A 6-million benchmark for ai-generated video detection. InProceedings of the AAAI Conference on Artificial Intelligence, volume 40, pages 15582–15590, 2026

    2026

  33. [42]

    Qwen3.5: Towards native multimodal agents.https://qwen.ai/blog?id= qwen3.5, February 2026

    Qwen Team. Qwen3.5: Towards native multimodal agents.https://qwen.ai/blog?id= qwen3.5, February 2026

    2026

  34. [43]

    Ltx-video: Realtime video latent diffusion.arXiv preprint arXiv:2501.00103, 2024

    Yoav HaCohen, Nisan Chiprut, Benny Brazowski, Daniel Shalem, Dudu Moshe, Eitan Richardson, Eran Levin, Guy Shiran, Nir Zabari, Ori Gordon, Poriya Panet, Sapir Weissbuch, Victor Kulikov, Yaki Bitterman, Zeev Melumian, and Ofir Bibi. Ltx-video: Realtime video latent diffusion.ar...

    2024 arXiv

  35. [44]

    Wan: Open and advanced large-scale video generative models.arXiv preprint arXiv:2503.20314, 2025

    Team Wan, Ang Wang, Baole Ai, Bin Wen, Chaojie Mao, Chen-Wei Xie, Di Chen, Feiwu Yu, Haiming Zhao, Jianxiao Yang, Jianyuan Zeng, Jiayu Wang, Jingfeng Zhang, Jingren Zhou, Jinkai Wang, Jixuan Chen, Kai Zhu, Kang Zhao, Keyu Yan, Lianghua Huang, Mengyang Feng, Ningyi Zhang, Pande...

    2025 arXiv

  36. [45]

    Skyreels-v2: Infinite-length film generative model, 2025

    Guibin Chen, Dixuan Lin, Jiangping Yang, Chunze Lin, Junchen Zhu, Mingyuan Fan, Hao Zhang, Sheng Chen, Zheng Chen, Chengcheng Ma, Weiming Xiong, Wei Wang, Nuo Pang, Kang Kang, Zhiheng Xu, Yuzhe Jin, Yupeng Liang, Yubing Song, Peng Zhao, Boyuan Xu, Di Qiu, Debang Li, Zhengcong ...

    2025 arXiv

  37. [46]

    Wan 2.7: Image-to-video api.https://www.alibabacloud.com/help/en/ model-studio/image-to-video-general-api-reference, 2026

    Alibaba Cloud. Wan 2.7: Image-to-video api.https://www.alibabacloud.com/help/en/ model-studio/image-to-video-general-api-reference, 2026

    2026

  38. [47]

    Seedance 1.0 Pro

    BytePlus. Seedance 1.0 Pro. https://docs.byteplus.com/en/docs/ModelArk/1587798, 2025

    2025

  39. [48]

    Introducing gpt-5.https://openai.com/index/introducing-gpt-5/, 2025

    OpenAI. Introducing gpt-5.https://openai.com/index/introducing-gpt-5/, 2025

    2025

  40. [49]

    Gemini 3.1 Pro model card

    Google DeepMind. Gemini 3.1 Pro model card. https://deepmind.google/models/ model-cards/gemini-3-1-pro/, 2026. 14

    2026

  41. [50]

    Dapo: An open-source llm reinforcement learning system at scale.Advances in Neural Information Processing Systems, 38:113222– 113244, 2026

    Qiying Yu, Zheng Zhang, Ruofei Zhu, Yufeng Yuan, Xiaochen Zuo, Yu Yue, Weinan Dai, Tiantian Fan, Gaohong Liu, Lingjun Liu, et al. Dapo: An open-source llm reinforcement learning system at scale.Advances in Neural Information Processing Systems, 38:113222– 113244, 2026. 15 A Th...

    2026

Pith tools

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