Pith. sign in

REVIEW 3 major objections 8 minor 2 cited by

FakeSV-VLM: Taming VLM for Detecting Fake Short-Video News via Progressive Mixture-Of-Experts Adapter

T0 review · 3 major / 8 minor · reviewed 2026-08-05 · deepseek-v4-flash

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

Pith's one-line read Fine-tuning a VLM with two expert stages beats prior fake-video news detectors.

desk verdict Solid SOTA numbers and a useful recipe, but the two central mechanism claims (Attribution MoE inferring manipulation types, ADEC capturing cross-modal inconsistency) are contradicted by the paper's own equations; deserves referee attention but not acceptance as written. read the letter →

arxiv 2508.19639 v1 pith:UKWH6KIJ submitted 2025-08-27 cs.MM

Junxi Wang , Yaxiong Wang , Lechao Cheng , Zhun Zhong This is my paper
classification cs.MM
keywords fakenewsdetectionshortvideovision-languagemodelmixtureofexpertsartifacttokenscross-modalconsistencycontrastivelearningSV/FakeTT
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 sets out to show that a large vision-language model, instead of being used only for zero-shot answers or to pre-generate clues, can be fine-tuned end-to-end to detect fake news in short videos. It organizes the problem into four video-text scenarios—both real, only the text fake, only the video fake, both fake—and builds a Progressive Mixture-of-Experts Adapter (PMOE) so the model first judges real/fake and then attributes the manipulation type. A second module, Alignment-driven Event Checking (ADEC), aligns visual and textual event embeddings with a contrastive objective. On the FakeSV and FakeTT benchmarks, the method reports 90.22% and 89.30% accuracy, improving over the previous best approaches by 3.32 and 5.02 percentage points, and the paper documents cases where the modules overturn mistakes made by earlier models.

What carries the argument

The load-bearing machinery is the PMOE adapter on top of the fine-tuned VLM: learnable Artifact Tokens are concatenated with visual and text features, passed through early LLM layers, then routed through two sparse mixtures of experts—Detection MoE (real/fake) followed by Attribution MoE (real, visual-only fake, textual-only fake, both fake)—with authenticity probability guidance and artifact classification losses. The second piece is ADEC, which average-pools LLM-contextualized text and visual features and trains them with a symmetric contrastive loss so that matched real pairs sit close in embedding space. Together they give the VLM a manipulation-aware reasoning path plus a cross-modal co

What would settle it

Look at the gradient of L_ADEC for a fake-labeled sample: since I(v,t)=0 for every fake and cross-news pair, the per-sample gradient is identically zero, so no inconsistency signal reaches the model through this loss. An experiment that replaces the matched-only contrastive loss with one that also penalizes mismatched pairs—or that measures ADEC's gradient contribution separated by label—would settle whether the reported ablation gains come from detecting inconsistencies or merely from aligning real pairs.

Watch

Extended reading notes

Core claim

The paper argues that a pretrained vision-language model, fine-tuned end-to-end rather than used zero-shot or as a text generator, can detect fake short-video news better than dedicated baselines. It introduces two mechanisms: PMOE, which inserts learnable Artifact Tokens into early LLM layers and routes them through a real/fake Detection MoE and then a four-expert Attribution MoE (real, text-only fake, video-only fake, both fake), and ADEC, which uses contrastive learning to align pooled visual and textual event embeddings. On the FakeSV and FakeTT benchmarks, the resulting system reports 90.22% and 89.30% accuracy, surpassing the previous best methods by 3.32 and 5.02 points, with ablation

Load-bearing premise

The crucial assumption is that the ADEC module's contrastive loss teaches the model to notice mismatched video-text pairs; but in Eqs. 17-18 every unmatched or fake pair is assigned I(v,t)=0, so those pairs contribute exactly zero gradient and the module as written can only pull matched real pairs closer together, not push inconsistent pairs apart.

Editorial extensions

If this is right

  • Fine-tuned open-source VLMs can outperform proprietary zero-shot models on this task, so top results do not require closed commercial APIs.
  • Because PMOE and ADEC are not needed at inference, the accuracy gain does not add per-video compute or latency.
  • The four-scenario expert decomposition gives the model a way to report the manipulation type, not just the real/fake label.
  • The gains on both a Chinese and an English dataset suggest the recipe transfers across languages and platforms.

Reading between the lines

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

  • Editorial: Eqs. 17-18 assign zero weight to every fake and cross-news pair, so ADEC as written can only pull matched real pairs together; a variant with explicit negative pairs would clarify whether the ablation gain comes from alignment of real pairs or from separation of inconsistent ones.
  • The authors note that no fine-grained dataset exists to label which part of a video is fabricated, so the Attribution MoE's four-way diagnoses remain validated only on a 100-sample manual set; building such annotations would be a direct next step.
  • A testable extension is to apply the same progressive MoE adapter to out-of-context image-news pairs or audio-video mismatches, where event-level inconsistency is the dominant failure mode.
  • The reported margins are relative to specific prior methods; scaling beyond the 8B backbone used here may shift the gap, since the paper could not test larger VLMs under its compute budget.
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 / 8 minor

Summary. The paper proposes FakeSV-VLM, a Vision-Language Model framework for fake short-video news detection. The method inserts learnable Artifact Tokens into an InternVL2.5-8B backbone, feeds them through a two-stage Progressive Mixture-of-Experts Adapter (PMOE) consisting of a Detection MoE and an Attribution MoE, and adds an Alignment-driven Event Checking (ADEC) contrastive module. Training uses L_total = L_CE + L_PMOE + L_ADEC. On FakeSV and FakeTT, the paper reports Accuracy gains of +3.32% and +5.02% over the previous SOTA, with ablations attributing gains to PMOE and ADEC. The paper also reports a manual-annotation evaluation of the Attribution MoE's four-way manipulation-type predictions, and includes case studies and failure cases.

Significance. If the reported results and mechanism claims held, this would be a meaningful advance: it would establish a new SOTA on two standard fake short-video benchmarks and provide a reusable VLM-based architecture with fine-grained manipulation attribution. The manuscript's strengths include a broad comparison against 13 baselines, repeated runs, component ablations, code release, and a candid discussion of failure cases. However, the paper's central explanatory claims are not supported by the equations defining the losses. The Attribution MoE has no supervision or gradient path for its four-way routing, and the ADEC contrastive loss, as written, minimizes an objective over positive real pairs only and does not directly learn event-level inconsistency. These are load-bearing issues: the reported gains cannot, on the evidence in the manuscript, be attributed to the proposed mechanisms.

major comments (3)
  1. [§2.3, Eqs. (7)–(8), and Appendix C.5.2] The Attribution MoE is claimed to infer four manipulation types (real, visual forgery, textual forgery, full forgery), but no loss is defined on the four routing probabilities p_i^r, p_i^{fv}, p_i^{ft}, p_i^{fm}. Eq. (7) selects the expert by argmax and Eq. (8) applies only the selected expert. L_APG (Eq. 6) supervises the two Detection-MoE probabilities, and L_ACL (Eq. 11) supervises the attention-pooled binary output, not the attribution gate. Because argmax is piecewise constant, gradients from L_CE/L_ACL cannot train the Attribution-MoE router or the four expert networks through Eq. (8). Thus the four experts cannot specialize to the four forgery scenarios under the described training scheme. The 82% manual-annotation result in Table 7 is not derivable from the specified losses and routing; either the manuscript omits a differentiable or supervised routing mechanism, or the Attributi
  2. [§2.4, Eqs. (15)–(19)] The ADEC module is claimed to capture event-level cross-modal inconsistency, but the match label I(v,t) is defined as 1 only when the video and text belong to the same news and the news is real. For fake pairs and cross-news pairs, I(v,t)=0, so the corresponding term -I log s in Eqs. (17)–(18) is identically zero. Consequently, the loss provides no positive gradient signal that directly separates inconsistent video-text pairs; fake items enter only as negatives in the denominators of real-anchor contrastive terms. This does not amount to learning 'inconsistency' in the claimed sense. The ablation gains credited to ADEC in Table 2 therefore cannot be attributed to the stated event-level inconsistency-checking mechanism. The authors should either reformulate the contrastive loss (e.g., with an explicit mismatch class or with I=0 paired with a push-away term) or revise the claim.
  3. [§3.3, Table 2 and overall interpretation] Because both central mechanisms are not realized as written, the component ablation in Table 2 does not establish that the PMOE/ADEC designs improve performance for the stated reasons. The empirical gains over the fine-tuned backbone may come from the artifact tokens, the additional binary classification heads, LoRA fine-tuning, or the prompt template. As the paper stands, the novelty claim — 'the PMOE and ADEC modules enable robust manipulation reasoning and event-level cross-modal consistency checking' — is unsupported.
minor comments (8)
  1. [Table 2] Typo: 'Detetction MoE' should be 'Detection MoE'.
  2. [§3.4] Typo: 'The t case' should be 'The second case'.
  3. [Appendix C.2] The text says 'To validate the effectiveness of CFGE' but the model is FakeSV-VLM; this appears to be a leftover from another draft.
  4. [Figure 1 and Figure 3] Figure 1 labels and Figure 3 axis text render as unreadable '/uni...' glyph paths in the PDF; the figures need to be regenerated with proper fonts.
  5. [§2.3, Eq. (1)] The notation fc ⊗ A is used for both the input to LLM_{1:l} and its output, which is confusing; use a different symbol for the contextualized features.
  6. [§2.4] Typo: 'preceive' should be 'perceive'.
  7. [Appendix C.5.2] The manual annotation of 100 samples lacks details on annotator agreement, selection bias, and class imbalance; even as an auxiliary evaluation, more protocol detail is needed.
  8. [Tables 4–5] These tables are only referenced as 'Appendix C.3, C.4' in the main text; the reader would benefit from explicit in-text discussion of the entropy-loss and layer-placement ablations.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the empirical pipeline is self-contained; cited concerns are learnability/validity issues, not reductions to inputs.

full rationale

The paper's derivation chain is empirical: a VLM backbone (InternVL2.5-8B) is fine-tuned with supervised classification losses (Eqs. 6, 11, 20), and performance is evaluated on held-out test splits of FakeSV/FakeTT against externally defined baselines. No prediction is obtained by fitting a parameter to the target quantity and then re-announcing it; the SOTA numbers in Table 1 are measured, not derived from the method's assumptions. The self-citations (CA-FVD, ASAP, etc.) appear only as baselines or related work and are not invoked to justify the framework's correctness, so they are not load-bearing. The reader's and skeptic's concerns are about whether the modules can learn what is claimed: the Attribution MoE's hard-argmax routing (Eq. 7) receives no direct supervision on the four manipulation-type probabilities, and ADEC's match label (Eqs. 17-18) gives zero explicit loss for fake pairs. These are internal-consistency/learnability objections, not cases where an output is equivalent to an input by construction. Appendix C.5.2's manual annotation of 100 samples is a post-hoc evaluation, not a training signal, and the Limitations section explicitly notes the absence of fine-grained annotations; this supports the validity concern but does not constitute circular reasoning. Therefore the circularity score is 0.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The system depends on standard pretrained VLM knowledge and on a set of hyperparameters tuned on validation. The main nontrivial assumptions are the exhaustiveness of the four-scenario taxonomy and the correctness of the labels/annotations. No new physical or unnatural entities are introduced; the 'Artifact Tokens' are learnable parameters, not claims about the world.

free parameters (5)
  • artifact token count q = 32
    Chosen by validation performance; Figure 3a shows best at 32.
  • PMOE extraction layer l = 12
    Chosen by validation; Table 5 shows layer 12 alone best.
  • contrastive temperature tau = 0.07
    Set from common contrastive learning practice; Sec 3.1.1.
  • LoRA rank/alpha = rank 8, alpha 32
    Hyperparameters chosen for fine-tuning; Sec 3.1.1.
  • learning rate / schedule = 8e-5, cosine warmup
    Standard training hyperparameters; Sec 3.1.1.
assumptions (4)
  • domain assumption FakeSV and FakeTT ground-truth labels are correct and the temporal split prevents leakage.
    The model is trained and evaluated on these labels; wrong labels would invalidate the reported accuracy.
  • domain assumption The four-scenario taxonomy (real, visual-fake, text-fake, both-fake) is exhaustive for fake short news videos.
    The Attribution MoE design assumes these four categories cover all fake news video manipulations (Sec 2.1, 2.3).
  • domain assumption A VLM pretrained on general multimodal data retains useful fact-checking knowledge after LoRA fine-tuning.
    The motivation relies on VLMs having absorbed real-world knowledge; no evidence is given that this holds for short-video news beyond the empirical results.
  • domain assumption The manual annotation of 100 FakeTT samples in Appendix C.5.2 is accurate enough to evaluate attribution routing.
    The authors annotated these samples themselves; no second annotator or agreement measure is reported.

how reviews work

0 comments
Cite this review

Pith. "Pith review of FakeSV-VLM: Taming VLM for Detecting Fake Short-Video News via Progressive Mixture-Of-Experts Adapter." pith.science (2026). https://pith.science/paper/UKWH6KIJ

@misc{pith2026250819639,
  author       = {Pith},
  title        = {Pith review of: FakeSV-VLM: Taming VLM for Detecting Fake Short-Video News via Progressive Mixture-Of-Experts Adapter},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/UKWH6KIJ}},
  note         = {Machine review of arXiv:2508.19639}
}
read the original abstract

We present FakeSV-VLM in this paper, a new VLM-based framework for detecting fake news on short video platforms. Despite significant efforts to combat this issue due to the severe threat that fake news videos pose to public information security, existing methods still fall short in detection accuracy, often due to lack of knowledge to verify the news is real or not. However, large Vision Language Models (VLMs) have absorbed extensive real-world knowledge from massive multimodal datasets. Motivated by this, we adapt advanced VLMs for fake news detection in short videos. Upon close examination of news samples, we observe that short video samples can be categorized into four distinct scenarios: both video and text are real (for real samples), or both are fake, or either the video or text is fake (for fake samples). Inspired by this insight, we design four experts tailored to handle each scenario and integrate them into VLM via Mixture of Experts. Specifically, we develop the Progressive MoE Adapter (PMOE) module where detection experts first provide an initial analysis, followed by attribution experts for a comprehensive diagnosis, leading to a robust decision. Additionally, we also note the fake news videos often show inconsistency between two modalities. Consequently, we further design the Alignment-driven Event Checking (ADEC) module, which perceives the fake news by capturing the inconsistency between different modalities. Extensive experiments on two benchmark datasets, FakeSV and FakeTT, verify the superiority of our model. It significantly outperforms current state-of-the-art models by +3.32% and +5.02%, establishing a new benchmark in the field.

Figures

Figures reproduced from arXiv: 2508.19639 by the authors.

Figure 1
Figure 1. Illustration of the proposed FakeSV-VLM framework, which is built upon a Vision Language Model [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 3
Figure 3. Ablation studies of different architectural [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figure 4
Figure 4. Four successful cases of fake news videos selected from the FakeSV and FakeTT datasets. [PITH_FULL_IMAGE:figures/full_fig_p016_4.png] view at source ↗
Figures from the paper (1 more)
Figure 5
Figure 5. Figure 5: Two failure cases of fake news videos selected from the FakeSV and FakeTT datasets. [PITH_FULL_IMAGE:figures/full_fig_p016_5.png]

Discussion (0). Sign in to comment.

Forward citations

Cited by 2 Pith papers

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

  1. Learning to Solve, Forgetting to Retain: Correct-Set Turnover in RLVR

    cs.LG 2026-06 unverdicted novelty 6.0 of 10

    RLVR exhibits correct-set turnover where solved problems regress during training, and a periodic review mechanism exploiting a repair-window principle improves retention and performance over baselines.

  2. Efficient RAG with Intent-Aware Retrieval and Semantics-Preserving Chunking

    cs.CL 2026-05 unverdicted novelty 4.0 of 10

    InSemRAG combines dynamic intent-aware hybrid retrieval and semantics-preserving chunk repair in an iterative loop, yielding 2.65 F1 gain on HotPotQA and 1.5 accuracy gain on FEVER with 4.32x lower latency than Multi-...

Reference graph

Works this paper leans on

46 extracted references · 30 canonical work pages · cited by 2 Pith papers

  1. [1]

    Javad Abbasi Aghamaleki and Alireza Behrad. 2017. Malicious inter-frame video tampering detection in mpeg videos using time and spatial domain analysis of quantization effects. Multimedia Tools and Applications, 76:20691--20717

    work page 2017

  2. [2]

    Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, and 1 others. 2023. Gpt-4 technical report. arXiv preprint arXiv:2303.08774

    arXiv 2023

  3. [3]

    Omar Ismael Al-Sanjary, Ahmed Abdullah Ahmed, Adam Amril Bin Jaharadak, Musab AM Ali, and Hewa Majeed Zangana. 2018. Detection clone an object movement using an optical flow approach. In 2018 IEEE Symposium on Computer Applications & Industrial Electronics (ISCAIE), pages 388--394. IEEE

    work page 2018

  4. [4]

    Shuai Bai, Keqin Chen, Xuejing Liu, Jialin Wang, Wenbin Ge, Sibo Song, Kai Dang, Peng Wang, Shijie Wang, Jun Tang, and 1 others. 2025. Qwen2. 5-vl technical report. arXiv preprint arXiv:2502.13923

    arXiv 2025

  5. [5]

    Yuyan Bu, Qiang Sheng, Juan Cao, Peng Qi, Danding Wang, and Jintao Li. 2024. Fakingrecipe: Detecting fake news on short video platforms from the perspective of creative process. In Proceedings of the 32nd ACM International Conference on Multimedia, pages 1351--1360

    work page 2024

  6. [6]

    Luka Bunti \'c , Mate Dami \'c , and Dalibor Gregani \'c . 2020. Impact of fake news on the global economy. Trade perspectives, pages 73--81

    work page 2020

  7. [7]

    Zhe Chen, Weiyun Wang, Yue Cao, Yangzhou Liu, Zhangwei Gao, Erfei Cui, Jinguo Zhu, Shenglong Ye, Hao Tian, Zhaoyang Liu, and 1 others. 2024. Expanding performance boundaries of open-source multimodal models with model, data, and test-time scaling. arXiv preprint arXiv:2412.05271

    arXiv 2024

  8. [8]

    Hyewon Choi and Youngjoong Ko. 2021. Using topic modeling and adversarial neural networks for fake news video detection. In Proceedings of the 30th ACM international conference on information & knowledge management, pages 2950--2954

    work page 2021

Show all 46 references
  1. [9]

    Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. Bert: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of the 2019 conference of the North American chapter of the association for computational linguistics: human...

    2019

  2. [10]

    Xingjian Diao, Ming Cheng, Wayner Barrios, and SouYoung Jin. 2025 a . Ft2tf: First-person statement text-to-talking face generation. In Proceedings of the Winter Conference on Applications of Computer Vision (WACV), pages 4821--4830

    2025

  3. [11]

    Xingjian Diao, Tianzhen Yang, Chunhui Zhang, Weiyi Wu, Ming Cheng, and Jiang Gui. 2025 b . Learning sparsity for effective and efficient music performance question answering. arXiv preprint arXiv:2506.01319

    2025 arXiv

  4. [12]

    Xingjian Diao, Chunhui Zhang, Keyi Kong, Weiyi Wu, Chiyu Ma, Zhongyu Ouyang, Peijun Qing, Soroush Vosoughi, and Jiang Gui. 2025 c . Soundmind: Rl-incentivized logic reasoning for audio-language models. arXiv preprint arXiv:2506.12935

    2025

  5. [13]

    learning musical representations for music performance question answering

    Xingjian Diao, Chunhui Zhang, Tingxuan Wu, Ming Cheng, Zhongyu Ouyang, Weiyi Wu, and Jiang Gui. 2024. "learning musical representations for music performance question answering". In "Findings of the Association for Computational Linguistics: EMNLP 2024"

    2024

  6. [14]

    Xingjian Diao, Chunhui Zhang, Weiyi Wu, Zhongyu Ouyang, Peijun Qing, Ming Cheng, Soroush Vosoughi, and Jiang Gui. 2025 d . Temporal working memory: Query-guided segment refinement for enhanced multimodal understanding. arXiv preprint arXiv:2502.06020

    2025 arXiv

  7. [15]

    Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, and 1 others. 2020. An image is worth 16x16 words: Transformers for image recognition at scale. arXiv p...

    2020 arXiv

  8. [16]

    William Fedus, Barret Zoph, and Noam Shazeer. 2022. Switch transformers: Scaling to trillion parameter models with simple and efficient sparsity. Journal of Machine Learning Research, 23(120):1--39

    2022

  9. [17]

    Rongpei Hong, Jian Lang, Jin Xu, Zhangtao Cheng, Ting Zhong, and Fan Zhou. 2025. Following clues, approaching the truth: Explainable micro-video rumor detection via chain-of-thought reasoning. In Proceedings of the ACM on Web Conference 2025, pages 4684--4698

    2025

  10. [18]

    Beizhe Hu, Qiang Sheng, Juan Cao, Yuhui Shi, Yang Li, Danding Wang, and Peng Qi. 2024. Bad actor, good advisor: Exploring the role of large language models in fake news detection. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 38, pages 22105--22113

    2024

  11. [19]

    Edward J Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, Weizhu Chen, and 1 others. 2022. Lora: Low-rank adaptation of large language models. ICLR, 1(2):3

    2022

  12. [20]

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

    2023

  13. [21]

    Yanshu Li, Yi Cao, Hongyang He, Qisen Cheng, Xiang Fu, Xi Xiao, Tianyang Wang, and Ruixiang Tang. 2025 a . https://arxiv.org/abs/2504.04633 M ^2 iv: Towards efficient and fine-grained multimodal in-context learning via representation engineering . Preprint, arXiv:2504.04633

    2025 arXiv

  14. [22]

    Yanshu Li, Tian Yun, Jianjiang Yang, Pinyuan Feng, Jinfa Huang, and Ruixiang Tang. 2025 b . Taco: Enhancing multimodal in-context learning via task mapping-guided sequence configuration. arXiv preprint arXiv:2505.17098

    2025

  15. [23]

    Jingchun Lian, Lingyu Liu, Yaxiong Wang, Yujiao Wu, Li Zhu, and Zhedong Zheng. 2024. https://doi.org/10.48550/ARXIV.2412.19685 A large-scale interpretable multi-modality benchmark for facial image forgery localization . CoRR, abs/2412.19685

    2024 doi arXiv

  16. [24]

    Yaoyuan Liang, Zhuojun Cai, Jian Xu, Guanbo Huang, Yiran Wang, Xiao Liang, Jiahao Liu, Ziran Li, Jingang Wang, and Shao-Lun Huang. 2024. Unleashing region understanding in intermediate layers for mllm-based referring expression generation. Advances in Neural Information Proces...

    2024

  17. [25]

    Xuannan Liu, Peipei Li, Huaibo Huang, Zekun Li, Xing Cui, Jiahao Liang, Lixiong Qin, Weihong Deng, and Zhaofeng He. 2024. Fka-owl: Advancing multimodal fake news detection through knowledge-augmented lvlms. In Proceedings of the 32nd ACM International Conference on Multimedia,...

    2024

  18. [26]

    Shuo Niu, Dilasha Shrestha, Abhisan Ghimire, and Zhicong Lu. 2023. A survey on watching social issue videos among youtube and tiktok users. arXiv preprint arXiv:2310.19193

    2023 arXiv

  19. [27]

    OpenAI. 2025. https://openai.com/index/gpt-4-1/ Introducing gpt-4.1 in the api . Accessed: 2025-05-01

    2025

  20. [28]

    Peng Qi, Yuyan Bu, Juan Cao, Wei Ji, Ruihao Shui, Junbin Xiao, Danding Wang, and Tat-Seng Chua. 2023 a . Fakesv: A multimodal benchmark with rich social context for fake news detection on short video platforms. In Proceedings of the AAAI Conference on Artificial Intelligence, ...

    2023

  21. [29]

    Peng Qi, Zehong Yan, Wynne Hsu, and Mong Li Lee. 2024. Sniffer: Multimodal large language model for explainable out-of-context misinformation detection. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 13052--13062

    2024

  22. [30]

    Peng Qi, Yuyang Zhao, Yufeng Shen, Wei Ji, Juan Cao, and Tat-Seng Chua. 2023 b . Two heads are better than one: Improving fake news video detection by correlating with neighbors. arXiv preprint arXiv:2306.05241

    2023 arXiv

  23. [31]

    Mubbashar Saddique, Khurshid Asghar, Usama Ijaz Bajwa, Muhammad Hussain, and Zulfiqar Habib. 2019. Spatial video forgery detection and localization using texture analysis of consecutive frames. Advances in Electrical & Computer Engineering, 19(3)

    2019

  24. [32]

    Lanyu Shang, Ziyi Kou, Yang Zhang, and Dong Wang. 2021. A multimodal misinformation detector for covid-19 short videos on tiktok. In 2021 IEEE international conference on big data (big data), pages 899--908. IEEE

    2021

  25. [33]

    K Sitara and BM Mehtre. 2017. A comprehensive approach for exposing inter-frame video forgeries. In 2017 IEEE 13th International Colloquium on Signal Processing & its Applications (CSPA), pages 73--78. IEEE

    2017

  26. [34]

    Sukrit Venkatagiri, Joseph S Schafer, and Stephen Prochaska. 2023. The challenges of studying misinformation on video-sharing platforms during crises and mass-convergence events. arXiv preprint arXiv:2303.14309

    2023 arXiv

  27. [35]

    Ivan Vykopal, Mat \'u s Pikuliak, Ivan Srba, Robert Moro, Dominik Macko, and Maria Bielikova. 2023. Disinformation capabilities of large language models. arXiv preprint arXiv:2311.08838

    2023 arXiv

  28. [36]

    Junxi Wang, Na Zhang, Yaxiong Wang, and 1 others. 2025. Consistency-aware fake videos detection on short video platforms. arXiv preprint arXiv:2504.21495

    2025 arXiv

  29. [37]

    Peng Wang, Shuai Bai, Sinan Tan, Shijie Wang, Zhihao Fan, Jinze Bai, Keqin Chen, Xuejing Liu, Jialin Wang, Wenbin Ge, and 1 others. 2024. Qwen2-vl: Enhancing vision-language model's perception of the world at any resolution. arXiv preprint arXiv:2409.12191

    2024 arXiv

  30. [38]

    Yaxiong Wang, Hao Yang, Xueming Qian, Lin Ma, Jing Lu, Biao Li, and Xin Fan. 2019. https://doi.org/10.24963/IJCAI.2019/526 Position focused attention network for image-text matching . In Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence...

    2019 doi

  31. [39]

    Chloe Wittenberg, Ben M Tappin, Adam J Berinsky, and David G Rand. 2021. The (minimal) persuasive advantage of political video over text. Proceedings of the National Academy of Sciences, 118(47):e2114388118

    2021

  32. [40]

    Jiaying Wu, Jiafeng Guo, and Bryan Hooi. 2024. Fake news in sheep's clothing: Robust fake news detection against llm-empowered style attacks. In Proceedings of the 30th ACM SIGKDD conference on knowledge discovery and data mining, pages 3367--3378

    2024

  33. [41]

    Markos Zampoglou, Foteini Markatopoulou, Gregoire Mercier, Despoina Touska, Evlampios Apostolidis, Symeon Papadopoulos, Roger Cozien, Ioannis Patras, Vasileios Mezaris, and Ioannis Kompatsiaris. 2019. Detecting tampered videos with multimedia forensics and deep learning. In Mu...

    2019

  34. [42]

    Xian Zhang, Haokun Wen, Jianlong Wu, Pengda Qin, Hui Xue', and Liqiang Nie. 2024 a . Differential-perceptive and retrieval-augmented mllm for change captioning. In Proceedings of the 32nd ACM International Conference on Multimedia, pages 4148--4157

    2024

  35. [43]

    Yuchen Zhang, Yaxiong Wang, Yujiao Wu, Lianwei Wu, and Li Zhu. 2025. https://doi.org/10.48550/ARXIV.2505.17476 The coherence trap: When mllm-crafted narratives exploit manipulated visual contexts . CoRR, abs/2505.17476

    2025 doi

  36. [44]

    Zhenxing Zhang, Yaxiong Wang, Lechao Cheng, Zhun Zhong, Dan Guo, and Meng Wang. 2024 b . Asap: Advancing semantic alignment promotes multi-modal manipulation detecting and grounding. arXiv preprint arXiv:2412.12718

    2024

  37. [45]

    online" 'onlinestring :=

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint eprinttype howpublished institution journal key month note number organization pages publisher school series title type volume year doi pubmed url lastchecked label extra.label sort.label short.list...

  38. [46]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 gl...

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.