Pith. sign in

REVIEW 3 major objections 6 minor 2 cited by

How Good is ChatGPT at Audiovisual Deepfake Detection: A Comparative Study of ChatGPT, AI Models and Human Perception

T0 review · 3 major / 6 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read A general-purpose chatbot with no forensic training can detect audiovisual deepfakes at about the same accuracy as human viewers, provided the prompt directs it to look for concrete artifacts.

desk verdict First systematic look at ChatGPT for audiovisual deepfakes, but the human-parity claim rests on 40 videos and an unverified audio mechanism. read the letter →

arxiv 2411.09266 v1 pith:I6XNPUOX submitted 2024-11-14 cs.CV cs.AIcs.HCcs.LGcs.MM

Sahibzada Adil Shahzad , Ammarah Hashmi , Yan-Tsung Peng , Yu Tsao , Hsin-Min Wang This is my paper · ORCID
classification cs.CVcs.AIcs.HCcs.LGcs.MM
keywords audiovisualdeepfakedetectionChatGPTGPT-4multimodalforensicspromptengineeringzero-shotFakeAVCelebhumanperception
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

ChatGPT, a general-purpose chatbot with no training on forensic data, can flag audiovisual deepfakes about as reliably as human viewers once the prompt points it at specific artifacts, this paper argues. On a balanced 40-video subset of the FakeAVCeleb benchmark, the two artifact-focused prompts reach 65.00% accuracy, nearly matching the 65.64% human average, while simple 'is this real?' prompts hover near chance. The paper positions this as evidence that prompt engineering can unlock an LLM's implicit multimodal knowledge for video forgery triage, with the bonus of interpretable explanations naming the offending artifact. It also stresses the gap to state-of-the-art audiovisual forensic models (87.50–97.50%), which remain far more accurate.

What carries the argument

The mechanism is prompt-conditioned multimodal analysis: the video (frames and audio) and a text prompt are fed to GPT-4, and the prompt dictates which low-level and cross-modal cues the model must check. The paper's operative tool is the artifact attribute list, ten numbered cues spanning visual forgery signs (unnatural face edges, inconsistent lighting/shading, irregular teeth, lip-movement problems, skin-texture inconsistencies), auditory signs (spectral artifacts, phoneme artifacts, speech-pattern inconsistencies, voice-quality issues), and one synchronization cue (audio-video mismatch). Prompts P6 and P7 force a structured answer—either [region, artifact] pairs or a bare attribute number—which eliminates rejected answers and raises recall compared with binary 'yes/no' prompts; the paper attributes the gain to the model's use of hand-crafted features from OpenCV, librosa, and related libraries for visual and acoustic analysis.

What would settle it

Re-test prompts P6 and P7 on the same 40 videos after replacing every audio track with a single unrelated soundtrack, and also report ChatGPT's accuracy on the real-video/fake-audio (RVFA) subset separately; if audio-track replacement leaves accuracy unchanged or RVFA accuracy is near chance, then the model is deciding from visual content alone and the acoustic-analysis mechanism is falsified.

Watch

Extended reading notes

Core claim

The paper's central claim is that ChatGPT (GPT-4) performs audiovisual deepfake detection at a human-competitive level when the input prompt enumerates the artifacts to look for, and that it does so without any task-specific fine-tuning. Using the same 40 videos from FakeAVCeleb that a prior human-perception study used, the authors report 65.00% accuracy for prompts P6 and P7—where the model must report synthesis artifacts in the visual frames or audio, or must return attribute numbers from a ten-item artifact list—against a 65.64% human average and state-of-the-art model scores between 87.50% and 97.50%. The authors interpret the result as showing that an LLM can generalize across manipulations, explain its decisions by naming visual, acoustic, or synchronization artifacts, and serve as an interpretable first-pass screening tool, while acknowledging that its reliance on hand-crafted signal-processing features and prompt sensitivity cap its accuracy well below specialized detectors.

Load-bearing premise

The load-bearing premise is that ChatGPT genuinely processes each video's audio track—extracting features like spectral centroid, MFCCs, and silence ratio—rather than producing its audio commentary from visual frames and learned priors; if the model is not actually analyzing sound, the paper's cross-modal and audio-artifact claims are unsupported.

Editorial extensions

If this is right

  • A zero-shot LLM can serve as a first-pass deepfake screener on audiovisual content, producing a human-readable artifact report before a specialized model or human examiner is called in.
  • Because the model names the artifact it sees, LLM triage could categorize manipulations by type (face swap vs. voice clone vs. both) without separate per-type classifiers.
  • The sharp drop from P6/P7 to the naive prompts implies that any user-facing deepfake tool built on an LLM must embed artifact-specific scaffolding; otherwise accuracy falls to near chance.
  • The accuracy gap between ChatGPT and trained forensic models confirms that LLMs are not a substitute for dedicated detectors, but the interpretability difference suggests hybrid pipelines—LLM rationale plus deep-model score—as the natural next step.

Reading between the lines

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

  • We infer that the paper's accuracy numbers do not by themselves prove the model is analyzing the audio track; a direct test would swap each video's audio with an unrelated soundtrack and check whether ChatGPT's decisions change, which would separate visual-cue-based inference from genuine cross-modal analysis.
  • We infer that the 'human-level' comparison on 40 videos should be treated as an existence proof rather than a stable ranking; repeating the prompt protocol on other audiovisual deepfake datasets or on larger subsets could move the 65% figure substantially.
  • We infer that the probability prompts (P3/P4) could be recalibrated against human confidence judgments from the same benchmark, giving a practical way to combine LLM probability scores with human oversight in a triage system.
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 / 6 minor

Summary. The paper investigates whether ChatGPT (GPT-4) can detect audiovisual deepfakes by analyzing both video and audio streams. Using 40 videos from the FakeAVCeleb benchmark, the authors test seven prompting strategies, compare the best results against human accuracy from their prior study [19] and against several audiovisual forensic models, and report that artifact-focused prompts (P6 and P7) achieve 65.00% accuracy, similar to the human overall accuracy of 65.64%, while the specialized models reach 87.50–97.50%. The paper also claims that ChatGPT provides interpretable artifact attributions and concludes that, with appropriate prompting, ChatGPT performs on par with humans but well below end-to-end forensic detectors.

Significance. If fully supported, the central claim would be a significant and timely result: a general-purpose LLM performing zero-shot audiovisual forgery triage at human level, with interpretable outputs, would be practically useful for initial screening. The paper has clear strengths: it uses an external benchmark dataset, a systematic set of prompts, and an explicit comparison table that imports human and AI baselines from a prior study. The authors also include a limitations section, which is good scientific practice. However, the load-bearing evidence is presently too thin to support the strong conclusions. The mechanism by which ChatGPT processes audio is not documented, the human-comparative claim lacks any uncertainty quantification, and per-manipulation results are not reported. As an exploratory study the paper is valuable, but as a comparative claim it needs substantial additional support.

major comments (3)
  1. [Section III-B, III-C, V-B] The claim that ChatGPT performs genuine acoustic analysis is not supported by the reported experimental setup. Section III-B states that videos are fed directly 'without performing any preprocessing' and that 'the model extracts audio from the video and performs visual and acoustic analysis based on input prompts,' while Section III-C lists concrete acoustic features such as spectral centroid, MFCC, and silence ratio. However, the paper does not report the ChatGPT interface or API version, whether any code-execution or tool-use mode was enabled, or how the audio track is actually ingested. Section V-B then asserts that 'ChatGPT mainly relies on hand-crafted features and traditional functions in computer vision and speech processing libraries, including OpenCV, librosa, numpy, wav, and skimage,' which is inconsistent with a plain LLM operating without tool access. The example outputs in Fig. 4 (e.g., 'Clipping detected,' 'Silence detected') are consistent with plausible text generation rather than actual measurements. Unless the authors document the audio-processing mechanism, the 'audiovisual' interpretation of the 65% result is not established; the result may reflect visual-only or prior-based detection.
  2. [Table II / Section IV-C-2] The headline claim that ChatGPT 'performs on par with humans' is not statistically supported. The comparison uses a single run of seven prompts on 40 videos; P6 and P7 yield 65.00% accuracy, which is only 0.64 percentage points below the human overall average of 65.64% taken from the authors' prior study [19]. No confidence intervals, repeated trials, or significance tests are provided. For n=40, the standard error of an accuracy near 65% is roughly 7.5 percentage points, so the observed difference is negligible. In addition, P6 and P7 were selected after inspecting results across all seven prompts, which inflates the expected best-case accuracy. The human baseline is also imported from a different experimental protocol (a two-phase subjective test) without human variance being reported. To support the 'comparable to human' claim, the authors should provide per-run variability, statistical tests, and a clear statement of the prompt-selection procedure.
  3. [Abstract / Section IV-A / Fig. 1] The abstract and conclusions claim that ChatGPT can 'account for spatial and spatiotemporal artifacts and inconsistencies that may exist within or across modalities,' but the results are reported only as aggregate accuracy on the 40-video set. The FakeAVCeleb subset spans three manipulation types (FVRA, RVFA, FVFA) described in Fig. 1, and the paper never reports accuracy per manipulation type. Without this breakdown, the reader cannot determine whether the 65% result reflects genuine cross-modal detection (e.g., of audio artifacts in RVFA) or mostly visual/prior-based behavior. Reporting per-manipulation accuracy is essential to substantiate the audiovisual forgery triage claim.
minor comments (6)
  1. [Table II / Section IV-C-2] The table header and the main text contain the typo 'ChatGP' instead of 'ChatGPT'; for example, 'ChatGP performs on par with humans' in Section IV-C-2.
  2. [Fig. 4] Fig. 4 contains two typos: 'Predication' should be 'Prediction' and 'lable' should be 'label.'
  3. [Section V-A] The ablation prompts are described as 'video-only mention prompts,' yet the last prompt in the list includes '10-lack of synchronization between audio and video,' which is an audiovisual attribute; this is confusing and should be clarified or removed from the video-only set.
  4. [Section V-A] The seventh prompt in the ablation list says 'Check if the video has the artifact attribute in the following list and ONLY return the attribute number in this image,' but the task is video-based; the phrase 'in this image' appears to be copied from an image-forensics study and should be adapted to the video setting.
  5. [Section III] The paper states 'we used OpenAI's GPT-4' but does not specify the exact model version, the interface (API vs. web), the date of deployment, or any parameter settings; this information is needed for reproducibility.
  6. [Section V] The heading 'Ablation Study' is somewhat misleading because the section varies prompt design rather than ablating model components; consider renaming it to 'Prompt-Modality Analysis' or 'Prompt Design Experiments.'

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the ChatGPT benchmark is an external task and no fitted parameter is renamed as a prediction.

full rationale

The paper's central result is an external measurement: 40 videos from FakeAVCeleb with ground-truth labels are fed to ChatGPT under different prompts, and the outputs are scored with the standard metrics in Eqs. 2-5. There is no fitted parameter, no training loop, and no equation whose output is defined from the target labels. The prompt designs P1-P7 are input variations, not quantities inferred from the test labels; selecting the best-performing prompt after seeing the test set is a statistical multiple-comparison concern, not a circularity. The human and AI baselines in Table II are imported from the authors' prior work [19] and from related papers that share authors with this study ([19], [26], [27], [28], [30]). However, those baselines are external empirical measurements on the same fixed video subset, not quantities derived from or defined in terms of the ChatGPT outputs. The self-citation overlap therefore does not make the ChatGPT result tautological; the citation is real evidence because the human and model accuracies were measured outside the present paper's fitted values. The paper's claim that ChatGPT genuinely performs audiovisual analysis rests on an unverified mechanism (Section III-C claims spectral/MFCC/silence-ratio analysis while Section V-B says ChatGPT relies on hand-crafted library functions, with no API or tool-use configuration reported). That is an assumption-validity and correctness concern, not a circular derivation, because the measured accuracy numbers do not logically presuppose that mechanism. Overall, no circular step can be identified from the paper's own equations or citations.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The study rests on assumptions that GPT-4's multimodal input includes actual audio waveform processing, that the 40-video subset is representative, and that baseline numbers from prior self-authored studies transfer. No numerical parameters are fitted to data, but the probability threshold and the choice of best prompt are implicit selection decisions that affect the reported headline accuracy.

free parameters (3)
  • Probability-to-label threshold for P3 and P4
    The paper reports accuracy for probability prompts but does not specify the threshold used to map the 0 to 100 probability score to a real or fake binary label, or how the average of audio and video scores is combined (Sections III-D and V-B).
  • Rejection handling denominator
    For prompts with a nonzero rejection rate, accuracy appears to be computed after dropping rejected samples, but the paper never states whether the denominator is 40 or the number of accepted responses; Table III shows similar ambiguity.
  • Best-prompt selection (P6 and P7)
    The headline "comparable to humans" uses the best-performing prompts chosen after evaluating seven hand-designed prompts; the paper reports all prompts, which mitigates this, but the summary emphasizes the maximum rather than an average or a held-out prompt.
assumptions (5)
  • domain assumption GPT-4 can perceive and analyze the audio track of an input video.
    The method description in Sections III-B and III-C assumes ChatGPT extracts and analyzes audio (spectral features, MFCC, silence ratio) from the raw video. GPT-4's actual audio perception capabilities are not demonstrated.
  • domain assumption The 40-video subset from FakeAVCeleb is representative of audiovisual deepfakes.
    Section IV-A selects the same 40 videos as [19] for fair comparison; no argument is given that this subset spans manipulation types or difficulty levels beyond the three categories in Fig. 1.
  • domain assumption Ground-truth labels of the selected videos are correct.
    Labels are inherited from FakeAVCeleb [46] and the prior study [19]; no independent verification is performed.
  • domain assumption Human and AI baseline numbers from [19] transfer to this evaluation context.
    Table II imports human phase accuracies and AI model accuracies from the authors' prior study without re-running them, assuming identical conditions such as playback, model versions, and evaluation protocol.
  • ad hoc to paper ChatGPT's textual analysis reflects computation rather than plausible text generation.
    Sections III-C and V-B attribute concrete signal-processing calculations (zero crossing rate, MFCC, SSIM) to ChatGPT; no evidence shows the model executes such tools, and the claim that it relies on OpenCV and librosa is not verifiable.

how reviews work

0 comments
Cite this review

Pith. "Pith review of How Good is ChatGPT at Audiovisual Deepfake Detection: A Comparative Study of ChatGPT, AI Models and Human Perception." pith.science (2026). https://pith.science/paper/I6XNPUOX

@misc{pith2026241109266,
  author       = {Pith},
  title        = {Pith review of: How Good is ChatGPT at Audiovisual Deepfake Detection: A Comparative Study of ChatGPT, AI Models and Human Perception},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/I6XNPUOX}},
  note         = {Machine review of arXiv:2411.09266}
}
read the original abstract

Multimodal deepfakes involving audiovisual manipulations are a growing threat because they are difficult to detect with the naked eye or using unimodal deep learningbased forgery detection methods. Audiovisual forensic models, while more capable than unimodal models, require large training datasets and are computationally expensive for training and inference. Furthermore, these models lack interpretability and often do not generalize well to unseen manipulations. In this study, we examine the detection capabilities of a large language model (LLM) (i.e., ChatGPT) to identify and account for any possible visual and auditory artifacts and manipulations in audiovisual deepfake content. Extensive experiments are conducted on videos from a benchmark multimodal deepfake dataset to evaluate the detection performance of ChatGPT and compare it with the detection capabilities of state-of-the-art multimodal forensic models and humans. Experimental results demonstrate the importance of domain knowledge and prompt engineering for video forgery detection tasks using LLMs. Unlike approaches based on end-to-end learning, ChatGPT can account for spatial and spatiotemporal artifacts and inconsistencies that may exist within or across modalities. Additionally, we discuss the limitations of ChatGPT for multimedia forensic tasks.

Figures

Figures reproduced from arXiv: 2411.09266 by the authors.

Figure 1
Figure 1. Illustration of audiovisual deepfake manipulations. Original video [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Illustration of the multimodal capabilities of a large language [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Bar graph comparing the number of True Positives, False Positives, [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Demonstration of ChatGPT responses, which takes video and text prompts as input and produces audiovisual analysis, including explanations and [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

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

  1. Digital Forensic Investigation of the ChatGPT Windows Application

    cs.CR 2025-05 conditional novelty 4.0 of 10

    This paper catalogs recoverable forensic traces left by the ChatGPT Windows desktop app, including chat prompts in RAM, registry entries, disk files, network endpoints, and export metadata, and shows they can survive ...

  2. Survey on AI-Generated Media Detection: From Non-MLLM to MLLM

    cs.CV 2025-02 unverdicted novelty 3.0 of 10

    A survey organizing AI-generated media detection into Non-MLLM and MLLM based methods, with task and benchmark taxonomies.

Reference graph

Works this paper leans on

50 extracted references · 43 canonical work pages · cited by 2 Pith papers

  1. [19]

    Hashmi, S

    A. Hashmi, S. A. Shahzad, C.-W. Lin, Y . Tsao, H.-M. Wang, Unmasking Illusions: Understanding human perception of audiovisual deepfakes, arXiv preprint arXiv:2405.04097 (2024)

    arXiv 2024

  2. [1]

    Goodfellow, J

    I. Goodfellow, J. Pouget-Abadie, M. Mirza, B. Xu, D. Warde-Farley, S. Ozair, A. Courville, Y . Bengio, Generative adversarial networks, Communications of the ACM 63 (11) (2020) 139–144

    work page 2020

  3. [2]

    J. Ho, A. Jain, P. Abbeel, Denoising diffusion probabilistic models, in: Proceedings of the Advances in Neural Information Processing Systems, V ol. 33, 2020, pp. 6840–6851

    work page 2020

  4. [3]

    A. O. Kwok, S. G. Koh, Deepfake: a social construction of technology perspective, Current Issues in Tourism 24 (13) (2021) 1798–1802

    work page 2021

  5. [4]

    Ray, Disinformation, deepfakes and democracies: The need for legislative reform, The University of New South Wales Law Journal 44 (3) (2021) 983–1013

    A. Ray, Disinformation, deepfakes and democracies: The need for legislative reform, The University of New South Wales Law Journal 44 (3) (2021) 983–1013

    work page 2021

  6. [5]

    Figueira, L

    ´A. Figueira, L. Oliveira, The current state of fake news: challenges and opportunities, Procedia Computer Science 121 (2017) 817–825

    work page 2017

  7. [6]

    Vaccari, A

    C. Vaccari, A. Chadwick, Deepfakes and disinformation: Exploring the impact of synthetic political video on deception, uncertainty, and trust in news, Social Media+ Society 6 (1) (2020)

    work page 2020

  8. [7]

    Afchar, V

    D. Afchar, V . Nozick, J. Yamagishi, I. Echizen, MesoNet: a compact facial video forgery detection network, in: Proceedings of the IEEE International Workshop on Information Forensics and Security, 2018, pp. 1–7

    work page 2018

Show all 50 references
  1. [8]

    H. H. Nguyen, J. Yamagishi, I. Echizen, Capsule-Forensics: Using capsule networks to detect forged images and videos, in: Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing, 2019, pp. 2307–2311

    2019

  2. [9]

    K. Lutz, R. Bassett, Deepfake Detection with Inconsistent Head Poses: Reproducibility and analysis, arXiv preprint arXiv:2108.12715 (2021)

    2021 arXiv

  3. [10]

    Haliassos, K

    A. Haliassos, K. V ougioukas, S. Petridis, M. Pantic, Lips Don’t Lie: A generalisable and robust approach to face forgery detection, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021, pp. 5039–5049

    2021

  4. [11]

    L. Wang, Y . Yoshida, Y . Kawakami, S. Nakagawa, Relative phase infor- mation for detecting human speech and spoofed speech., in: Proceedings of the Interspeech Conference, 2015, pp. 2092–2096

    2015

  5. [12]

    Todisco, H

    M. Todisco, H. Delgado, N. Evans, A new feature for automatic speaker verification anti-spoofing: Constant Q cepstral coefficients, in: Proceed- ings of the Speaker and Language Recognition Workshop (Odyssey), V ol. 45, 2016, p. 283

    2016

  6. [13]

    T. B. Patel, H. A. Patil, Combining evidences from mel cepstral, cochlear filter cepstral and instantaneous frequency features for detection of natural vs. spoofed speech., in: Proceedings of the Interspeech Conference, 2015, pp. 2062–2066

    2015

  7. [14]

    Wu, H.-C

    H. Wu, H.-C. Kuo, N. Zheng, K.-H. Hung, H.-Y . Lee, Y . Tsao, H.- M. Wang, H. Meng, Partially fake audio detection by self-attention- based fake span discovery, in: Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing, 2022, pp. 9236–9240

    2022

  8. [15]

    Karnouskos, Artificial intelligence in digital media: The era of deepfakes, IEEE Transactions on Technology and Society 1 (3) (2020) 138–147

    S. Karnouskos, Artificial intelligence in digital media: The era of deepfakes, IEEE Transactions on Technology and Society 1 (3) (2020) 138–147

    2020

  9. [16]

    G. P. Zachary, Digital manipulation and the future of electoral democracy in the US, IEEE Transactions on Technology and Society 1 (2) (2020) 104–112. 8

    2020

  10. [17]

    J. R. Carvalko, Generative AI, ingenuity, and law, IEEE Transactions on Technology and Society 5 (2) (2024) 169–182

    2024

  11. [18]

    Apolo, K

    Y . Apolo, K. Michael, Beyond a reasonable doubt? audiovisual evi- dence, AI manipulation, deepfakes, and the law, IEEE Transactions on Technology and Society 5 (2) (2024) 156–168

    2024

  12. [20]

    Korshunova, W

    I. Korshunova, W. Shi, J. Dambre, L. Theis, Fast face-swap using convolutional neural networks, in: Proceedings of the IEEE International Conference on Computer Vision, 2017, pp. 3677–3685

    2017

  13. [21]

    Nirkin, Y

    Y . Nirkin, Y . Keller, T. Hassner, FSGAN: Subject agnostic face swap- ping and reenactment, in: Proceedings of the IEEE/CVF International Conference on Computer Vision, 2019, pp. 7184–7193

    2019

  14. [22]

    Prajwal, R

    K. Prajwal, R. Mukhopadhyay, V . P. Namboodiri, C. Jawahar, A lip sync expert is all you need for speech to lip generation in the wild, in: Proceedings of the ACM International Conference on Multimedia, 2020, pp. 484–492

    2020

  15. [23]

    Y . Jia, Y . Zhang, R. Weiss, Q. Wang, J. Shen, F. Ren, P. Nguyen, R. Pang, I. Lopez Moreno, Y . Wu, et al., Transfer learning from speaker verification to multispeaker text-to-speech synthesis, in: Proceedings of the Advances in Neural Information Processing Systems, V ol. 31...

    2018

  16. [24]

    Zhou, S.-N

    Y . Zhou, S.-N. Lim, Joint audio-visual deepfake detection, in: Proceed- ings of the IEEE/CVF International Conference on Computer Vision, 2021, pp. 14800–14809

    2021

  17. [25]

    Khalid, M

    H. Khalid, M. Kim, S. Tariq, S. S. Woo, Evaluation of an audio-video multimodal deepfake dataset using unimodal and multimodal detectors, in: Proceedings of the Workshop on Synthetic Multimedia-Audiovisual Deepfake Generation and Detection, 2021, pp. 7–15

    2021

  18. [26]

    Hashmi, S

    A. Hashmi, S. A. Shahzad, W. Ahmad, C. W. Lin, Y . Tsao, H.-M. Wang, Multimodal forgery detection using ensemble learning, in: Proceedings of the Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, 2022, pp. 1524–1532

    2022

  19. [27]

    S. A. Shahzad, A. Hashmi, S. Khan, Y .-T. Peng, Y . Tsao, H.-M. Wang, Lip sync matters: A novel multimodal forgery detector, in: Proceedings of the Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, 2022, pp. 1885–1892

    2022

  20. [28]

    S. A. Shahzad, A. Hashmi, Y .-T. Peng, Y . Tsao, H.-M. Wang, A V-Lip- Sync+: Leveraging av-hubert to exploit multimodal inconsistency for video deepfake detection, arXiv preprint arXiv:2311.02733 (2023)

    2023

  21. [29]

    Ilyas, A

    H. Ilyas, A. Javed, K. M. Malik, A VFakeNet: A unified end-to-end dense swin transformer deep learning model for audio-visual deepfakes detection, Applied Soft Computing 136 (2023) 110124

    2023

  22. [30]

    Hashmi, S

    A. Hashmi, S. A. Shahzad, C.-W. Lin, Y . Tsao, H.-M. Wang, A VTENet: Audio-visual transformer-based ensemble network exploiting multiple experts for video deepfake detection, arXiv preprint arXiv:2310.13103 (2023)

    2023 arXiv

  23. [31]

    W. Yang, X. Zhou, Z. Chen, B. Guo, Z. Ba, Z. Xia, X. Cao, K. Ren, AvoiD-DF: Audio-visual joint learning for detecting deepfake, IEEE Transactions on Information Forensics and Security 18 (2023) 2015– 2029

    2023

  24. [32]

    Y . Xu, P. Terh¨ost, M. Pedersen, K. Raja, Analyzing fairness in deepfake detection with massively annotated databases, IEEE Transactions on Technology and Society 5 (1) (2024) 93–106

    2024

  25. [33]

    Y . Bang, S. Cahyawijaya, N. Lee, W. Dai, D. Su, B. Wilie, H. Lovenia, Z. Ji, T. Yu, W. Chung, et al., A multitask, multilingual, multimodal evaluation of ChatGPT on reasoning, hallucination, and interactivity, in: Proceedings of the International Joint Conference on Natural L...

    2023

  26. [34]

    Vaikunta Pai, P

    T. Vaikunta Pai, P. Nethravathi, R. Birau, V . Popescu, B. Karthik Pai, P. V . Naik, Multimodal ChatGPT: Extending ChatGPT to enable rich multimodal conversations using deep neural network, Journal of Intelli- gent & Fuzzy Systems (2024) 1–17

    2024

  27. [35]

    Z. Yang, L. Li, J. Wang, K. Lin, E. Azarnasab, F. Ahmed, Z. Liu, C. Liu, M. Zeng, L. Wang, MM-REACT: Prompting ChatGPT for multimodal reasoning and action, arXiv preprint arXiv:2303.11381 (2023)

    2023 arXiv

  28. [36]

    S. Jia, R. Lyu, K. Zhao, Y . Chen, Z. Yan, Y . Ju, C. Hu, X. Li, B. Wu, S. Lyu, Can ChatGPT detect deepfakes? a study of using multimodal large language models for media forensics, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024, pp....

    2024

  29. [37]

    DeAndres-Tame, R

    I. DeAndres-Tame, R. Tolosana, R. Vera-Rodriguez, A. Morales, J. Fier- rez, J. Ortega-Garcia, How good is ChatGPT at face biometrics? a first look into recognition, soft biometrics, and explainability, IEEE Access 12 (2024) 34390–34401

    2024

  30. [38]

    X. Yang, J. Zhou, Research about the ability of LLM in the tamper- detection area, arXiv preprint arXiv:2401.13504 (2024)

    2024 arXiv

  31. [39]

    K. M. Caramancion, Harnessing the power of ChatGPT to decimate mis/disinformation: Using ChatGPT for fake news detection, in: Pro- ceedings of the IEEE World AI IoT Congress, 2023, pp. 0042–0046

    2023

  32. [40]

    Koco ´n, I

    J. Koco ´n, I. Cichecki, O. Kaszyca, M. Kochanek, D. Szydło, J. Baran, J. Bielaniewicz, M. Gruza, A. Janz, K. Kanclerz, et al., ChatGPT: Jack of all trades, master of none, Information Fusion 99 (2023) 101861

    2023

  33. [41]

    G. Wu, W. Wu, X. Liu, K. Xu, T. Wan, W. Wang, Cheap-fake detection with LLM using prompt engineering, in: Proceedings of the IEEE International Conference on Multimedia and Expo Workshops, 2023, pp. 105–109

    2023

  34. [42]

    S. S. Biswas, Potential use of Chat GPT in global warming, Annals of Biomedical Engineering 51 (6) (2023) 1126–1127

    2023

  35. [43]

    J. He, L. Li, W. Yao, H. Gao, Exploring Future Education: The innovative integration and practice of multimodal learning and ChatGPT, in: Proceedings of the International Conference on Computer Science, Engineering, and Education, 2024, pp. 18–23

    2024

  36. [44]

    S. S. Biswas, Role of Chat GPT in public health, Annals of Biomedical Engineering 51 (5) (2023) 868–869

    2023

  37. [45]

    Z. Yan, K. Zhang, R. Zhou, L. He, X. Li, L. Sun, Multimodal ChatGPT for medical applications: an experimental study of GPT-4V, arXiv preprint arXiv:2310.19061 (2023)

    2023 arXiv

  38. [46]

    Khalid, S

    H. Khalid, S. Tariq, M. Kim, S. S. Woo, FakeA VCeleb: A novel audio-video multimodal deepfake dataset, in: Proceedings of the Neural Information Processing Systems Track on Datasets and Benchmarks, 2021

    2021

  39. [47]

    Shi, W.-N

    B. Shi, W.-N. Hsu, K. Lakhotia, A. Mohamed, Learning audio-visual speech representation by masked multimodal cluster prediction, in: Pro- ceedings of the International Conference on Learning Representations, 2021

    2021

  40. [48]

    Dosovitskiy, An Image is Worth 16x16 Words: Transformers for image recognition at scale, arXiv preprint arXiv:2010.11929 (2020)

    A. Dosovitskiy, An Image is Worth 16x16 Words: Transformers for image recognition at scale, arXiv preprint arXiv:2010.11929 (2020)

    2020 arXiv

  41. [49]

    Arnab, M

    A. Arnab, M. Dehghani, G. Heigold, C. Sun, M. Lu ˇci´c, C. Schmid, ViViT: A video vision transformer, in: Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021, pp. 6836–6846

    2021

  42. [50]

    Gong, Y .-A

    Y . Gong, Y .-A. Chung, J. Glass, AST: Audio spectrogram transformer, in: Proceedings of the Interspeech Conference, 2021, pp. 571–575

    2021

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.