Pith. sign in

REVIEW 3 major objections 4 minor 6 cited by

Holmes-VAU: Towards Long-term Video Anomaly Understanding at Any Granularity

T0 review · 3 major / 4 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read The paper argues that hierarchical clip, event, and video annotations let a single model both localize and explain anomalies across time scales.

desk verdict HIVAU-70k is a genuinely useful hierarchical anomaly-understanding benchmark; the paper's detection and reasoning evaluations do not support the headline claims as currently written. read the letter →

arxiv 2412.06171 v2 pith:UFOUKWYK submitted 2024-12-09 cs.CV

Huaxin Zhang , Xiaohao Xu , Xiang Wang , Jialong Zuo , Xiaonan Huang , Changxin Gao , Shanjun Zhang , Li Yu
show 1 more author
Nong Sang
This is my paper · ORCID
classification cs.CV
keywords videoanomalydetectionunderstandinghierarchicalannotationsmultimodallargelanguagemodelsAnomaly-focusedTemporalSamplerlong-videoreasoningbenchmarkconstructionvisual-language
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper tries to establish that video anomaly understanding should be treated as a multi-scale reasoning problem, not just frame-level scoring. It introduces HIVAU-70k, a benchmark with more than 70,000 annotations organized at clip, event, and video levels, built from the UCF-Crime and XD-Violence collections using a semi-automated engine that pairs manual event segmentation with LLM-generated summaries. The paper also proposes the Anomaly-focused Temporal Sampler (ATS), which uses a lightweight anomaly scorer to concentrate a visual-language model's sampled frames in anomaly-dense stretches of long videos. On its test set, the resulting Holmes-VAU model reports higher anomaly-reasoning metrics than general video MLLMs, and higher detection scores than previously reported video anomaly detectors, with a much smaller model size. The point of the work is that hierarchical annotations plus adaptive sampling let one model both find and explain anomalies across temporal scales.

What carries the argument

The load-bearing object is the Anomaly-focused Temporal Sampler (ATS): a two-part module that combines an anomaly scorer with a density-aware sampler. The scorer, following the UR-DMU architecture, maps each frame's class token to an anomaly score; the sampler treats those scores as a probability mass, accumulates them along time as $S_{\text{cumsum}}(t)=\sum_{i=1}^{t}(s_i+\tau)$, and uniformly samples $N$ points on the cumulative axis to choose which frames enter the language model. The $\tau$ term keeps a floor on uniform coverage, so sampling clusters around anomaly-rich moments without discarding context. The other load-bearing piece is the annotation engine: manual event segmentation followed by recursive LLM summarization from clip captions to event summaries to video summaries, which yields the Judgment, Description, and Analysis instruction data. Together they let a fixed 2B visual-language model spend its visual tokens where anomalies actually occur and learn to verbalize judgments at any timescale.

What would settle it

Re-annotate a random sample of the videos with an independent team that marks event boundaries and checks whether the existing summaries correctly describe the anomalies; if inter-annotator agreement on boundaries is low, or if the LLM summaries frequently conflict with independent event descriptions, then both the training signal and the reasoning-test ground truth are unreliable. Separately, a controlled test with long videos containing two well-separated anomaly events can show whether ATS misses the second event after concentrating on the first at an equal frame budget.

Watch

Extended reading notes

Core claim

The central claim is that anomaly comprehension improves when a model is trained and evaluated at three temporal granularities at once: short clips, anomaly events, and whole videos. The paper constructs this hierarchy automatically at scale by having humans mark event boundaries, using a video captioning model for clips, and then asking an LLM to recursively summarize clip captions into event summaries and event summaries into video summaries, with manual checking. The second claim is that the frame budget of a video-language model should be allocated by anomaly content: ATS treats anomaly scores as a temporal distribution, forms a cumulative sum, and samples frames by an inverse-transform-like mapping, so more frames land where anomaly scores are high. In experiments, training on the full hierarchy outperforms training on any single granularity, and ATS beats uniform and top-K sampling at equal frame counts. The paper reports an AP of 87.68% on XD-Violence, an AUC of 88.96% on UCF-Crime, and substantially higher BLEU, CIDEr, METEOR, and ROUGE scores than general video-language baselines on its 3,300-item reasoning test set.

Load-bearing premise

The benchmark's value depends on the manual event-boundary labels and the LLM-written summaries being accurate and consistent enough to serve both as frame-level training supervision for the anomaly scorer and as ground truth for evaluating anomaly reasoning.

Editorial extensions

If this is right

  • The full hierarchy becomes the training recipe: clip data supplies perception, event data supplies anomaly judgment, and video data supplies long-range analysis; dropping any level lowers measured reasoning quality.
  • With sixteen sampled frames, the ATS-equipped model beats uniform and top-K sampling at the same visual-token budget, so long-video anomaly reasoning can be run with far fewer frames than dense sampling requires.
  • Because the anomaly scorer is trained on HIVAU-70k's event boundaries, it can serve as a standalone frame-level detector; the paper reports it outperforms weakly-supervised baselines on UCF-Crime and XD-Violence.
  • The released benchmark lets other groups fine-tune video-language models on hierarchical anomaly instructions without repeating the annotation pipeline.

Reading between the lines

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

  • The cumulative-sum sampler is inverse-transform sampling over the frame salience distribution; making the scorer differentiable and training the sampler end-to-end with the language model is a natural next step.
  • ATS could transfer to other long-video tasks that have a relevance score, such as action localization, highlight detection, or question-guided retrieval.
  • Because the reasoning ground truth is LLM-generated and only lightly checked, the reported BLEU and CIDEr gaps may partly reward lexical similarity to those summaries; a human-preference or factual-consistency probe would test whether the reasoning gains are real.
  • Replacing manual event segmentation with a segmentation model and human verification of boundaries could scale the same engine to more anomaly classes and domains at lower cost.
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 HIVAU-70k, a hierarchical video anomaly understanding benchmark with clip-, event-, and video-level annotations, built from UCF-Crime and XD-Violence using a semi-automated pipeline that combines manual event-boundary labeling with LLM-generated summaries. The paper also proposes Holmes-VAU, a multimodal LLM pipeline with an Anomaly-focused Temporal Sampler (ATS) that scores frames with a trained anomaly scorer and non-uniformly samples frames for the LLM. Experiments report state-of-the-art detection results on UCF-Crime and XD-Violence, large gains over general MLLMs on the new reasoning test set, and ablations supporting the ATS and hierarchical instruction data.

Significance. If properly validated, HIVAU-70k would be a valuable community resource: it is large, publicly released, and addresses a real gap by providing multi-granular anomaly annotations. The ATS idea, using anomaly scores to guide temporal sampling for a VLM, is also interesting and could transfer to other long-video understanding tasks. However, the current evidence for the two headline claims ('outperforms traditional methods' and 'hierarchical instruction data markedly improves anomaly comprehension') is weakened by a supervision confound in the detection comparison and by a self-consistent evaluation protocol for reasoning. The paper would be significantly strengthened by controlled experiments that separate the contribution of the extra frame-level supervision from the contribution of the ATS/VLM.

major comments (3)
  1. [Table 1 / Sec 4.3 / Eq. (5)] The detection comparison is not apples-to-apples. Holmes-VAU's anomaly scorer is initialized from UR-DMU and trained with frame-level labels derived from manual event boundaries using binary cross-entropy (Eq. 5), whereas all non-explainable baselines in Table 1 are unsupervised or weakly supervised with only video-level labels. The reported gains (87.68 AP on XD-Violence, 88.96 AUC on UCF-Crime) could therefore be due entirely to the added frame-level supervision. I recommend adding a fully supervised baseline, most naturally UR-DMU trained with the same frame-level labels and the same visual features, and/or a weakly supervised variant of the proposed scorer. Without such a control, the abstract's claim that the integrated ATS and visual-language model outperform traditional methods is not established.
  2. [Sec 5.1 / Sec 3.2 / Table 2] The reasoning evaluation is strongly self-referential. The test ground-truth text is generated with the same LLM pipeline (LLaMA3-70B summaries built from clip captions with the same prompt templates) that produced the instruction-tuning data, and the model is fine-tuned to imitate that style. Under this protocol, high BLEU/CIDEr scores may reflect style mimicry rather than anomaly comprehension. The manual review described in Sec 5.1 does not remove this bias, since the underlying text and its style are inherited from the annotation engine. Please add an evaluation on independently written human annotations, a human preference study, or an existing external anomaly reasoning benchmark; also consider reporting judgment accuracy as a binary classification metric rather than only lexical overlap.
  3. [Table 4 / Sec 5.3] The efficiency and accuracy claims for ATS are only partially supported. In Table 4, latency values are reported only for Uniform sampling; the Top-K and ATS rows have empty latency cells, so the paper does not demonstrate that ATS is more efficient than the alternatives. In addition, the accuracy comparison is limited to video-level BLEU and CIDEr on the authors' own test set; no detection metrics (AUC/AP) are reported for different samplers, even though detection is a central claim of the paper. Please report latency for all samplers and, if detection is claimed to improve, the corresponding detection results.
minor comments (4)
  1. [Sec 3.2] Typo: 'off-the-shell' should be 'off-the-shelf'.
  2. [Table 3] The checkmark alignment in Table 3 is hard to read; the table would be clearer if each row explicitly listed the included granularities (e.g., 'C only', 'E only', 'C+E') rather than leaving the reader to infer them from the checkmark columns.
  3. [Fig. 6] Figure 6(a) shows loss curves but the axes labels are minimal ('iter' and 'loss'), and Figure 6(b) does not report the numerical values for Video-MME; please provide the exact scores so readers can judge the trade-off quantitatively.
  4. [Sec 3.3 / Supp. A.2] The annotation quality control is described qualitatively ('manual checking', 'carefully reviewed'), but no inter-annotator agreement or consistency statistics are reported. Even a small-scale agreement study on event boundaries and summary correctness would increase confidence in the benchmark.

Circularity Check

2 steps flagged · score 4.0 of 10

Detection comparison is confounded by frame-level supervision, and the reasoning benchmark shares its LLM annotation pipeline with the training data, producing partial circularity; the ATS and hierarchical-data ablations remain internally controlled.

  1. fitted input called prediction [Sec. 4.3 (Eq. 5), Sec. 5.1 (Detection metric), Sec. 5.2 (Table 1)]
    "In the first step, we use the video data and annotated frame-level label (ˆy ∈ RT) from HIVAU-70k to train the anomaly scorer, which provides more accurate anomaly supervision compared to previous unsupervised and weakly-supervised methods. ... For anomaly detection, we use the anomaly scores output by the Anomaly Scorer as the prediction and perform the evaluation."

    The anomaly scorer is UR-DMU trained with binary cross-entropy (Eq. 5) on frame-level labels derived from manual event boundaries in HIVAU-70k. Table 1 then reports this scorer's AP/AUC as 'Holmes-VAU (Ours)' and uses it to support the claim that the integrated ATS and VLM outperform traditional methods. The baselines (RTFM, S3R, UR-DMU, VadCLIP, etc.) are weakly supervised or unsupervised and never receive frame-level labels. The reported gains (87.68 AP on XD-Violence, 88.96 AUC on UCF-Crime) are therefore attributable to the added frame-level supervision, not to ATS or the VLM.

  2. self definitional [Sec. 5.1 (Reasoning test set), Sec. 4.3 (Instruction tuning), Sec. A.2 (Event/Video Summary prompts)]
    "For anomaly reasoning, we annotate instruction data from the UCF-Crime and XD-Violence test sets, which have been carefully reviewed and filtered by annotators. ... Event Summary. We combined all captions and video-level category labels to generate anomaly-related summaries for each event using an LLM."

    The instruction-tuning targets are all HIVAU-70k instruction data produced by the recursive LLM summarization pipeline (LLaMA3-70B with the A.2 prompts, then manual review). The test references for anomaly reasoning are generated by the same pipeline on the test videos and only lightly human-filtered. The model is trained to maximize p(Xa|Xins) on the former and scored by BLEU/CIDEr/METEOR/ROUGE against the latter. Thus the 'anomaly comprehension' metric measures how well the model reproduces the annotation LLM's output distribution, not an independently established ground truth. The evaluation target is self-defined by the same process that generated the training signal, so the large reasoning gains are partly circular; human checking does not remove the shared generative pipeline.

full rationale

The paper contains genuinely independent empirical content: the ATS ablation in Table 4 compares ATS against Uniform and Top-K sampling under identical supervision, and the hierarchical-data ablation in Table 3 varies training granularities with the same evaluation protocol. Detection is evaluated on the external UCF-Crime and XD-Violence test splits, so it is not a pure self-benchmark. However, the headline claim that the integrated ATS and VLM outperform traditional methods rests on Table 1, where the anomaly scorer is trained with frame-level labels while the baselines use weaker supervision; that comparison is confounded and cannot isolate the contribution of ATS or the VLM. Separately, the reasoning benchmark and the training set are produced by the same LLM annotation engine, so the reported reasoning improvements partly reflect distribution matching to that engine rather than independently verified anomaly understanding. There is no load-bearing self-citation or imported uniqueness theorem; the issues are evaluation-loop and supervision-confound circularity rather than derivation-by-definition. Overall, partial circularity warrants a score of 4.

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

The central claims rest on the quality of manual event boundaries and LLM-generated text annotations. The hyperparameters listed are typical training choices, and the domain assumptions are not formally established in the paper.

free parameters (5)
  • tau (sampling uniformity) = 0.1
    Controls the uniformity of the density-aware sampler in Eq. 2; set to 0.1 during testing and not varied.
  • Number of sampled frames N = 16
    Default number of frames fed to the LLM; chosen to balance latency and performance in Table 4.
  • LoRA rank r = 64
    Rank of LoRA adaptation; chosen via ablation in Fig. 6 to trade off VAU performance and general Video-MME capability.
  • Anomaly scorer learning rate = 1e-4
    Optimizer hyperparameter for the anomaly scorer training step.
  • Instruction tuning learning rate = 4e-5
    AdamW learning rate with cosine decay for LoRA tuning.
assumptions (4)
  • domain assumption Anomalous frames contain more information and exhibit greater variation than normal frames
    Stated in Sec 4.2 to motivate density-aware sampling; supported by a citation to [49], not an in-paper proof.
  • domain assumption LLaVA-Next-Video provides sufficiently accurate clip-level captions
    Used in the data engine for XD-Violence clips in Sec 3.2 and Sec A.2; no manual verification rate reported.
  • domain assumption LLaMA3-70B summarization produces reliable event and video summaries after manual checking
    The whole benchmark's text annotations depend on this; only 'manual review' is mentioned without quantitative agreement, Sec 3.2.
  • domain assumption The manual event boundaries from five annotators are correct and consistent
    These boundaries generate the frame-level labels used to train the anomaly scorer, Sec 3.2 and Sec 4.3.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Holmes-VAU: Towards Long-term Video Anomaly Understanding at Any Granularity." pith.science (2026). https://pith.science/paper/UFOUKWYK

@misc{pith2026241206171,
  author       = {Pith},
  title        = {Pith review of: Holmes-VAU: Towards Long-term Video Anomaly Understanding at Any Granularity},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/UFOUKWYK}},
  note         = {Machine review of arXiv:2412.06171}
}
read the original abstract

How can we enable models to comprehend video anomalies occurring over varying temporal scales and contexts? Traditional Video Anomaly Understanding (VAU) methods focus on frame-level anomaly prediction, often missing the interpretability of complex and diverse real-world anomalies. Recent multimodal approaches leverage visual and textual data but lack hierarchical annotations that capture both short-term and long-term anomalies. To address this challenge, we introduce HIVAU-70k, a large-scale benchmark for hierarchical video anomaly understanding across any granularity. We develop a semi-automated annotation engine that efficiently scales high-quality annotations by combining manual video segmentation with recursive free-text annotation using large language models (LLMs). This results in over 70,000 multi-granular annotations organized at clip-level, event-level, and video-level segments. For efficient anomaly detection in long videos, we propose the Anomaly-focused Temporal Sampler (ATS). ATS integrates an anomaly scorer with a density-aware sampler to adaptively select frames based on anomaly scores, ensuring that the multimodal LLM concentrates on anomaly-rich regions, which significantly enhances both efficiency and accuracy. Extensive experiments demonstrate that our hierarchical instruction data markedly improves anomaly comprehension. The integrated ATS and visual-language model outperform traditional methods in processing long videos. Our benchmark and model are publicly available at https://github.com/pipixin321/HolmesVAU.

Figures

Figures reproduced from arXiv: 2412.06171 by the authors.

Figure 1
Figure 1. Motivation. Left: Existing datasets lack the hierarchical structure to capture transient and sustained anomalies across varying temporal scales. Our HIVAU-70k dataset addresses this by providing multi-granularity annotations—clip, event, and video levels—that enable detailed anomaly analysis in complex real-world scenarios. Right: Inspired by Sherlock Holmes’s knack for zeroing in on critical details, our Holmes-VAU… view at source ↗
Figure 2
Figure 2. Data Engine. We present a structured workflow for generating hierarchical annotations across video, event, and clip levels. Clips are first captioned, then processed through a large language model (LLM) with prompts for event summarization. The outputs include clip captions, event summaries, and video summaries, followed by manual checking and refinement. This multi-step approach enriches the dataset with detailed j… view at source ↗
Figure 3
Figure 3. HIVAU-70k dataset. (a) Duration distributions for clips, events, and full videos, showing dominance of short clips. (b) Hierarchical data organization from clip-level to video-level, enabling perception-to-reasoning insights. (c) Word count variations across annotation levels, with more detailed descriptions at the video level. (d) Sample annotations capturing captioning, judgment, description, and anomaly analysis,… view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Holmes-VAU: a multi-modal-LLM-based video anomaly detection framework with adaptive anomaly focus. integrating it into the VLM, and fine-tuning it via instruc￾tion on HIVAU-70k to form our Holmes-VAU model. 4.1. Pipeline Overview The overall pipeline of our Holmes-VAU …
Figure 5
Figure 5. Figure 5: Qualitative comparison of anomaly understanding explanation. Compared with state-of-the-art general MLLMs, i.e., In￾ternVL2 [6] and QwenVL2 [53], our proposed Holmes-VAU demonstrates more accurate anomaly judgment, along with more detailed and comprehensive anomaly-rel…
Figure 6
Figure 6. Figure 6: Ablation study of trainable parameters. (a) Loss curve during instruction-tuning. (b) We tuned the LoRA [15] pa￾rameter r to control trainable parameters, evaluating its impact on VAU capability, and General performance on Video-MME [12]. tion. Our proposed ATS mitigat…

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 6 Pith papers

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

  1. From Detection to Understanding: TAR and TAR-Bench for Multi-Task Traffic Anomaly Reasoning

    cs.CV 2026-08 conditional novelty 6.0 of 10

    TAR and TAR-Bench provide a ten-task traffic anomaly reasoning dataset and benchmark, and fine-tuning on the multi-task chain-of-thought data raises VLM mean scores by about 21 points.

  2. VAU-R1: Advancing Video Anomaly Understanding via Reinforcement Fine-Tuning

    cs.CV 2025-05 reject novelty 6.0 of 10

    VAU-R1 uses Group Relative Policy Optimization with accuracy, format, and temporal-IoU rewards to improve video anomaly reasoning on a new LLM-generated benchmark, VAU-Bench.

  3. Vad-R1: Towards Video Anomaly Reasoning via Perception-to-Cognition Chain-of-Thought

    cs.CV 2025-05 conditional novelty 6.0 of 10

    A two-stage trained MLLM with a perception-to-cognition chain-of-thought and a self-verification RL reward outperforms prior models on video anomaly detection and reasoning.

  4. Visual Anomaly Detection under Complex View-Illumination Interplay: A Large-Scale Benchmark

    cs.CV 2025-05 conditional novelty 6.0 of 10

    M2AD, a large-scale benchmark with 120 view-illumination configurations per object, shows that state-of-the-art visual anomaly detection methods drop markedly when viewpoint and lighting vary together.

  5. Uncertainty-Weighted Image-Event Multimodal Fusion for Video Anomaly Detection

    cs.CV 2025-05 conditional novelty 4.0 of 10

    IEF-VAD fuses CLIP image and synthetic-event features via learned inverse-variance weighting with Kalman-style updates and iterative refinement, reporting state-of-the-art AUC/AP on UCF-Crime, XD-Violence, ShanghaiTec...

  6. Quo Vadis, Anomaly Detection? LLMs and VLMs in the Spotlight

    cs.CV 2024-12 conditional novelty 2.0 of 10

    A survey of 13 recent LLM/VLM-based video anomaly detection methods, organized by interpretability, temporal modeling, few-shot learning, and open-world detection.

Reference graph

Works this paper leans on

128 extracted references · 64 canonical work pages · cited by 6 Pith papers

  1. [1]

    Robust real-time unusual event detection using mul- tiple fixed-location monitors

    Amit Adam, Ehud Rivlin, Ilan Shimshoni, and Daviv Reinitz. Robust real-time unusual event detection using mul- tiple fixed-location monitors. IEEE transactions on pattern analysis and machine intelligence, 30(3):555–560, 2008. 2

    2008

  2. [2]

    Llama 3 model card

    AI@Meta. Llama 3 model card. 2024. 3, 13

    2024

  3. [3]

    Hiervl: Learning hierarchical video- language embeddings

    Kumar Ashutosh, Rohit Girdhar, Lorenzo Torresani, and Kristen Grauman. Hiervl: Learning hierarchical video- language embeddings. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 23066–23078, 2023. 3

    2023

  4. [4]

    Meteor: An automatic metric for mt evaluation with improved correlation with hu- man judgments

    Satanjeev Banerjee and Alon Lavie. Meteor: An automatic metric for mt evaluation with improved correlation with hu- man judgments. In Proceedings of the acl workshop on in- trinsic and extrinsic evaluation measures for machine trans- lation and/or summarization, pages 65–72, 2005. 6, 7

    2005

  5. [5]

    Mgfn: Magnitude- contrastive glance-and-focus network for weakly-supervised video anomaly detection

    Yingxian Chen, Zhengzhe Liu, Baoheng Zhang, Wilton Fok, Xiaojuan Qi, and Yik-Chung Wu. Mgfn: Magnitude- contrastive glance-and-focus network for weakly-supervised video anomaly detection. In Proceedings of the AAAI Con- ference on Artificial Intelligence, pages 387–395, 2023. 6

    2023

  6. [6]

    How far are we to gpt-4v? closing the gap to commercial multimodal models with open-source suites

    Zhe Chen, Weiyun Wang, Hao Tian, Shenglong Ye, Zhang- wei Gao, Erfei Cui, Wenwen Tong, Kongzhi Hu, Jiapeng Luo, Zheng Ma, et al. How far are we to gpt-4v? closing the gap to commercial multimodal models with open-source suites. arXiv preprint arXiv:2404.16821, 2024. 2, 5, 6, 7, 8, 17

    arXiv 2024

  7. [7]

    Videollama 2: Advancing spatial- temporal modeling and audio understanding in video-llms

    Zesen Cheng, Sicong Leng, Hang Zhang, Yifei Xin, Xin Li, Guanzheng Chen, Yongxin Zhu, Wenqi Zhang, Ziyang Luo, Deli Zhao, et al. Videollama 2: Advancing spatial- temporal modeling and audio understanding in video-llms. arXiv preprint arXiv:2406.07476, 2024. 7

    arXiv 2024

  8. [8]

    Adaptive sparse mem- ory networks for efficient and robust video object segmenta- tion

    Jisheng Dang, Huicheng Zheng, Xiaohao Xu, Longguang Wang, Qingyong Hu, and Yulan Guo. Adaptive sparse mem- ory networks for efficient and robust video object segmenta- tion. IEEE Transactions on Neural Networks and Learning Systems, 2024. 3

    2024

Show all 128 references
  1. [9]

    Uncovering what why and how: A comprehensive benchmark for causation understanding of video anomaly

    Hang Du, Sicheng Zhang, Binzhu Xie, Guoshun Nan, Ji- ayang Zhang, Junrui Xu, Hangyu Liu, Sicong Leng, Jiang- ming Liu, Hehe Fan, et al. Uncovering what why and how: A comprehensive benchmark for causation understanding of video anomaly. In Proceedings of the IEEE/CVF Conferenc...

    2024

  2. [10]

    Mmbench-video: A long-form multi-shot benchmark for holistic video under- standing

    Xinyu Fang, Kangrui Mao, Haodong Duan, Xiangyu Zhao, Yining Li, Dahua Lin, and Kai Chen. Mmbench-video: A long-form multi-shot benchmark for holistic video under- standing. arXiv preprint arXiv:2406.14515, 2024. 3

    2024 arXiv

  3. [11]

    Mist: Multiple instance self-training framework for video anomaly detection

    Jia-Chang Feng, Fa-Ting Hong, and Wei-Shi Zheng. Mist: Multiple instance self-training framework for video anomaly detection. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages 14009– 14018, 2021. 2, 6

    2021

  4. [12]

    Video-mme: The first-ever compre- hensive evaluation benchmark of multi-modal llms in video analysis

    Chaoyou Fu, Yuhan Dai, Yondong Luo, Lei Li, Shuhuai Ren, Renrui Zhang, Zihan Wang, Chenyu Zhou, Yunhang Shen, Mengdan Zhang, et al. Video-mme: The first-ever compre- hensive evaluation benchmark of multi-modal llms in video analysis. arXiv preprint arXiv:2405.21075, 2024. 3, 8

    2024 arXiv

  5. [13]

    Memorizing normality to detect anomaly: Memory-augmented deep autoencoder for unsupervised anomaly detection

    Dong Gong, Lingqiao Liu, Vuong Le, Budhaditya Saha, Moussa Reda Mansour, Svetha Venkatesh, and Anton van den Hengel. Memorizing normality to detect anomaly: Memory-augmented deep autoencoder for unsupervised anomaly detection. In Proceedings of the IEEE/CVF Inter- national Con...

  6. [14]

    Learning temporal reg- ularity in video sequences

    Mahmudul Hasan, Jonghyun Choi, Jan Neumann, Amit K Roy-Chowdhury, and Larry S Davis. Learning temporal reg- ularity in video sequences. In Proceedings of the IEEE con- ference on computer vision and pattern recognition , pages 733–742, 2016. 2, 6

    2016

  7. [15]

    Lora: Low-rank adaptation of large language models

    Edward J Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen- Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen. Lora: Low-rank adaptation of large language models. arXiv preprint arXiv:2106.09685, 2021. 6, 8

    2021 arXiv

  8. [16]

    Video re- cap: Recursive captioning of hour-long videos

    Md Mohaiminul Islam, Ngan Ho, Xitong Yang, Tushar Na- garajan, Lorenzo Torresani, and Gedas Bertasius. Video re- cap: Recursive captioning of hour-long videos. In Proceed- ings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 18198–18208, 2024. 3

    2024

  9. [17]

    Clip-tsa: Clip-assisted temporal self-attention for weakly-supervised video anomaly detection

    Hyekang Kevin Joo, Khoa V o, Kashu Yamazaki, and Ngan Le. Clip-tsa: Clip-assisted temporal self-attention for weakly-supervised video anomaly detection. In 2023 IEEE International Conference on Image Processing (ICIP), pages 3230–3234. IEEE, 2023. 3, 6

    2023

  10. [18]

    Coarse-fine net- works for temporal activity detection in videos

    Kumara Kahatapitiya and Michael S Ryoo. Coarse-fine net- works for temporal activity detection in videos. In Proceed- ings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 8385–8394, 2021. 3

    2021

  11. [19]

    Observe locally, infer globally: a space-time mrf for detecting abnormal activi- ties with incremental updates

    Jaechul Kim and Kristen Grauman. Observe locally, infer globally: a space-time mrf for detecting abnormal activi- ties with incremental updates. In 2009 IEEE conference on computer vision and pattern recognition, pages 2921–2928. IEEE, 2009. 2

    2009

  12. [20]

    Anomaly locality in video surveillance

    Federico Landi, Cees GM Snoek, and Rita Cucchiara. Anomaly locality in video surveillance. arXiv preprint arXiv:1901.10364, 2019. 2

    1901 arXiv

  13. [21]

    Videochat: Chat-centric video understanding

    KunChang Li, Yinan He, Yi Wang, Yizhuo Li, Wenhai Wang, Ping Luo, Yali Wang, Limin Wang, and Yu Qiao. Videochat: Chat-centric video understanding. arXiv preprint arXiv:2305.06355, 2023. 2, 4, 7

    2023 arXiv

  14. [22]

    Self-training multi- sequence learning with transformer for weakly supervised video anomaly detection

    Shuo Li, Fang Liu, and Licheng Jiao. Self-training multi- sequence learning with transformer for weakly supervised video anomaly detection. In Proceedings of the AAAI Con- ference on Artificial Intelligence, pages 1395–1403, 2022. 2, 6

    2022

  15. [23]

    Anomaly detection and localization in crowded scenes.IEEE transactions on pattern analysis and machine intelligence , 36(1):18–32, 2013

    Weixin Li, Vijay Mahadevan, and Nuno Vasconcelos. Anomaly detection and localization in crowded scenes.IEEE transactions on pattern analysis and machine intelligence , 36(1):18–32, 2013. 2 9

    2013

  16. [24]

    Resound: To- wards action recognition without representation bias

    Yingwei Li, Yi Li, and Nuno Vasconcelos. Resound: To- wards action recognition without representation bias. InPro- ceedings of the European Conference on Computer Vision (ECCV), pages 513–528, 2018. 3

    2018

  17. [25]

    Video-llava: Learning united visual represen- tation by alignment before projection

    Bin Lin, Bin Zhu, Yang Ye, Munan Ning, Peng Jin, and Li Yuan. Video-llava: Learning united visual represen- tation by alignment before projection. arXiv preprint arXiv:2311.10122, 2023. 2, 4, 7

    2023 arXiv

  18. [26]

    Rouge: A package for automatic evaluation of summaries

    Chin-Yew Lin. Rouge: A package for automatic evaluation of summaries. In Text summarization branches out , pages 74–81, 2004. 6, 7

    2004

  19. [27]

    Improved baselines with visual instruction tuning

    Haotian Liu, Chunyuan Li, Yuheng Li, and Yong Jae Lee. Improved baselines with visual instruction tuning. arXiv preprint arXiv:2310.03744, 2023. 2, 4, 6

    2023 arXiv

  20. [28]

    Llava-next: Im- proved reasoning, ocr, and world knowledge, 2024

    Haotian Liu, Chunyuan Li, Yuheng Li, Bo Li, Yuanhan Zhang, Sheng Shen, and Yong Jae Lee. Llava-next: Im- proved reasoning, ocr, and world knowledge, 2024. 3

    2024

  21. [29]

    Visual instruction tuning

    Haotian Liu, Chunyuan Li, Qingyang Wu, and Yong Jae Lee. Visual instruction tuning. Advances in neural information processing systems, 36, 2024. 2, 4

    2024

  22. [30]

    Exploring background-bias for anomaly detection in surveillance videos

    Kun Liu and Huadong Ma. Exploring background-bias for anomaly detection in surveillance videos. In Proceedings of the 27th ACM International Conference on Multimedia , pages 1490–1499, 2019. 2

    2019

  23. [31]

    Fu- ture frame prediction for anomaly detection–a new baseline

    Wen Liu, Weixin Luo, Dongze Lian, and Shenghua Gao. Fu- ture frame prediction for anomaly detection–a new baseline. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 6536–6545, 2018. 2

    2018

  24. [32]

    Fineaction: A fine-grained video dataset for temporal action localization

    Yi Liu, Limin Wang, Yali Wang, Xiao Ma, and Yu Qiao. Fineaction: A fine-grained video dataset for temporal action localization. IEEE transactions on image processing , 31: 6937–6950, 2022. 3

    2022

  25. [33]

    A hybrid video anomaly detection framework via memory-augmented flow reconstruction and flow-guided frame prediction

    Zhian Liu, Yongwei Nie, Chengjiang Long, Qing Zhang, and Guiqing Li. A hybrid video anomaly detection framework via memory-augmented flow reconstruction and flow-guided frame prediction. In Proceedings of the IEEE/CVF interna- tional conference on computer vision , pages 13588–13597,

  26. [34]

    Abnormal event detec- tion at 150 fps in matlab

    Cewu Lu, Jianping Shi, and Jiaya Jia. Abnormal event detec- tion at 150 fps in matlab. In Proceedings of the IEEE inter- national conference on computer vision , pages 2720–2727,

  27. [35]

    A review of skeleton-based human action recognition

    Jian Lu, Xuanfeng Li, Bo Zhao, and Jian Zhou. A review of skeleton-based human action recognition. Journal of Image and Graphics, 28(12):3651–3669, 2023. 3

    2023

  28. [36]

    Video anomaly detection and explanation via large language models

    Hui Lv and Qianru Sun. Video anomaly detection and explanation via large language models. arXiv preprint arXiv:2401.05702, 2024. 2, 16

    2024 arXiv

  29. [37]

    Unbiased multiple instance learning for weakly supervised video anomaly detection

    Hui Lv, Zhongqi Yue, Qianru Sun, Bin Luo, Zhen Cui, and Hanwang Zhang. Unbiased multiple instance learning for weakly supervised video anomaly detection. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pat- tern Recognition, pages 8022–8031, 2023. 2

    2023

  30. [38]

    Temporal global correlation network for end-to-end action proposal generation

    Baiteng Ma, Shiwei Zhang, Changxin Gao, and Nong Sang. Temporal global correlation network for end-to-end action proposal generation. Acta Electronica Sinica, 50(10):2452– 2461, 2022. 3

    2022

  31. [39]

    Video-chatgpt: Towards detailed video understanding via large vision and language models

    Muhammad Maaz, Hanoona Rasheed, Salman Khan, and Fa- had Shahbaz Khan. Video-chatgpt: Towards detailed video understanding via large vision and language models. In Pro- ceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (ACL 2024), 2024. 7

    2024

  32. [40]

    Ab- normal crowd behavior detection using social force model

    Ramin Mehran, Alexis Oyama, and Mubarak Shah. Ab- normal crowd behavior detection using social force model. In 2009 IEEE conference on computer vision and pattern recognition, pages 935–942. IEEE, 2009. 2

    2009

  33. [41]

    Bleu: a method for automatic evaluation of machine translation

    Kishore Papineni, Salim Roukos, Todd Ward, and Wei-Jing Zhu. Bleu: a method for automatic evaluation of machine translation. In Proceedings of the 40th annual meeting of the Association for Computational Linguistics , pages 311–318,

  34. [42]

    Learning prompt-enhanced context features for weakly- supervised video anomaly detection

    Yujiang Pu, Xiaoyu Wu, Lulu Yang, and Shengjin Wang. Learning prompt-enhanced context features for weakly- supervised video anomaly detection. IEEE Transactions on Image Processing, 2024. 2, 3

    2024

  35. [43]

    Learning from untrimmed videos: Self-supervised video representation learning with hierarchical consistency

    Zhiwu Qing, Shiwei Zhang, Ziyuan Huang, Yi Xu, Xiang Wang, Mingqian Tang, Changxin Gao, Rong Jin, and Nong Sang. Learning from untrimmed videos: Self-supervised video representation learning with hierarchical consistency. In Proceedings of the IEEE/CVF Conference on Computer V...

    2022

  36. [44]

    Learning transferable visual models from natural language supervi- sion

    Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, et al. Learning transferable visual models from natural language supervi- sion. In International conference on machine learning, ...

    2021

  37. [45]

    Finegym: A hierarchical video dataset for fine-grained action under- standing

    Dian Shao, Yue Zhao, Bo Dai, and Dahua Lin. Finegym: A hierarchical video dataset for fine-grained action under- standing. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 2616–2625,

  38. [46]

    Real-world anomaly detection in surveillance videos

    Waqas Sultani, Chen Chen, and Mubarak Shah. Real-world anomaly detection in surveillance videos. In Proceedings of the IEEE conference on computer vision and pattern recog- nition, pages 6479–6488, 2018. 2, 3, 6, 13

    2018

  39. [47]

    Hawk: Learning to understand open-world video anomalies

    Jiaqi Tang, Hao Lu, Ruizheng Wu, Xiaogang Xu, Ke Ma, Cheng Fang, Bin Guo, Jiangbo Lu, Qifeng Chen, and Ying- Cong Chen. Hawk: Learning to understand open-world video anomalies. arXiv preprint arXiv:2405.16886 , 2024. 2, 3, 4, 16, 17

    2024 arXiv

  40. [48]

    Dyannet: A scene dynamicity guided self-trained video anomaly detection net- work

    Kamalakar Vijay Thakare, Yash Raghuwanshi, Debi Prosad Dogra, Heeseung Choi, and Ig-Jae Kim. Dyannet: A scene dynamicity guided self-trained video anomaly detection net- work. In Proceedings of the IEEE/CVF Winter conference on applications of computer vision, pages 5541–5550,...

    2023

  41. [49]

    Weakly-supervised video anomaly detection with robust temporal feature magni- tude learning

    Yu Tian, Guansong Pang, Yuanhong Chen, Rajvinder Singh, Johan W Verjans, and Gustavo Carneiro. Weakly-supervised video anomaly detection with robust temporal feature magni- tude learning. In Proceedings of the IEEE/CVF international conference on computer vision, pages 4975–49...

    2021

  42. [50]

    Exploring diffusion models for unsupervised video anomaly detection

    Anil Osman Tur, Nicola Dall’Asen, Cigdem Beyan, and Elisa Ricci. Exploring diffusion models for unsupervised video anomaly detection. In 2023 IEEE International Conference on Image Processing (ICIP), pages 2540–2544. IEEE, 2023. 2

    2023

  43. [51]

    Cider: Consensus-based image description evalua- tion

    Ramakrishna Vedantam, C Lawrence Zitnick, and Devi Parikh. Cider: Consensus-based image description evalua- tion. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 4566–4575, 2015. 6, 7

    2015

  44. [52]

    Gods: Generalized one-class discriminative subspaces for anomaly detection

    Jue Wang and Anoop Cherian. Gods: Generalized one-class discriminative subspaces for anomaly detection. In Proceed- ings of the IEEE/CVF International Conference on Com- puter Vision, pages 8201–8211, 2019. 2, 6

    2019

  45. [53]

    Qwen2-vl: Enhancing vision-language model’s perception of the world at any resolution

    Peng Wang, Shuai Bai, Sinan Tan, Shijie Wang, Zhihao Fan, Jinze Bai, Keqin Chen, Xuejing Liu, Jialin Wang, Wenbin Ge, Yang Fan, Kai Dang, Mengfei Du, Xuancheng Ren, Rui Men, Dayiheng Liu, Chang Zhou, Jingren Zhou, and Jun- yang Lin. Qwen2-vl: Enhancing vision-language model’s ...

    2024 arXiv

  46. [54]

    Review of action recognition based on multimodal data

    SC Wang, Q Huang, YF Zhang, X Li, YQ Nie, and GC Luo. Review of action recognition based on multimodal data. Im- age Graph, 27(11):3139–3159, 2022. 3

    2022

  47. [55]

    Self-supervised sparse representa- tion for video anomaly detection

    Jhih-Ciang Wu, He-Yen Hsieh, Ding-Jie Chen, Chiou-Shann Fuh, and Tyng-Luh Liu. Self-supervised sparse representa- tion for video anomaly detection. In European Conference on Computer Vision, pages 729–745. Springer, 2022. 2, 5, 6

    2022

  48. [56]

    Not only look, but also listen: Learning multimodal violence detection under weak supervision

    Peng Wu, Jing Liu, Yujia Shi, Yujia Sun, Fangtao Shao, Zhaoyang Wu, and Zhiwei Yang. Not only look, but also listen: Learning multimodal violence detection under weak supervision. In Computer Vision–ECCV 2020: 16th Euro- pean Conference, Glasgow, UK, August 23–28, 2020, Pro- c...

    2020

  49. [57]

    Open-vocabulary video anomaly detection

    Peng Wu, Xuerong Zhou, Guansong Pang, Yujia Sun, Jing Liu, Peng Wang, and Yanning Zhang. Open-vocabulary video anomaly detection. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 18297–18307, 2024. 2, 3, 6

    2024

  50. [58]

    Vadclip: Adapting vision-language models for weakly supervised video anomaly detection

    Peng Wu, Xuerong Zhou, Guansong Pang, Lingru Zhou, Qingsen Yan, Peng Wang, and Yanning Zhang. Vadclip: Adapting vision-language models for weakly supervised video anomaly detection. In Proceedings of the AAAI Con- ference on Artificial Intelligence, pages 6074–6082, 2024. 2, 3, 5, 6

    2024

  51. [59]

    Detecting anomalous events in videos by learning deep representations of appearance and motion

    Dan Xu, Yan Yan, Elisa Ricci, and Nicu Sebe. Detecting anomalous events in videos by learning deep representations of appearance and motion. Computer Vision and Image Un- derstanding, 156:117–127, 2017. 2

    2017

  52. [60]

    To- wards robust video object segmentation with adaptive object calibration

    Xiaohao Xu, Jinglu Wang, Xiang Ming, and Yan Lu. To- wards robust video object segmentation with adaptive object calibration. In Proceedings of the 30th ACM International Conference on Multimedia, pages 2709–2718, 2022. 3

    2022

  53. [61]

    Video event restoration based on keyframes for video anomaly detection

    Zhiwei Yang, Jing Liu, Zhaoyang Wu, Peng Wu, and Xiaotao Liu. Video event restoration based on keyframes for video anomaly detection. In Proceedings of the IEEE/CVF Con- ference on Computer Vision and Pattern Recognition, pages 14592–14601, 2023. 2

    2023

  54. [62]

    Text prompt with nor- mality guidance for weakly supervised video anomaly detec- tion

    Zhiwei Yang, Jing Liu, and Peng Wu. Text prompt with nor- mality guidance for weakly supervised video anomaly detec- tion. In Proceedings of the IEEE/CVF Conference on Com- puter Vision and Pattern Recognition , pages 18899–18908,

  55. [63]

    Dota: Unsupervised de- tection of traffic anomaly in driving videos

    Yu Yao, Xizi Wang, Mingze Xu, Zelin Pu, Yuchen Wang, Ella Atkins, and David J Crandall. Dota: Unsupervised de- tection of traffic anomaly in driving videos. IEEE transac- tions on pattern analysis and machine intelligence , 45(1): 444–459, 2022. 2

    2022

  56. [64]

    Towards surveillance video-and-language understanding: New dataset, baselines, and challenges, 2023

    Tongtong Yuan, Xuange Zhang, Kun Liu, Bo Liu, Chen Chen, Jian Jin, and Zhenzhen Jiao. Towards surveillance video-and-language understanding: New dataset, baselines, and challenges, 2023. 3, 16

    2023

  57. [65]

    Towards surveillance video-and-language understanding: New dataset baselines and challenges

    Tongtong Yuan, Xuange Zhang, Kun Liu, Bo Liu, Chen Chen, Jian Jin, and Zhenzhen Jiao. Towards surveillance video-and-language understanding: New dataset baselines and challenges. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 22052–...

    2024

  58. [66]

    Generative cooperative learning for unsupervised video anomaly detection

    M Zaigham Zaheer, Arif Mahmood, M Haris Khan, Mattia Segu, Fisher Yu, and Seung-Ik Lee. Generative cooperative learning for unsupervised video anomaly detection. In Pro- ceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 14744–14754, 2022. 2, 6

    2022

  59. [67]

    Harnessing large language mod- els for training-free video anomaly detection

    Luca Zanella, Willi Menapace, Massimiliano Mancini, Yim- ing Wang, and Elisa Ricci. Harnessing large language mod- els for training-free video anomaly detection. InProceedings of the IEEE/CVF Conference on Computer Vision and Pat- tern Recognition, pages 18527–18536, 2024. 2, ...

    2024

  60. [68]

    Video-llama: An instruction-tuned audio-visual language model for video un- derstanding

    Hang Zhang, Xin Li, and Lidong Bing. Video-llama: An instruction-tuned audio-visual language model for video un- derstanding. arXiv preprint arXiv:2306.02858, 2023. 2, 7

    2023 arXiv

  61. [69]

    Hr-pro: Point-supervised temporal action localization via hierarchical reliability prop- agation

    Huaxin Zhang, Xiang Wang, Xiaohao Xu, Zhiwu Qing, Changxin Gao, and Nong Sang. Hr-pro: Point-supervised temporal action localization via hierarchical reliability prop- agation. arXiv preprint arXiv:2308.12608, 2023. 3

    2023 arXiv

  62. [70]

    Glancevad: Exploring glance su- pervision for label-efficient video anomaly detection

    Huaxin Zhang, Xiang Wang, Xiaohao Xu, Xiaonan Huang, Chuchu Han, Yuehuan Wang, Changxin Gao, Shanjun Zhang, and Nong Sang. Glancevad: Exploring glance su- pervision for label-efficient video anomaly detection. arXiv preprint arXiv:2403.06154, 2024. 2

    2024 arXiv

  63. [71]

    Video instruction tuning with synthetic data

    Yuanhan Zhang, Jinming Wu, Wei Li, Bo Li, Zejun Ma, Zi- wei Liu, and Chunyuan Li. Video instruction tuning with synthetic data. arXiv preprint arXiv:2410.02713, 2024. 7, 13

    2024 arXiv

  64. [72]

    Online detection of unusual events in videos via dynamic sparse coding

    Bin Zhao, Li Fei-Fei, and Eric P Xing. Online detection of unusual events in videos via dynamic sparse coding. In CVPR 2011, pages 3313–3320. IEEE, 2011. 2

    2011

  65. [73]

    Graph convolutional label noise cleaner: Train a plug-and-play action classifier for anomaly detection

    Jia-Xing Zhong, Nannan Li, Weijie Kong, Shan Liu, Thomas H Li, and Ge Li. Graph convolutional label noise cleaner: Train a plug-and-play action classifier for anomaly detection. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 1237–1246,

  66. [74]

    Dual memory units with uncertainty regulation for weakly supervised video 11 anomaly detection

    Hang Zhou, Junqing Yu, and Wei Yang. Dual memory units with uncertainty regulation for weakly supervised video 11 anomaly detection. In Proceedings of the AAAI Conference on Artificial Intelligence , pages 3769–3777, 2023. 2, 5, 6, 14, 15 12 Holmes-V AU: Towards Long-term Vide...

    2023

  67. [75]

    ”Describe the video briefly.”

  68. [76]

    ”Describe the main events that take place in this video.”

  69. [77]

    ”Give a short description of the video.”

  70. [78]

    ”What happened in this video?”

  71. [79]

    ”Generate a brief caption for the video.”

  72. [80]

    ”Can you provide a brief description of the video?”

  73. [81]

    ”Briefly describe the main subjects and their actions in the video.”

  74. [82]

    ”Provide a short overview of what happens in the video?”

  75. [83]

    ”Describe the key moments that showcase the subjects’ activities in the video.”

  76. [84]

    ”Describe the sequence of events involving the main subjects in the video.”

  77. [85]

    ”What activities happen throughout the video?”

  78. [86]

    ”Describe the main subjects and their roles in the video.”

  79. [87]

    ”What key moments stand out in the video?”

  80. [88]

    ”What are the primary activities showcased in the video?”

  81. [89]

    ”What happens to the main subjects as the video progresses?”

  82. [90]

    ”What is a brief overview of what happens in the video?”

  83. [91]

    ”Describe the main subjects and their contributions to the video.”

  84. [92]

    ”Describe the key events in the video.”

  85. [93]

    ”Describe the video’s main activities.”

  86. [94]

    ”Can you describe the main action in this video briefly?”

  87. [95]

    ”Describe the video clip concisely.”

  88. [96]

    ”Provide a brief description of the given video clip.”

  89. [97]

    ”Summarize the visual content of the video clip.”

  90. [98]

    ”Give a short and clear explanation of the subsequent video clip.” Judgement

  91. [99]

    ”What types of anomalies are shown in the video clip?”

  92. [100]

    ”Are there any anomaly events detected in the video?”

  93. [101]

    ”Detect and classify the anomaly events in the video.”

  94. [102]

    ”Identify any abnormal behaviors depicted in the video.”

  95. [103]

    ”Determine whether there are anomaly events in the video and the specific name of the anomaly.”

  96. [104]

    ”What anomalies can be identified in the video?”

  97. [105]

    ”What categories of anomalies can be found in the video?”

  98. [106]

    ”Could you point out any abnormal actions in the video?”

  99. [107]

    video":

    ”Point out the abnormal actions in the video.” 13 1 { 2 "video": "v=2rfyeR-YaJw__#1_label_G-0-0", 3 "n_frames": 1940, 4 "fps": 24.0, 5 "label": ["Explosion"], 6 "clips": [[[5.583, 11.903], [11.903, 18.222], [18.222, 24.542]], 7 [[36.167, 43.48], [43.48, 50.792]]], 8 "clip_capt...

    1940

  100. [108]

    ”Are there anomalies observed in the video clip?” Description

  101. [109]

    ”Describe the anomaly events observed in the video.”

  102. [110]

    ”Could you describe the anomaly events observed in the video?”

  103. [111]

    ”Could you specify the anomaly events present in the video?”

  104. [112]

    ”Give a description of the detected anomaly events in this video.”

  105. [113]

    ”Could you give a description of the anomaly events in the video?”

  106. [114]

    ”Provide a summary of the anomaly events in the video.”

  107. [115]

    ”Could you provide a summary of the anomaly events in this video?””

  108. [116]

    ”What details can you provide about the anomaly in the video?”

  109. [117]

    ”How would you detail the anomaly events found in the video?”

  110. [118]

    ”How would you describe the particular anomaly events in the video?” Analysis

  111. [119]

    ”Why do you judge this event to be anomalous?”

  112. [120]

    ”Can you provide the reasons for considering it anomalous?”

  113. [121]

    ”Can you give the basis for your judgment of this event as an anomaly?”

  114. [122]

    ”What led you to classify this event as an anomaly?”

  115. [123]

    ”Could you provide the reasons for considering this event as abnormal?”

  116. [124]

    ”What evidence do you have to support your judgment of this event as an anomaly?”

  117. [125]

    ”Can you analyze the factors contributing to this anomalous event?”

  118. [126]

    ”Could you share your analysis of the anomalous event?”

  119. [127]

    ”What patterns did you observe that contributed to your conclusion about this event being an anomaly?”

  120. [128]

    id": "Abuse007_x264_E0C2

    ”How do the characteristics of this event support its classification as an anomaly?” A.4. Data Samples. To facilitate understanding, we provide the final constructed instruction data at various temporal granularities, including clip-level, event-level, and video-level, as show...

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.