Pith. sign in

REVIEW 6 major objections 6 minor 1 cited by

Datasets and Recipes for Video Temporal Grounding via Reinforcement Learning

T0 review · 6 major / 6 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read A two-stage recipe of filtered supervised fine-tuning followed by reinforcement learning lets a 7B open video model outperform prior temporal grounding models on three benchmarks.

desk verdict A genuinely useful data-recipe paper for temporal grounding that releases open resources and runs clean ablations, but its headline comparison omits the two closest RL baselines and the 'consistently outperforms' claim is broader than the evidence supports. read the letter →

arxiv 2507.18100 v1 pith:3LATIR55 submitted 2025-07-24 cs.CV cs.AI

Ruizhe Chen , Zhiting Fan , Tianze Luo , Heqing Zou , Zhaopeng Feng , Guiyang Xie , Hansheng Zhang , Zhuochen Wang
show 2 more authors
Zuozhu Liu Huaijian Zhang
This is my paper
classification cs.CVcs.AI
keywords videotemporalgroundingreinforcementlearningGRPOchain-of-thoughtsupervisedfine-tuningIoUrewarddatasetcurationlargevision-languagemodels
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 argues that video temporal grounding in open 7B vision-language models can be made state-of-the-art by a two-stage training recipe rather than by larger models or more data. The first stage is a cold-start supervised fine-tuning on a small, high-quality set of chain-of-thought examples whose predicted segments overlap the ground truth by more than 0.8 IoU. The second stage applies reinforcement learning on a difficulty-filtered set, with samples below 0.4 IoU removed, rewarding temporal overlap plus correct reasoning format. On three benchmarks the final model beats prior instruction-tuned baselines, and the authors release the curated datasets, models, and code.

What carries the argument

The load-bearing mechanism is an IoU-gated data-and-reward pipeline. A high-capacity teacher model writes chain-of-thought rationales for a pooled set of moment-retrieval and query-grounding samples; only samples whose predicted interval overlaps the ground truth by more than 0.8 IoU enter the cold-start SFT set, and samples with IoU below 0.4 are excluded from RL. Training then runs Group Relative Policy Optimization, a PPO variant that normalizes rewards inside a sampled group of responses, on a reward $r_i = \lambda_{\text{tIoU}} \cdot r_{\text{tIoU}} + \lambda_{\text{form}} \cdot r_{\text{form}}$, where $r_{\text{tIoU}}$ is the temporal Intersection-over-Union and $r_{\text{form}}$ checks for the required reasoning tags. The cold start provides the initial formatting and reasoning behavior that RL then amplifies.

What would settle it

Run the exact two-stage recipe with the teacher's chain-of-thought rationales replaced by plain timestamp answers in the cold-start set; if final R@0.5 scores do not drop on all three benchmarks, the claim that explicit reasoning drives the gains is false.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that explicit, verifiable reasoning can be injected into a video temporal grounding model through reinforcement learning, provided the model starts from a carefully filtered cold start. Supervised fine-tuning alone on the same data does not consistently help and can even regress; the gains appear only after RL, and they are larger when the cold-start data are filtered for high IoU and when the RL data exclude very hard or mislabeled samples. The reward that carries this is a weighted combination of Intersection-over-Union between the predicted and ground-truth temporal segment and a binary check that the response contains the required reasoning and answer tags. With this recipe, a 7B open model reaches the reported top R@0.3 and mean IoU numbers on three temporal grounding benchmarks.

Load-bearing premise

The whole recipe assumes the ground-truth segments in the pooled source datasets are accurate enough that IoU filtering and the IoU reward are trustworthy signals; noisy annotations would be amplified by both the cold-start filter and the RL reward.

Editorial extensions

If this is right

  • If the recipe generalizes, a small open model can rival much larger or specially designed grounding models without new architectures.
  • The finding that SFT alone can regress while RL after cold start improves suggests that RL, not instruction tuning, is the stage that should absorb scale in this task.
  • Because the data and code are released, the same cold-start-plus-filtered-RL pipeline can be reproduced and adapted to other video-language tasks.
  • The difficulty thresholds of 0.8 and 0.4 give a concrete starting point for tuning data curation in low-resource settings.

Reading between the lines

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

  • A natural extension would make the difficulty filter a curriculum: instead of dropping samples below 0.4 IoU, train on them later or down-weight them, which may improve hard-sample robustness.
  • The same IoU-gated reward could be applied to other dense prediction tasks with a verifiable overlap metric, such as referring video segmentation or action segmentation.
  • Because the teacher rationales are filtered by IoU against source annotations, the recipe may silently under-represent genuinely hard cases where the teacher is right but the source segment is imprecise; auditing failures on such cases would be a useful stress test.
  • The released RL set is small, so a natural question is how performance scales as the filtered pool grows; the reported curves suggest diminishing returns may set in differently with and without cold start.
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

6 major / 6 minor

Summary. The paper proposes a two-stage training recipe for video temporal grounding (VTG) with an open 7B LVLM (Qwen2.5-VL-7B): first, an SFT cold start on 13K Gemini-2.5-Pro-generated chain-of-thought samples filtered to IoU > 0.8 against source ground truth; second, GRPO reinforcement learning with a weighted IoU-plus-format reward on an 18K subset that excludes samples with IoU < 0.4. The paper reports results on NExT-GQA, ReXTime, and Charades-STA, ablates the cold-start and RL-data filtering choices, and releases data, code, and models.

Significance. If the recipe holds, the contribution is practically valuable: it provides an open, reproducible pipeline for RL-based temporal grounding on a 7B backbone, with ablations that isolate two design choices (cold-start quality and RL difficulty filtering) that are often confounded in prior work. The release of intermediate datasets and checkpoints enables direct comparison and further research. However, the significance is currently limited by missing comparisons with the closest RL-based baselines, an unexplained SFT degradation, and internal inconsistencies between the main tables and the ablation tables.

major comments (6)
  1. [Tables 3 and 4 vs Tables 2 and 5] The mIoU values reported for the same model configurations are inconsistent between the main results and the ablations. For example, TVG-ColdStart is reported with NExT-GQA mIoU 15.09 and ReXTime mIoU 10.18 in Table 2, but with 26.14 and 26.26 in Table 3; TVG-R1 appears as 29.25/28.20/46.73 in Tables 2 and 5 and as 30.41/26.38/48.78 in Tables 3 and 4. Since Tables 3 and 4 are the central evidence for the filtering claims, this discrepancy must be resolved (e.g., by stating which checkpoint, split, or seed each table uses) or the conclusions drawn from those tables are unsupported.
  2. [Section 4.2, Tables 2 and 5] The claim that VTG-R1 'consistently outperforms previous models on most benchmarks' is not supported relative to the closest RL-based baselines. Time-R1 [42] and Video-R1 [12] are cited in Section 2 but never evaluated, and on NExT-GQA VTG-R1's R@0.5 of 20.78 is below VideoChat-TPO (23.40) and Qwen2.5-VL-32B (22.26). Please add these comparisons or restrict the claim to the evaluated baseline set.
  3. [Section 4.2, Tables 2 and 5] The SFT cold start degrades performance substantially below the base model on all three benchmarks (e.g., Table 2: NExT-GQA mIoU 15.09 vs 20.87 for Qwen2.5-VL-7B; Table 5: Charades-STA R@0.3 42.23 vs 68.98). The paper's explanation ('possibly due to overfitting or limited generalization') is speculative and not backed by analysis. Because the recipe's first stage is claimed to provide 'robust multimodal alignment and structured reasoning,' the manuscript should analyze this degradation (e.g., output-format adherence, per-category errors, training loss) or reframe the cold start as a necessary initialization for RL rather than a performance-improving stage.
  4. [Section 4.1 and Tables 2-5] All experiments appear to be single runs with no error bars or significance testing. Several reported gains are small (e.g., Charades-STA R@0.3: VTG-R1 70.75 vs Base 68.98; mIoU 46.73 vs 46.14). Without multiple seeds per configuration, the stability of the recipe's improvements cannot be assessed. Please report repeated runs (at least 2-3 seeds) for the main comparisons.
  5. [Section 3.1, Appendix A.1] The data filtering relies on thresholds ϵ1=0.8 and ϵ2=0.4 applied to the IoU between Gemini-2.5-Pro's predicted segment and the source ground truth. This conflates difficulty with annotation noise, and the source datasets include weakly supervised collections (e.g., TACoS, DiDeMo, InternVid-VTime). The thresholds are fixed with no sensitivity analysis, and the CoT annotations are not validated (e.g., by human evaluation). Since the ablations show filtering is a key ingredient, a threshold sensitivity study or annotation-quality check is needed to support the recipe.
  6. [Section 3.1 vs Section 3.3.1] The reward specification is inconsistent with the prompt template: the template requires the time span to be enclosed in ⟨time⟩...⟨/time⟩ tags, while the format reward rform checks for ⟨think⟩, ⟨/think⟩, ⟨answer⟩, and ⟨/answer⟩ tags and does not mention ⟨time⟩. Please clarify which format is actually used during training and align the description.
minor comments (6)
  1. [Section 4.1] The metric is written 'R1@m' but the standard notation in temporal grounding is R@m; please define it consistently and use a single notation throughout.
  2. [References] References [19], [20], and [21] all cite the same DiDeMo paper, while [28] (used for NExT-GQA) is listed as a VideoChat paper; please correct and de-duplicate the bibliography.
  3. [Section 3.1] The prompt template has formatting issues (e.g., 'system You MUST reason...' without spacing) and the placeholder {query / instance} is not defined.
  4. [Figure 2] The three panels do not indicate the number of runs or seeds, and the y-axis ranges make the differences appear larger than they are; please add error bars or state explicitly that single curves are shown.
  5. [Appendix A.1] β is set to 0.0, which removes the KL regularization from the GRPO objective; the manuscript should comment on whether the reference policy is then used at all and on the implications for training stability.
  6. [Tables 3 and 4] The column headers 'NExTG.', 'RexT.', 'Charad.' should be defined (presumably mIoU); please clarify in the captions.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the two-stage SFT+RL recipe is validated on external VTG benchmarks, and the IoU reward used in training is a standard objective rather than a fitted predictor.

full rationale

The paper's derivation chain is self-contained: it trains Qwen2.5-VL-7B with a cold-start SFT stage on Gemini-annotated CoT data filtered by IoU, followed by GRPO with a composite IoU-plus-format reward, and evaluates on three external benchmarks (NExT-GQA, ReXTime, Charades-STA) that are not among the training sources listed in Table 1 (HiREST, QuerYD, TACoS, DiDeMo, InternVid-VTime, Grounded-VLLM). The fact that the RL reward rtIoU is the same IoU metric used for evaluation is standard reward specification, not a fitted parameter renamed as a prediction. The central claims are supported by held-out benchmark numbers rather than by construction. The 'high-quality' cold-start definition (IoU > 0.8) is a data-selection criterion, and the ablated comparisons of filtered vs unfiltered data are empirical, not definitional identities. Self-citations in the related-work section (refs 4-8, 10, 11, 13) are descriptive context for RLHF/RLVR and are not load-bearing for the VTG results. No uniqueness theorem or ansatz is imported from the authors' prior work. The main weaknesses, such as omitted RL-style temporal-grounding baselines (Time-R1, Video-R1) and single-run results without error bars, are comparison and completeness concerns, not circularity.

Assumptions & free parameters 6 free parameters · 5 assumptions · 0 invented entities

The recipe depends on hand-chosen thresholds (epsilon_1, epsilon_2), reward weights (lambda_tIoU, lambda_form), group size G, and KL coefficient beta. The main domain assumptions are the accuracy of source ground truth and of Gemini-generated CoT rationales. No new entities are invented.

free parameters (6)
  • epsilon_1 (coldstart IoU threshold) = 0.8
    Samples with Gemini-generated prediction IoU > 0.8 are kept for the coldstart SFT set; chosen by hand, not swept.
  • epsilon_2 (RL data exclusion threshold) = 0.4
    Samples with IoU < 0.4 are excluded from RL data as too hard or erroneous; chosen by hand.
  • lambda_tIoU (IoU reward weight) = 0.9
    Weight of the IoU reward in the composite reward.
  • lambda_form (format reward weight) = 0.1
    Weight of the reasoning format reward.
  • beta (KL regularization coefficient) = 0.0
    KL divergence coefficient in the GRPO objective; set to 0, meaning no KL constraint against the reference policy.
  • GRPO group size G = 8
    Number of sampled responses per query in RL.
assumptions (5)
  • domain assumption Ground-truth temporal segments in the source datasets are correct and complete.
    Used in IoU filtering (Section 3.1) and IoU reward (Section 3.3.1). If noisy, filtering and reward propagate errors.
  • domain assumption Gemini-2.5-Pro generated CoT rationales accurately reflect temporal evidence in the video.
    Coldstart SFT trains on these rationales (Section 3.1); inaccurate rationales would teach wrong reasoning patterns.
  • domain assumption IoU with a single ground-truth segment is an adequate reward signal for temporal grounding.
    The composite reward is dominated by IoU (Section 3.3.1); this metric may not capture semantic correctness of reasoning.
  • standard math The GRPO objective with the stated loss function is valid.
    GRPO from DeepSeekMath [37]; standard RL algorithm, no new derivation.
  • domain assumption The evaluation benchmarks (NExT-GQA, ReXTime, Charades-STA) measure temporal grounding faithfully.
    Used for all main results (Section 4.2).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Datasets and Recipes for Video Temporal Grounding via Reinforcement Learning." pith.science (2026). https://pith.science/paper/3LATIR55

@misc{pith2026250718100,
  author       = {Pith},
  title        = {Pith review of: Datasets and Recipes for Video Temporal Grounding via Reinforcement Learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/3LATIR55}},
  note         = {Machine review of arXiv:2507.18100}
}
read the original abstract

Video Temporal Grounding (VTG) aims to localize relevant temporal segments in videos given natural language queries. Despite recent progress with large vision-language models (LVLMs) and instruction-tuning, existing approaches often suffer from limited temporal awareness and poor generalization. In this work, we introduce a two-stage training framework that integrates supervised fine-tuning with reinforcement learning (RL) to improve both the accuracy and robustness of VTG models. Our approach first leverages high-quality curated cold start data for SFT initialization, followed by difficulty-controlled RL to further enhance temporal localization and reasoning abilities. Comprehensive experiments on multiple VTG benchmarks demonstrate that our method consistently outperforms existing models, particularly in challenging and open-domain scenarios. We conduct an in-depth analysis of training strategies and dataset curation, highlighting the importance of both high-quality cold start data and difficulty-controlled RL. To facilitate further research and industrial adoption, we release all intermediate datasets, models, and code to the community.

Discussion (0). Sign in to comment.

Forward citations

Cited by 1 Pith paper

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

  1. TAR: Temporal Anchor-Constrained Reasoning for Video Temporal Grounding

    cs.CV 2025-08 conditional novelty 7.0 of 10

    A reinforcement-learning method that forces video grounding models to emit progressively more accurate intermediate timestamps, improving accuracy and reasoning faithfulness without large teacher models.

Reference graph

Works this paper leans on

46 extracted references · 22 canonical work pages · cited by 1 Pith paper

  1. [42]

    Time-r1: Post-training large vision language model for temporal video grounding

    Ye Wang, Ziheng Wang, Boshen Xu, Yang Du, Kejun Lin, Zihan Xiao, Zihao Yue, Jianzhong Ju, Liang Zhang, Dingyi Yang, et al. Time-r1: Post-training large vision language model for temporal video grounding. arXiv preprint arXiv:2503.13377, 2025

    arXiv 2025

  2. [12]

    Video-r1: Reinforcing video reasoning in mllms

    Kaituo Feng, Kaixiong Gong, Bohao Li, Zonghao Guo, Yibing Wang, Tianshuo Peng, Junfei Wu, Xiaoying Zhang, Benyou Wang, and Xiangyu Yue. Video-r1: Reinforcing video reasoning in mllms. arXiv preprint arXiv:2503.21776, 2025

    arXiv 2025

  3. [1]

    Localizing moments in video with natural language

    Lisa Anne Hendricks, Oliver Wang, Eli Shechtman, Josef Sivic, Trevor Darrell, and Bryan Russell. Localizing moments in video with natural language. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2017

    work page 2017

  4. [2]

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

    arXiv 2025

  5. [3]

    Training a helpful and harmless assistant with reinforcement learning from human feedback

    Yuntao Bai, Andy Jones, Kamal Ndousse, Amanda Askell, Anna Chen, Nova DasSarma, Dawn Drain, Stanislav Fort, Deep Ganguli, Tom Henighan, et al. Training a helpful and harmless assistant with reinforcement learning from human feedback. arXiv preprint arXiv:2204.05862, 2022

    arXiv 2022

  6. [4]

    Fast model debias with machine unlearning

    Ruizhe Chen, Jianfei Yang, Huimin Xiong, Jianhong Bai, Tianxiang Hu, Jin Hao, Yang Feng, Joey Tianyi Zhou, Jian Wu, and Zuozhu Liu. Fast model debias with machine unlearning. Advances in Neural Information Processing Systems, 36:14516–14539, 2023

    work page 2023

  7. [5]

    Learnable Privacy Neurons Localization in Language Models

    Ruizhe Chen, Tianxiang Hu, Yang Feng, and Zuozhu Liu. Learnable privacy neurons localization in language models. arXiv preprint arXiv:2405.10989, 2024

    work page Pith review arXiv 2024

  8. [6]

    Identifying and Mitigating Social Bias Knowledge in Language Models

    Ruizhe Chen, Yichen Li, Jianfei Yang, Joey Tianyi Zhou, Jian Wu, and Zuozhu Liu. Identifying and mitigating social bias knowledge in language models. arXiv preprint arXiv:2408.11843, 2024

    work page Pith review arXiv 2024

Show all 46 references
  1. [7]

    Pad: Personalized alignment of llms at decoding-time

    Ruizhe Chen, Xiaotian Zhang, Meng Luo, Wenhao Chai, and Zuozhu Liu. Pad: Personalized alignment of llms at decoding-time. arXiv preprint arXiv:2410.04070, 2024

    2024 arXiv

  2. [8]

    Diffpo: Diffusion-styled preference optimization for efficient inference- time alignment of large language models

    Ruizhe Chen, Wenhao Chai, Zhifei Yang, Xiaotian Zhang, Joey Tianyi Zhou, Tony Quek, Soujanya Poria, and Zuozhu Liu. Diffpo: Diffusion-styled preference optimization for efficient inference- time alignment of large language models. arXiv preprint arXiv:2503.04240, 2025

    2025 arXiv

  3. [9]

    Wenliang Dai, Junnan Li, Dongxu Li, Anthony M. H. Tiong, Junqi Zhao, Weisheng Wang, Boyang Li, Pascale Fung, and Steven C. H. Hoi. Instructblip: Towards general-purpose vision-language models with instruction tuning. arXiv preprint arXiv:2305.06500, 2023

    2023 arXiv

  4. [10]

    Fairmt-bench: Benchmarking fairness for multi-turn dialogue in conversational llms

    Zhiting Fan, Ruizhe Chen, Tianxiang Hu, and Zuozhu Liu. Fairmt-bench: Benchmarking fairness for multi-turn dialogue in conversational llms. arXiv preprint arXiv:2410.19317, 2024

    2024 arXiv

  5. [11]

    Biasalert: A plug-and-play tool for social bias detection in llms

    Zhiting Fan, Ruizhe Chen, Ruiling Xu, and Zuozhu Liu. Biasalert: A plug-and-play tool for social bias detection in llms. arXiv preprint arXiv:2407.10241, 2024

    2024 arXiv

  6. [13]

    Mt-r1-zero: Advancing llm-based machine translation via r1-zero-like reinforcement learning

    Zhaopeng Feng, Shaosheng Cao, Jiahan Ren, Jiayuan Su, Ruizhe Chen, Yan Zhang, Zhe Xu, Yao Hu, Jian Wu, and Zuozhu Liu. Mt-r1-zero: Advancing llm-based machine translation via r1-zero-like reinforcement learning. arXiv preprint arXiv:2504.10160, 2025. 10 Datasets and Recipes fo...

    2025 arXiv

  7. [14]

    Temporal sen- tence grounding in streaming videos

    Tian Gan, Xiao Wang, Yan Sun, Jianlong Wu, Qingpei Guo, and Liqiang Nie. Temporal sen- tence grounding in streaming videos. In Proceedings of the 31st ACM International Conference on Multimedia, pages 4637–4646, 2023

    2023

  8. [15]

    Tall: Temporal activity localization via language query

    Jiyang Gao, Chen Sun, Zhenheng Yang, and Ram Nevatia. Tall: Temporal activity localization via language query. In Proceedings of the IEEE International Conference on Computer Vision (ICCV), pages 5277–5285. IEEE, 2017. doi: 10.1109/ICCV .2017.563

    2017 doi

  9. [16]

    Ego4D: Around the world in 3,000 hours of egocentric video

    Kristen Grauman, Andrew Westbury, Eugene Byrne, Zachary Chavis, Antonino Furnari, Rohit Girdhar, Jackson Hamburger, Hao Jiang, Miao Liu, and Xingyu Liu et al. Ego4D: Around the world in 3,000 hours of egocentric video. In Proceedings of the IEEE/CVF Conference on Computer Visi...

    2022

  10. [17]

    Deepseek-r1: Incentivizing reasoning capability in llms via reinforcement learning

    Daya Guo, Dejian Yang, Haowei Zhang, Junxiao Song, Ruoyu Zhang, Runxin Xu, Qihao Zhu, Shirong Ma, Peiyi Wang, Xiao Bi, et al. Deepseek-r1: Incentivizing reasoning capability in llms via reinforcement learning. arXiv preprint arXiv:2501.12948, 2025

    2025 arXiv

  11. [18]

    VTG-LLM: Integrating timestamp knowledge into video LLMs for enhanced video temporal grounding

    Yongxin Guo, Jingyu Liu, Mingda Li, Dingxin Cheng, Xiaoying Tang, Dianbo Sui, Qingbin Liu, Xi Chen, and Kevin Zhao. VTG-LLM: Integrating timestamp knowledge into video LLMs for enhanced video temporal grounding. arXiv preprint arXiv:2405.13382, 2024

    2024 arXiv

  12. [19]

    Rus- sell

    Lisa Anne Hendricks, Oliver Wang, Eli Shechtman, Josef Sivic, Trevor Darrell, and Bryan C. Rus- sell. Localizing moments in video with natural language. In Proceedings of the IEEE International Conference on Computer Vision (ICCV), pages 5804–5813. IEEE, 2017

    2017

  13. [21]

    Rus- sell

    Lisa Anne Hendricks, Oliver Wang, Eli Shechtman, Josef Sivic, Trevor Darrell, and Bryan C. Rus- sell. Localizing moments in video with natural language. In Proceedings of the IEEE International Conference on Computer Vision (ICCV), pages 5804–5813, 2017

    2017

  14. [22]

    Rextime: Temporal grounding benchmark for reasoning-intensive videos

    Bin Huang and et al. Rextime: Temporal grounding benchmark for reasoning-intensive videos. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 14313–14323, 2024

    2024

  15. [23]

    Vtimellm: Empower llm to grasp video moments

    Bin Huang, Xin Wang, Hong Chen, Zihan Song, and Wenwu Zhu. Vtimellm: Empower llm to grasp video moments. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 14271–14280, 2024

    2024

  16. [24]

    Lita: Language instructed temporal-localization assistant

    De-An Huang, Shijia Liao, Subhashree Radhakrishnan, Hongxu Yin, Pavlo Molchanov, Zhiding Yu, and Jan Kautz. Lita: Language instructed temporal-localization assistant. In Computer Vision – ECCV 2024, Part XXX (Proc. 18th European Conference on Computer Vision), pages 202–218. S...

    2024

  17. [25]

    Vision-r1: Incentivizing reasoning capability in multimodal large language models

    Wenxuan Huang, Bohan Jia, Zijie Zhai, Shaosheng Cao, Zheyu Ye, Fei Zhao, Zhe Xu, Yao Hu, and Shaohui Lin. Vision-r1: Incentivizing reasoning capability in multimodal large language models. arXiv preprint arXiv:2503.06749, 2025. 11 Datasets and Recipes for Video Temporal Ground...

    2025 arXiv

  18. [26]

    Vision-based abnormal event detection in industrial manufacturing processes: A review

    Hao Li, Shuhui Wang, Ying Zhou, and et al. Vision-based abnormal event detection in industrial manufacturing processes: A review. Computers in Industry , 130:103469, 2021. doi: 10.1016/j. compind.2021.103469

    2021

  19. [28]

    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. In arXiv preprint arXiv:2305.06355, 2023

    2023 arXiv

  20. [29]

    Videomind: A chain-of- lora agent for long video reasoning

    Ye Liu, Kevin Qinghong Lin, Chang Wen Chen, and Mike Zheng Shou. Videomind: A chain-of- lora agent for long video reasoning. arXiv preprint arXiv:2503.13444, 2025

    2025

  21. [30]

    Understanding r1-zero-like training: A critical perspective

    Zichen Liu, Changyu Chen, Wenjun Li, Penghui Qi, Tianyu Pang, Chao Du, Wee Sun Lee, and Min Lin. Understanding r1-zero-like training: A critical perspective. arXiv preprint arXiv:2503.20783, 2025

    2025 arXiv

  22. [31]

    Mm-eureka: Exploring visual aha moment with rule-based large-scale reinforcement learning

    Fanqing Meng, Lingxiao Du, Zongkai Liu, Zhixiang Zhou, Quanfeng Lu, Daocheng Fu, Botian Shi, Wenhai Wang, Junjun He, Kaipeng Zhang, et al. Mm-eureka: Exploring visual aha moment with rule-based large-scale reinforcement learning. CoRR, 2025

    2025

  23. [32]

    Queryd: A video dataset with high-quality text and audio narrations

    Andreea-Maria Oncescu, Joao F Henriques, Yang Liu, Andrew Zisserman, and Samuel Albanie. Queryd: A video dataset with high-quality text and audio narrations. In ICASSP 2021-2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 2265–2269....

    2021

  24. [33]

    Training language models to follow instructions with human feedback

    Long Ouyang, Jeffrey Wu, Xu Jiang, Diogo Almeida, Carroll Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, et al. Training language models to follow instructions with human feedback. Advances in neural information processing systems, 35: 277...

    2022

  25. [34]

    Momentor: Advancing video large language model with fine-grained temporal reasoning

    Long Qian, Juncheng Li, Yu Wu, Yaobo Ye, Hao Fei, Tat-Seng Chua, Yueting Zhuang, and Siliang Tang. Momentor: Advancing video large language model with fine-grained temporal reasoning. In Proceedings of the 41st International Conference on Machine Learning (ICML), 2024

    2024

  26. [35]

    Timechat: A time-sensitive multimodal large language model for long video understanding

    Shuhuai Ren, Linli Yao, Shicheng Li, Xu Sun, and Lu Hou. Timechat: A time-sensitive multimodal large language model for long video understanding. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 14313–14323, 2024

    2024

  27. [36]

    Proximal policy optimization algorithms

    John Schulman, Filip Wolski, Prafulla Dhariwal, Alec Radford, and Oleg Klimov. Proximal policy optimization algorithms. CoRR, abs/1707.06347, 2017. A simpler and effective first-order alternative to TRPO via surrogate objective and clipping

    2017 arXiv

  28. [37]

    Deepseekmath: Pushing the limits of mathematical reasoning in open language models

    Zhihong Shao et al. Deepseekmath: Pushing the limits of mathematical reasoning in open language models. arXiv preprint arXiv:2402.03300 , 2024. Introduces Group Relative Policy Optimization (GRPO) as a PPO variant enhancing reasoning ability and memory efficiency

    2024 arXiv

  29. [38]

    Learning to summarize with human feedback

    Nisan Stiennon, Long Ouyang, Jeffrey Wu, Daniel Ziegler, Ryan Lowe, Chelsea Voss, Alec Radford, Dario Amodei, and Paul F Christiano. Learning to summarize with human feedback. Advances in neural information processing systems, 33:3008–3021, 2020. 12 Datasets and Recipes for Vi...

    2020

  30. [39]

    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 Recognition (CVPR), pages 6479–6488, 2018

    2018

  31. [40]

    Endonet: A deep architecture for recognition tasks on laparoscopic videos

    Ap Twinanda, Sherif Shehata, Didier Mutter, Jacques Marescaux, Michel de Mathelin, and Nicolas Padoy. Endonet: A deep architecture for recognition tasks on laparoscopic videos. IEEE Transactions on Medical Imaging, 36(1):86–97, 2017. doi: 10.1109/TMI.2016.2593957

    2017

  32. [41]

    Grounded-videollm: Sharpening fine-grained temporal grounding in video large language models

    Haibo Wang, Zhiyang Xu, Yu Cheng, Shizhe Diao, Yufan Zhou, Yixin Cao, Qifan Wang, Weifeng Ge, and Lifu Huang. Grounded-videollm: Sharpening fine-grained temporal grounding in video large language models. arXiv preprint arXiv:2410.03290, 2024

    arXiv

  33. [43]

    Internvid: A large-scale video-text dataset for multimodal understand- ing and generation

    Yi Wang, Yinan He, Yizhuo Li, Kunchang Li, Jiashuo Yu, Xin Ma, Xinhao Li, Guo Chen, Xinyuan Chen, Yaohui Wang, et al. Internvid: A large-scale video-text dataset for multimodal understand- ing and generation. arXiv preprint arXiv:2307.06942, 2023

    2023 arXiv

  34. [44]

    Negative sample matters: A renaissance of metric learning for temporal grounding

    Zhenzhi Wang, Limin Wang, Tao Wu, Tianhao Li, and Gangshan Wu. Negative sample matters: A renaissance of metric learning for temporal grounding. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 36, pages 2613–2623, 2022

    2022

  35. [45]

    Task preference optimization: Improving multimodal large language models with vision task alignment

    Ziang Yan, Zhilin Li, Yinan He, Chenting Wang, Kunchang Li, Xinhao Li, Xiangyu Zeng, Zilei Wang, Yali Wang, Yu Qiao, et al. Task preference optimization: Improving multimodal large language models with vision task alignment. In Proceedings of the Computer Vision and Pattern Re...

    2025

  36. [46]

    Dapo: An open-source llm reinforcement learning system at scale

    Qiying Yu, Zheng Zhang, Ruofei Zhu, Yufeng Yuan, Xiaochen Zuo, Yu Yue, Weinan Dai, Tiantian Fan, Gaohong Liu, Lingjun Liu, et al. Dapo: An open-source llm reinforcement learning system at scale. arXiv preprint arXiv:2503.14476, 2025

    2025 arXiv

  37. [47]

    Hierarchical video-moment retrieval and step-captioning

    Abhay Zala, Jaemin Cho, Satwik Kottur, Xilun Chen, Barlas Oguz, Yashar Mehdad, and Mohit Bansal. Hierarchical video-moment retrieval and step-captioning. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 23056–23065, 2023

    2023

  38. [48]

    Easyr1: An efficient, scalable, multi-modality rl training framework

    Yaowei Zheng, Junting Lu, Shenzhi Wang, Zhangchi Feng, Dongdong Kuang, and Yuwen Xiong. Easyr1: An efficient, scalable, multi-modality rl training framework. https://github.com/ hiyouga/EasyR1, 2025. 13 Datasets and Recipes for Video Temporal Grounding via Reinforcement Learni...

    2025

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.