Pith. sign in

REVIEW 2 major objections 5 minor 2 cited by

Reasoning Segmentation for Images and Videos: A Survey

T0 review · 2 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read Reasoning segmentation asks models to segment objects named only by indirect text, and this survey provides the first comprehensive map of the field's methods, metrics, and datasets.

desk verdict Useful and well-organized survey of reasoning segmentation, but the 'first comprehensive' claim needs verification and the ReasonSeg numbers have a small internal contradiction. read the letter →

arxiv 2505.18816 v1 pith:AFPYFUQL submitted 2025-05-24 cs.CV

Yiqing Shen , Chenjia Li , Fei Xiong , Jeong-O Jeong , Tianpeng Wang , Michael Latman , Mathias Unberath This is my paper
classification cs.CV
keywords reasoningsegmentationmultimodallargelanguagemodelsimagevideoimplicittextqueriessurveybenchmarksevaluationmetrics
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

Reasoning segmentation (RS) asks a model to produce a pixel-level mask for an object named only indirectly in text, such as 'what could keep food fresh longer?' The paper claims to be the first comprehensive survey of this task, covering 26 methods, 29 datasets, and the associated evaluation metrics for both images and videos. It argues that RS is a distinct paradigm from semantic, instance, panoptic, and referring segmentation because the query demands multi-step reasoning and world knowledge, not just recognition or direct description. The survey organizes the field into image and video domains, traces the evolution from LISA's 'embedding-as-mask' design to conversational, multi-step, and agent-based systems, and identifies open problems in reasoning-aware metrics, interactive refinement, and domain applications. A sympathetic reader would come away with a map of where the field came from and what is missing.

What carries the argument

The organizing machinery is the task formulation itself plus the 'embedding-as-mask' mechanism introduced by LISA: given an image or video and an implicit text query, a multimodal LLM produces a special segmentation token (such as `<SEG>` or `<TRK>`) whose hidden embedding is turned into a mask by a decoder, often a frozen Segment Anything Model. The survey uses this mechanism as the reference point for its taxonomy, classifying methods by whether they fine-tune the LLM, add tokens for rejection or tracking, disentangle reasoning from segmentation, or replace token-based masks with prompt generation and agent-style planning. This single axis lets the survey compare otherwise heterogeneous methods and explain the field's evolution.

What would settle it

Audit the survey's tables against the primary sources: total the ReasonSeg split counts (the text states 1,218 samples while the table lists 239 + 200 + 799 = 1,238), check whether any RS method or benchmark published before May 2025 is absent from the timeline and tables, and look for an earlier or concurrent RS survey. A single omitted method, a misattributed capability, or an earlier survey would falsify the 'first comprehensive survey' claim.

Watch

Extended reading notes

Core claim

The paper's central claim is that reasoning segmentation has become a recognizable research area with its own lineage, whose founding move was LISA's 'embedding-as-mask' paradigm: a multimodal LLM emits a special `<SEG>` token, and the token's hidden embedding is decoded into a binary mask. The survey asserts that this single mechanism anchors most image methods, with variants adding instance discrimination, empty-target rejection, multi-round dialogue, chain-of-thought, and reinforcement learning, while video methods add temporal tokens, keyframe selection, and mask propagation. It further claims that datasets have shifted from manually annotated referring expressions such as RefCOCO to LLM-generated implicit queries at scale, and that evaluation remains dominated by mask-overlap metrics (cIoU, gIoU, ncIoU, mIoU) that under-measure reasoning quality. If this characterization is right, RS is not an isolated trick but a coherent task family whose progress can be tracked along dimensions of reasoning depth, interactivity, temporal consistency, and generalization.

Load-bearing premise

The survey's claims stand or fall on whether its enumeration of 26 methods, 29 datasets, and 12 metrics is complete through April 2025 and whether the one-line characterizations in its tables faithfully reflect the cited papers.

Editorial extensions

If this is right

  • Later image methods can be read as responses to specific limitations of the founding token-based approach: instance-level separation, empty-target rejection, multi-round dialogue, and forgetting during fine-tuning.
  • Video reasoning segmentation is image reasoning segmentation plus temporal machinery, with text-guided keyframe selection, tracking tokens, and mask propagation as the recurring design choices.
  • Mask-only IoU metrics will not expose gains in reasoning quality, so the emerging text-aware metrics and LLM-judged scores will need standardization before the field can compare methods fairly.
  • Datasets built by large language models may train models that match machine-generated reasoning patterns well while causal, counterfactual, and analogical reasoning remain under-tested.
  • The survey's identified gaps point to concrete next steps: multi-step and interactive reasoning, uncertainty-aware outputs, reasoning-specific evaluation, and integration with audio, depth, or LiDAR modalities.

Reading between the lines

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

  • If this taxonomy is accurate, a new RS method can be classified by which founding limitation it attacks; adding a new token or training scheme without changing the core embedding-to-mask pathway is an incremental variant, not a new paradigm.
  • The field's dependence on LLM-generated queries creates a possible feedback loop: models trained and evaluated on similar LLM-authored language may overstate their ability to handle human-written implicit requests.
  • Because a segmentation mask forces the model to commit to one spatial interpretation, RS could be a sharper test of whether multimodal models genuinely reason than text-only benchmarks; comparing mask quality across reasoning categories might expose where reasoning is shallow.
  • The LLM-judged text metrics inherit their judge's biases; a natural next experiment would be to measure agreement between GPT-score, human ratings, and mask-based metrics across reasoning types.
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

2 major / 5 minor

Summary. This paper presents a survey of reasoning segmentation (RS), a task in which models segment objects based on implicit text queries that require reasoning and world knowledge. The authors organize the field into image and video domains, review 26 methods (17 image, 9 video), provide an overview of 12 evaluation metrics, and catalog 29 datasets and benchmarks. They also discuss existing applications (safety monitoring, underwater imaging, operating-room workflow, earth observation) and propose future directions. The paper positions itself as the first comprehensive survey of RS.

Significance. If the survey's enumeration and characterizations are accurate, it would serve as a valuable organizational reference for a rapidly growing field and would help researchers navigate the method space, metric zoo, and benchmark landscape. The paper takes a useful taxonomy-driven approach and includes a fine-grained comparison across architectural features, processing strategies, and implementation details. Its central contribution is the comprehensiveness and correctness of the catalog itself; therefore, accuracy of the numeric data and the reproducibility of the literature search are load-bearing for the paper's value. The internal inconsistency in dataset statistics and the absence of a search protocol are significant concerns in the current version.

major comments (2)
  1. [§2.4, Table 3, §5.1] The ReasonSeg dataset statistics are internally inconsistent. Section 2.4 states that ReasonSeg consists of 1,218 image-instruction-mask samples with a test split of 779, while Table 3 reports a total of 1,238 and a test split of 799, and Section 5.1 also cites the total as 1,238. This is a direct factual contradiction in a central descriptive claim of the survey. The authors should verify the original ReasonSeg paper, correct the numbers, and ensure that the same figures appear consistently in the text, the table, and the discussion.
  2. [Abstract, §1] The claim that this is 'the first comprehensive survey' of reasoning segmentation is load-bearing for the paper's contribution, but it is not supported by any reproducible search methodology. The paper provides no query strings, databases searched, inclusion/exclusion criteria, or a date-stamped coverage statement. Given the field's rapid growth, with preprints appearing within weeks of the submission date (e.g., POPEN 2504.00640 and JiT 2503.21056), a missed prior or concurrent survey would invalidate the 'first' claim, and a missed method or dataset would weaken the 'comprehensive' claim. The authors should either add a reproducible literature-search section or temper the first/comprehensive wording to match what is actually demonstrated.
minor comments (5)
  1. [§2.4] The example query 'After cooking, consuming food, and preparing for food, where can we throw away the rest of the food and scraps?' reads awkwardly; please revise the wording to improve clarity.
  2. [Table 3] In the GranDf row, the train column shows a dash while a total of 214,000 is given; please clarify the split counts (train/val/test) to make the total transparent.
  3. [§4.1] There is a typo in 'implict text query' in the Underwater Images Segmentation paragraph; it should be 'implicit text query'.
  4. [Table 1] The column header 'Fine-TuningFree of MLLM' is ambiguous; it appears to combine two separate concepts, 'fine-tuning' and 'free of MLLM.' Please split the header into two clear columns to avoid confusion when reading the check-mark entries.
  5. [§2.2] For LLaVASeg, the phrase 'representing the first attempt at free of MLLM-tuning' is grammatically awkward; suggest rewording to 'the first attempt to avoid MLLM fine-tuning.'

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation; the survey is organizational and its claims do not reduce to fitted inputs or self-citations.

full rationale

This paper is a literature survey; it derives no quantitative predictions and fits no parameters, so there is no derivation-to-input circularity of the kinds enumerated. The central claims are enumerative ('26 state-of-the-art methods', '29 datasets and benchmarks', 'first comprehensive survey'), and the taxonomy in Tables 1-4 is a classification of cited works rather than a result computed from those citations. Self-citations appear (JiT [74], ORDiRS [73], RVTBench [72], digital twin position [70]), but they are presented as surveyed methods, application examples, or supporting citations for a limitation about LLM-generated query diversity; none is used as load-bearing evidence for the survey's own completeness or as a forced uniqueness result. The 'first comprehensive survey' assertion lacks a documented search protocol, and the ReasonSeg sample-count inconsistency (Table 3 vs Section 2.4) raises correctness and completeness concerns, but those are accuracy risks, not circularity. Per the scoring rules, absence of circularity receives score 0.

Assumptions & free parameters 0 free parameters · 1 assumptions · 0 invented entities

As a survey, the paper introduces no fitted parameters, no new mathematical postulates, and no new entities. Its claims rest entirely on the accuracy and completeness of its literature characterization and on internal consistency, which is only partially met.

assumptions (1)
  • domain assumption The enumerated set of 26 methods, 29 datasets, and 12 metrics is comprehensive and accurately characterized as of the April 2025 cutoff.
    The survey's central contribution is its taxonomy; if coverage is incomplete or characterizations are inaccurate, the survey's utility is reduced. This is introduced in the Introduction and instantiated in Tables 1-4.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Reasoning Segmentation for Images and Videos: A Survey." pith.science (2026). https://pith.science/paper/AFPYFUQL

@misc{pith2026250518816,
  author       = {Pith},
  title        = {Pith review of: Reasoning Segmentation for Images and Videos: A Survey},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/AFPYFUQL}},
  note         = {Machine review of arXiv:2505.18816}
}
read the original abstract

Reasoning Segmentation (RS) aims to delineate objects based on implicit text queries, the interpretation of which requires reasoning and knowledge integration. Unlike the traditional formulation of segmentation problems that relies on fixed semantic categories or explicit prompting, RS bridges the gap between visual perception and human-like reasoning capabilities, facilitating more intuitive human-AI interaction through natural language. Our work presents the first comprehensive survey of RS for image and video processing, examining 26 state-of-the-art methods together with a review of the corresponding evaluation metrics, as well as 29 datasets and benchmarks. We also explore existing applications of RS across diverse domains and identify their potential extensions. Finally, we identify current research gaps and highlight promising future directions.

Figures

Figures reproduced from arXiv: 2505.18816 by the authors.

Figure 1
Figure 1. Illustration of various segmentation tasks on a street-crossing scene. [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Timeline of RS methods from October 2023 to April 2025, illustrating [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Overview of the LISA [40], which adopts an “embedding-as-mask” [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Comparison of training approaches for RS models: (a) Supervised [PITH_FULL_IMAGE:figures/full_fig_p012_4.png]
Figure 5
Figure 5. Figure 5: Illustrative example of image RS data. of RS datasets reflects a clear progression from simple referring expressions to increasingly complex reasoning tasks. Early datasets like RefCOCO and Re￾fCOCO+ [34] established foundational benchmarks with manually annotated natu…
Figure 6
Figure 6. Figure 6: Overview of the VISA [93]. 23 [PITH_FULL_IMAGE:figures/full_fig_p023_6.png]
Figure 7
Figure 7. Figure 7: Illustrative example of video RS data. seconds with pixel-level instance segmentation annotations at every 5 frames. Refer-Youtube-VOS has two distinct annotation types, namely “full video ex￾pressions” where annotators watch the entire video, and “first-frame expressi…

Discussion (0). Sign in to comment.

Forward citations

Cited by 2 Pith papers

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

  1. DGSeg: Dynamic Gating of Semantic-Spatial Guided Predictions for Reasoning Segmentation

    cs.CV 2026-07 accept novelty 6.0 of 10

    Separate semantic-spatial MLLM cues plus a quality-supervised dynamic gate yield 69.6/67.3 gIoU on ReasonSeg val/test, beating strong MLLM+SAM baselines.

  2. Temporally-Constrained Video Reasoning Segmentation and Automated Benchmark Construction

    cs.CV 2025-07 conditional novelty 6.0 of 10

    Temporally-constrained video reasoning segmentation is introduced, with an automated benchmark construction pipeline and a 52-sample dataset from the MVOR surgical videos.

Reference graph

Works this paper leans on

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

  1. [1]

    UI-Net: Interactive Artificial Neural Networks for Iterative Image Segmentation Based on a User Model

    Amrehn, M., Gaube, S., Unberath, M., Schebesch, F., Horz, T., Strumia, M., Steidl, S., Kowarschik, M., Maier, A., 2017. Ui-net: Interactive ar- tificial neural networks for iterative image segmentation based on a user model. arXiv preprint arXiv:1709.03450

    work page Pith review arXiv 2017

  2. [2]

    Vicas: A dataset for combining holistic and pixel-level video understanding using captions with grounded segmentation

    Athar, A., Deng, X., Chen, L.C., 2024. Vicas: A dataset for combining holistic and pixel-level video understanding using captions with grounded segmentation. arXiv preprint arXiv:2412.09754

    arXiv 2024

  3. [3]

    Burst: A benchmark for unifying object recognition, 39 segmentation and tracking in video, in: Proceedings of the IEEE/CVF winter conference on applications of computer vision, pp

    Athar, A., Luiten, J., Voigtlaender, P., Khurana, T., Dave, A., Leibe, B., Ramanan, D., 2023. Burst: A benchmark for unifying object recognition, 39 segmentation and tracking in video, in: Proceedings of the IEEE/CVF winter conference on applications of computer vision, pp. 1674–1683

    2023

  4. [4]

    One token to seg them all: Language instructed reasoning segmentation in videos

    Bai, Z., He, T., Mei, H., Wang, P., Gao, Z., Chen, J., Liu, L., Zhang, Z., Shou, M.Z., 2024. One token to seg them all: Language instructed reasoning segmentation in videos. arXiv preprint arXiv:2409.19603

    arXiv 2024

  5. [5]

    Cores: Orchestrating the dance of reasoning and segmentation, in: European Conference on Computer Vision, Springer

    Bao, X., Sun, S., Ma, S., Zheng, K., Guo, Y., Zhao, G., Zheng, Y., Wang, X., 2024. Cores: Orchestrating the dance of reasoning and segmentation, in: European Conference on Computer Vision, Springer. pp. 187–204

    2024

  6. [6]

    Xmem++: Production-level video segmentation from few annotated frames, in: Pro- ceedings of the IEEE/CVF International Conference on Computer Vision, pp

    Bekuzarov, M., Bermudez, A., Lee, J.Y., Li, H., 2023. Xmem++: Production-level video segmentation from few annotated frames, in: Pro- ceedings of the IEEE/CVF International Conference on Computer Vision, pp. 635–644

    2023

  7. [7]

    Coco-stuff: Thing and stuff classes in context, in: Proceedings of the IEEE conference on computer vision and pattern recognition, pp

    Caesar, H., Uijlings, J., Ferrari, V., 2018a. Coco-stuff: Thing and stuff classes in context, in: Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 1209–1218

  8. [8]

    Coco-stuff: Thing and stuff classes in context, in: Proceedings of the IEEE conference on computer vision and pattern recognition, pp

    Caesar, H., Uijlings, J., Ferrari, V., 2018b. Coco-stuff: Thing and stuff classes in context, in: Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 1209–1218

Show all 114 references
  1. [9]

    Pixel-level reasoning segmentation via multi-turn conversations

    Cai, D., Yang, X., Liu, Y., Wang, D., Feng, S., Zhang, Y., Poria, S., 2025. Pixel-level reasoning segmentation via multi-turn conversations. arXiv preprint arXiv:2502.09447

    2025 arXiv

  2. [10]

    End-to-end object detection with transformers, in: European conference on computer vision, Springer

    Carion, N., Massa, F., Synnaeve, G., Usunier, N., Kirillov, A., Zagoruyko, S., 2020. End-to-end object detection with transformers, in: European conference on computer vision, Springer. pp. 213–229

    2020

  3. [11]

    Chen, F.L., Zhang, D.Z., Han, M.L., Chen, X.Y., Shi, J., Xu, S., Xu, B.,

  4. [12]

    Chen, X., Mottaghi, R., Liu, X., Fidler, S., Urtasun, R., Yuille, A., 2014. Detect what you can: Detecting and representing objects using holistic models and body parts, in: Proceedings of the IEEE conference on com- puter vision and pattern recognition, pp. 1971–1978

    2014

  5. [13]

    Sam4mllm: Enhance multi-modal large language model for referring ex- pression segmentation, in: European Conference on Computer Vision, Springer

    Chen, Y.C., Li, W.H., Sun, C., Wang, Y.C.F., Chen, C.S., 2024. Sam4mllm: Enhance multi-modal large language model for referring ex- pression segmentation, in: European Conference on Computer Vision, Springer. pp. 323–340

    2024

  6. [14]

    Masked-attention mask transformer for universal image segmentation, in: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp

    Cheng, B., Misra, I., Schwing, A.G., Kirillov, A., Girdhar, R., 2022. Masked-attention mask transformer for universal image segmentation, in: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp. 1290–1299. 40

    2022

  7. [15]

    Xmem: Long-term video object seg- mentation with an atkinson-shiffrin memory model, in: European Confer- ence on Computer Vision, Springer

    Cheng, H.K., Schwing, A.G., 2022. Xmem: Long-term video object seg- mentation with an atkinson-shiffrin memory model, in: European Confer- ence on Computer Vision, Springer. pp. 640–658

    2022

  8. [16]

    Vocabulary-free image classification

    Conti, A., Fini, E., Mancini, M., Rota, P., Wang, Y., Ricci, E., 2023. Vocabulary-free image classification. Advances in Neural Information Pro- cessing Systems 36, 30662–30680

    2023

  9. [17]

    Scannet: Richly-annotated 3d reconstructions of indoor scenes, in: Proceedings of the IEEE conference on computer vision and pattern recognition, pp

    Dai, A., Chang, A.X., Savva, M., Halber, M., Funkhouser, T., Nießner, M., 2017. Scannet: Richly-annotated 3d reconstructions of indoor scenes, in: Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 5828–5839

    2017

  10. [18]

    Tao: A large-scale benchmark for tracking any object, in: Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part V 16, Springer

    Dave, A., Khurana, T., Tokmakov, P., Schmid, C., Ramanan, D., 2020. Tao: A large-scale benchmark for tracking any object, in: Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part V 16, Springer. pp. 436–454

    2020

  11. [19]

    Motion-grounded video reasoning: Understanding and perceiving motion at pixel level

    Deng, A., Chen, T., Yu, S., Yang, T., Spencer, L., Tian, Y., Mian, A.S., Bansal, M., Chen, C., 2024. Motion-grounded video reasoning: Understanding and perceiving motion at pixel level. arXiv preprint arXiv:2411.09921

    2024 arXiv

  12. [20]

    Mevis: A large-scale benchmark for video segmentation with motion expressions, in: Proceed- ings of the IEEE/CVF International Conference on Computer Vision, pp

    Ding, H., Liu, C., He, S., Jiang, X., Loy, C.C., 2023a. Mevis: A large-scale benchmark for video segmentation with motion expressions, in: Proceed- ings of the IEEE/CVF International Conference on Computer Vision, pp. 2694–2703

  13. [21]

    Mose: A new dataset for video object segmentation in complex scenes, in: Pro- ceedings of the IEEE/CVF international conference on computer vision, pp

    Ding, H., Liu, C., He, S., Jiang, X., Torr, P.H., Bai, S., 2023b. Mose: A new dataset for video object segmentation in complex scenes, in: Pro- ceedings of the IEEE/CVF international conference on computer vision, pp. 20224–20234

  14. [22]

    Panoptic segmentation: A review

    Elharrouss, O., Al-Maadeed, S., Subramanian, N., Ottakath, N., Al- maadeed, N., Himeur, Y., 2021. Panoptic segmentation: A review. arXiv preprint arXiv:2111.10250

    2021 arXiv

  15. [23]

    Oops! predicting unintentional action in video, in: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp

    Epstein, D., Chen, B., Vondrick, C., 2020. Oops! predicting unintentional action in video, in: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp. 919–929

    2020

  16. [24]

    A survey for foundation models in autonomous driving

    Gao, H., Wang, Z., Li, Y., Long, K., Yang, M., Shen, Y., 2024. A survey for foundation models in autonomous driving. arXiv preprint arXiv:2402.01105

    2024 arXiv

  17. [25]

    Part- aware panoptic segmentation, in: Proceedings of the IEEE/CVF Confer- ence on Computer Vision and Pattern Recognition, pp

    de Geus, D., Meletis, P., Lu, C., Wen, X., Dubbelman, G., 2021. Part- aware panoptic segmentation, in: Proceedings of the IEEE/CVF Confer- ence on Computer Vision and Pattern Recognition, pp. 5485–5494. 41

    2021

  18. [26]

    The devil is in temporal token: High quality video reasoning segmentation

    Gong, S., Zhuge, Y., Zhang, L., Yang, Z., Zhang, P., Lu, H., 2025. The devil is in temporal token: High quality video reasoning segmentation. arXiv preprint arXiv:2501.08549

    2025 arXiv

  19. [27]

    The iapr tc-12 benchmark: A new evaluation resource for visual information systems, in: International workshop ontoImage, pp

    Grubinger, M., Clough, P., Müller, H., Deselaers, T., 2006. The iapr tc-12 benchmark: A new evaluation resource for visual information systems, in: International workshop ontoImage, pp. 1–10

    2006

  20. [28]

    Lvis: A dataset for large vocabu- lary instance segmentation, in: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp

    Gupta, A., Dollar, P., Girshick, R., 2019. Lvis: A dataset for large vocabu- lary instance segmentation, in: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp. 5356–5364

    2019

  21. [29]

    A survey on instance segmentation: state of the art

    Hafiz, A.M., Bhat, G.M., 2020. A survey on instance segmentation: state of the art. International journal of multimedia information retrieval 9, 171–189

    2020

  22. [30]

    A brief survey on semantic segmentation with deep learning

    Hao, S., Zhou, Y., Guo, Y., 2020. A brief survey on semantic segmentation with deep learning. Neurocomputing 406, 302–321

    2020

  23. [31]

    Lora: Low-rank adaptation of large language models

    Hu, E.J., Shen, Y., Wallis, P., Allen-Zhu, Z., Li, Y., Wang, S., Wang, L., Chen, W., 2021. Lora: Low-rank adaptation of large language models. arXiv preprint arXiv:2106.09685

    2021 arXiv

  24. [32]

    Huang, Q., Dong, X., Zhang, P., Wang, B., He, C., Wang, J., Lin, D., Zhang, W., Yu, N., 2024. Opera: Alleviating hallucination in multi-modal large language models via over-trust penalty and retrospection-allocation, in: Proceedings of the IEEE/CVF Conference on Computer Visio...

    2024

  25. [33]

    Mmr: A large- scale benchmark dataset for multi-target and multi-granularity reasoning segmentation

    Jang, D., Cho, Y., Lee, S., Kim, T., Kim, D.S., 2025. Mmr: A large- scale benchmark dataset for multi-target and multi-granularity reasoning segmentation. arXiv preprint arXiv:2503.13881

    2025 arXiv

  26. [34]

    Referitgame: Referring to objects in photographs of natural scenes, in: Proceedings of the 2014 conference on empirical methods in natural language processing (EMNLP), pp

    Kazemzadeh, S., Ordonez, V., Matten, M., Berg, T., 2014. Referitgame: Referring to objects in photographs of natural scenes, in: Proceedings of the 2014 conference on empirical methods in natural language processing (EMNLP), pp. 787–798

    2014

  27. [35]

    Khan, S., Naseer, M., Hayat, M., Zamir, S.W., Khan, F.S., Shah, M.,

  28. [36]

    Khoreva, A., Rohrbach, A., Schiele, B., 2019. Video object segmentation with language referring expressions, in: Computer Vision–ACCV 2018: 14th Asian Conference on Computer Vision, Perth, Australia, December 2–6, 2018, Revised Selected Papers, Part IV 14, Springer. pp. 123–141

    2019

  29. [37]

    Segment anything

    Kirillov, A., et al., 2023. Segment anything. arXiv preprint arXiv:2304.02643 . 42

    2023 arXiv

  30. [38]

    Visual genome: Connecting language and vision using crowdsourced dense image annota- tions

    Krishna, R., Zhu, Y., Groth, O., Johnson, J., Hata, K., Kravitz, J., Chen, S., Kalantidis, Y., Li, L.J., Shamma, D.A., et al., 2017. Visual genome: Connecting language and vision using crowdsourced dense image annota- tions. International journal of computer vision 123, 32–73

    2017

  31. [39]

    The open images dataset v4: Unified image classification, object detection, and visual relationship detection at scale

    Kuznetsova, A., Rom, H., Alldrin, N., Uijlings, J., Krasin, I., Pont-Tuset, J., Kamali, S., Popov, S., Malloci, M., Kolesnikov, A., et al., 2020. The open images dataset v4: Unified image classification, object detection, and visual relationship detection at scale. Internation...

    2020

  32. [40]

    Lisa: Reasoning segmentation via large language model, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Lai, X., Tian, Z., Chen, Y., Li, Y., Yuan, Y., Liu, S., Jia, J., 2024. Lisa: Reasoning segmentation via large language model, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 9579–9589

    2024

  33. [41]

    Blip-2: Bootstrapping language- image pre-training with frozen image encoders and large language mod- els, in: International conference on machine learning, PMLR

    Li, J., Li, D., Savarese, S., Hoi, S., 2023a. Blip-2: Bootstrapping language- image pre-training with frozen image encoders and large language mod- els, in: International conference on machine learning, PMLR. pp. 19730– 19742

  34. [42]

    Segearth-r1: Geospatial pixel reasoning via large language model

    Li, K., Xin, Z., Pang, L., Pang, C., Deng, Y., Yao, J., Xia, G., Meng, D., Wang, Z., Cao, X., 2025. Segearth-r1: Geospatial pixel reasoning via large language model. arXiv preprint arXiv:2504.09644

    2025 arXiv

  35. [43]

    Robust referring video object segmentation with cyclic structural consensus, in: Proceed- ings of the IEEE/CVF International Conference on Computer Vision, pp

    Li, X., Wang, J., Xu, X., Li, X., Raj, B., Lu, Y., 2023b. Robust referring video object segmentation with cyclic structural consensus, in: Proceed- ings of the IEEE/CVF International Conference on Computer Vision, pp. 22236–22245

  36. [44]

    Llama-vid: An image is worth 2 tokens in large language models, in: European Conference on Computer Vision, Springer

    Li, Y., Wang, C., Jia, J., 2024. Llama-vid: An image is worth 2 tokens in large language models, in: European Conference on Computer Vision, Springer. pp. 323–340

    2024

  37. [45]

    Microsoft coco: Common objects in context, in: Computer Vision–ECCV 2014: 13th European Confer- ence, Zurich, Switzerland, September 6-12, 2014, Proceedings, Part V 13, Springer

    Lin, T.Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Dollár, P., Zitnick, C.L., 2014. Microsoft coco: Common objects in context, in: Computer Vision–ECCV 2014: 13th European Confer- ence, Zurich, Switzerland, September 6-12, 2014, Proceedings, Part V 13, Spri...

    2014

  38. [46]

    Gres: Generalized referring expression segmentation, in: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp

    Liu, C., Ding, H., Jiang, X., 2023. Gres: Generalized referring expression segmentation, in: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp. 23592–23601

    2023

  39. [47]

    Visual instruction tuning

    Liu, H., Li, C., Wu, Q., et al., 2024. Visual instruction tuning. Advances in neural information processing systems 36

    2024

  40. [48]

    Seg- zero: Reasoning-chain guided segmentation via cognitive reinforcement

    Liu, Y., Peng, B., Zhong, Z., Yue, Z., Lu, F., Yu, B., Jia, J., 2025. Seg- zero: Reasoning-chain guided segmentation via cognitive reinforcement. arXiv preprint arXiv:2503.06520 . 43

    2025 arXiv

  41. [49]

    Ground abstract structure concepts of scaffolding systems for automatic compliance check- ing based on reasoning segmentation

    Luo, X., Qin, M., Gao, Z., Yan, H., Yang, X., 2025. Ground abstract structure concepts of scaffolding systems for automatic compliance check- ing based on reasoning segmentation. Expert Systems with Applications 270, 126563

    2025

  42. [50]

    Video anomaly detection and explanation via large language models

    Lv, H., Sun, Q., 2024. Video anomaly detection and explanation via large language models

    2024

  43. [51]

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

    Maaz, M., Rasheed, H., Khan, S., Khan, F.S., 2023. Video-chatgpt: To- wards detailed video understanding via large vision and language models. arXiv preprint arXiv:2306.05424

    2023 arXiv

  44. [52]

    Mao, J., Huang, J., Toshev, A., Camburu, O., Yuille, A.L., Murphy, K.,

  45. [53]

    Large-scale video panoptic segmentation in the wild: A benchmark, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pat- tern Recognition, pp

    Miao, J., Wang, X., Wu, Y., Li, W., Zhang, X., Wei, Y., Yang, Y., 2022. Large-scale video panoptic segmentation in the wild: A benchmark, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pat- tern Recognition, pp. 21033–21043

    2022

  46. [54]

    Image segmentation using deep learning: A survey

    Minaee, S., Boykov, Y., Porikli, F., Plaza, A., Kehtarnavaz, N., Terzopou- los, D., 2021. Image segmentation using deep learning: A survey. IEEE transactions on pattern analysis and machine intelligence 44, 3523–3542

    2021

  47. [55]

    GPT-4 Technical Report

    OpenAI, 2023a. GPT-4 Technical Report. Technical Report. OpenAI. Technical report detailing the architecture and capabilities of GPT-4

  48. [56]

    GPT-4V(ision) System Card

    OpenAI, 2023b. GPT-4V(ision) System Card. Technical Report. OpenAI. System card outlining vision capabilities, limitations, and safety evalua- tions for GPT-4V

  49. [57]

    Di- nov2: Learning robust visual features without supervision

    Oquab, M., Darcet, T., Moutakanni, T., Vo, H., Szafraniec, M., Khalidov, V., Fernandez, P., Haziza, D., Massa, F., El-Nouby, A., et al., 2023. Di- nov2: Learning robust visual features without supervision. arXiv preprint arXiv:2304.07193

    2023 arXiv

  50. [58]

    Flickr30k entities: Collecting region-to-phrase correspondences for richer image-to-sentence models, in: Proceedings of the IEEE international conference on computer vision, pp

    Plummer, B.A., Wang, L., Cervantes, C.M., Caicedo, J.C., Hockenmaier, J., Lazebnik, S., 2015. Flickr30k entities: Collecting region-to-phrase correspondences for richer image-to-sentence models, in: Proceedings of the IEEE international conference on computer vision, pp. 2641–2649

    2015

  51. [59]

    The 2017 davis challenge on video object segmenta- tion

    Pont-Tuset, J., Perazzi, F., Caelles, S., Arbeláez, P., Sorkine-Hornung, A., Van Gool, L., 2017. The 2017 davis challenge on video object segmenta- tion. arXiv preprint arXiv:1704.00675

    2017 arXiv

  52. [60]

    Occluded video instance segmentation: A benchmark

    Qi, J., Gao, Y., Hu, Y., Wang, X., Liu, X., Bai, X., Belongie, S., Yuille, A., Torr, P.H., Bai, S., 2022. Occluded video instance segmentation: A benchmark. International Journal of Computer Vision 130, 2022–2039. 44

    2022

  53. [61]

    Reasoning to attend: Try to understand how< seg> token works

    Qian, R., Yin, X., Dou, D., 2024. Reasoning to attend: Try to understand how< seg> token works. arXiv preprint arXiv:2412.17741

    2024

  54. [62]

    Learning trans- ferable visual models from natural language supervision, in: International conference on machine learning, PMLR

    Radford, A., Kim, J.W., Hallacy, C., Ramesh, A., Goh, G., Agarwal, S., Sastry, G., Askell, A., Mishkin, P., Clark, J., et al., 2021. Learning trans- ferable visual models from natural language supervision, in: International conference on machine learning, PMLR. pp. 8748–8763

    2021

  55. [63]

    Paco: Parts and attributes of common objects, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Ramanathan, V., Kalia, A., Petrovic, V., Wen, Y., Zheng, B., Guo, B., Wang, R., Marquez, A., Kovvuri, R., Kadian, A., et al., 2023. Paco: Parts and attributes of common objects, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 7141–7151

    2023

  56. [64]

    Glamm: Pixelground- inglargemultimodalmodel, in: ProceedingsoftheIEEE/CVFConference on Computer Vision and Pattern Recognition, pp

    Rasheed, H., Maaz, M., Shaji, S., Shaker, A., Khan, S., Cholakkal, H., An- wer, R.M., Xing, E., Yang, M.H., Khan, F.S., 2024. Glamm: Pixelground- inglargemultimodalmodel, in: ProceedingsoftheIEEE/CVFConference on Computer Vision and Pattern Recognition, pp. 13009–13018

    2024

  57. [65]

    Sam 2: Segment anything in images and videos

    Ravi, N., Gabeur, V., Hu, Y.T., et al., 2024. Sam 2: Segment anything in images and videos. arXiv preprint arXiv:2408.00714

    2024 arXiv

  58. [66]

    Sentence-bert: Sentence embeddings using siamese bert-networks

    Reimers, N., 2019. Sentence-bert: Sentence embeddings using siamese bert-networks. arXiv preprint arXiv:1908.10084

    2019 arXiv

  59. [67]

    Pixellm: Pixel reasoning with large multimodal model, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Ren, Z., Huang, Z., Wei, Y., Zhao, Y., Fu, D., Feng, J., Jin, X., 2024. Pixellm: Pixel reasoning with large multimodal model, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 26374–26383

    2024

  60. [68]

    Objecthallucinationinimagecaptioning

    Rohrbach, A., Hendricks, L.A., Burns, K., Darrell, T., Saenko, K., 2018. Objecthallucinationinimagecaptioning. arXivpreprintarXiv:1809.02156

    2018 arXiv

  61. [69]

    Seo, S., Lee, J.Y., Han, B., 2020. Urvos: Unified referring video ob- ject segmentation network with a large-scale benchmark, in: Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part XV 16, Springer. pp. 208–223

    2020

  62. [70]

    Posi- tion: Foundation models need digital twin representations

    Shen, Y., Ding, H., Seenivasan, L., Shu, T., Unberath, M., 2025a. Posi- tion: Foundation models need digital twin representations. arXiv preprint arXiv:2505.03798

    arXiv

  63. [71]

    Shen, Y., Ding, H., Shao, X., Unberath, M., 2025b. Performance and nonadversarial robustness of the segment anything model 2 in surgical video segmentation, in: Medical Imaging 2025: Image-Guided Procedures, Robotic Interventions, and Modeling, SPIE. pp. 93–98

    2025

  64. [72]

    Rvtbench: A benchmark for visual reasoning tasks

    Shen, Y., Li, C., Fan, C., Unberath, M., 2025c. Rvtbench: A benchmark for visual reasoning tasks. arXiv preprint arXiv:2505.11838 . 45

    arXiv

  65. [73]

    Operating room workflow analysis via reasoning segmentation over digital twins

    Shen, Y., Li, C., Liu, B., Li, C.Y., Porras, T., Unberath, M., 2025d. Operating room workflow analysis via reasoning segmentation over digital twins. arXiv preprint arXiv:2503.21054

    arXiv

  66. [74]

    Online rea- soning video segmentation with just-in-time digital twins

    Shen, Y., Liu, B., Li, C., Seenivasan, L., Unberath, M., 2025e. Online rea- soning video segmentation with just-in-time digital twins. arXiv preprint arXiv:2503.21056

    arXiv

  67. [75]

    Sidorov, O., Hu, R., Rohrbach, M., Singh, A., 2020. Textcaps: a dataset for image captioning with reading comprehension, in: Computer Vision– ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part II 16, Springer. pp. 742–758

    2020

  68. [76]

    Eva-clip: Improved training techniques for clip at scale

    Sun, Q., Fang, Y., Wu, L., Wang, X., Cao, Y., 2023. Eva-clip: Improved training techniques for clip at scale. arXiv preprint arXiv:2303.15389

    2023 arXiv

  69. [77]

    Growcut: Interactive multi-label nd image segmentation by cellular automata, in: proc

    Vezhnevets, V., Konouchine, V., 2005. Growcut: Interactive multi-label nd image segmentation by cellular automata, in: proc. of Graphicon, Cite- seer. pp. 150–156

    2005

  70. [78]

    Reducing the annotation effort for video object segmentation datasets, in: Proceed- ings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pp

    Voigtlaender, P., Luo, L., Yuan, C., Jiang, Y., Leibe, B., 2021. Reducing the annotation effort for video object segmentation datasets, in: Proceed- ings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pp. 3060–3069

    2021

  71. [79]

    Prima: Multi-image vision-language models for reasoning segmentation

    Wahed, M., Nguyen, K.A., Juvekar, A.S., Li, X., Zhou, X., Shah, V., Yu, T., Yanardag, P., Lourentzou, I., 2024. Prima: Multi-image vision-language models for reasoning segmentation. arXiv preprint arXiv:2412.15209

    2024

  72. [80]

    Ov-vis: Open-vocabulary video instance segmentation

    Wang, H., Yan, C., Chen, K., Jiang, X., Tang, X., Hu, Y., Kang, G., Xie, W., Gavves, E., 2024a. Ov-vis: Open-vocabulary video instance segmentation. International Journal of Computer Vision 132, 5048–5065

  73. [81]

    Towards open-vocabulary video instance segmentation, in: proceedings of the IEEE/CVF international conference on computer vision, pp

    Wang, H., Yan, C., Wang, S., Jiang, X., Tang, X., Hu, Y., Xie, W., Gavves, E., 2023. Towards open-vocabulary video instance segmentation, in: proceedings of the IEEE/CVF international conference on computer vision, pp. 4057–4066

    2023

  74. [82]

    Llm-seg: Bridging image segmentation and large language model reasoning, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Wang, J., Ke, L., 2024. Llm-seg: Bridging image segmentation and large language model reasoning, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 1765–1774

    2024

  75. [83]

    Unidentified video objects: A benchmark for dense, open-world segmentation, in: Proceed- ings of the IEEE/CVF international conference on computer vision, pp

    Wang, W., Feiszli, M., Wang, H., Tran, D., 2021. Unidentified video objects: A benchmark for dense, open-world segmentation, in: Proceed- ings of the IEEE/CVF international conference on computer vision, pp. 10776–10785. 46

    2021

  76. [84]

    Segllm: Multi-roundreasoningsegmentation

    Wang, X., Zhang, S., Li, S., Kallidromitis, K., Li, K., Kato, Y., Kozuka, K., Darrell, T., 2024b. Segllm: Multi-roundreasoningsegmentation. arXiv preprint arXiv:2410.18923

    arXiv

  77. [85]

    Lasagna: Language-based segmentation assistant for complex queries

    Wei, C., Tan, H., Zhong, Y., Yang, Y., Ma, L., 2024a. Lasagna: Language-based segmentation assistant for complex queries. arXiv preprint arXiv:2404.08506

    arXiv

  78. [86]

    Instructseg: Unifying instructed visual segmentation with multi-modal large language models

    Wei, C., Zhong, Y., Tan, H., Zeng, Y., Liu, Y., Zhao, Z., Yang, Y., 2024b. Instructseg: Unifying instructed visual segmentation with multi-modal large language models. arXiv preprint arXiv:2412.14006

    arXiv

  79. [87]

    Chain-of-thought prompting elicits reasoning in large language models

    Wei, J., Wang, X., Schuurmans, D., Bosma, M., Xia, F., Chi, E., Le, Q.V., Zhou, D., et al., 2022. Chain-of-thought prompting elicits reasoning in large language models. Advances in neural information processing systems 35, 24824–24837

    2022

  80. [88]

    Ov- parts: Towards open-vocabulary part segmentation

    Wei, M., Yue, X., Zhang, W., Kong, S., Liu, X., Pang, J., 2024c. Ov- parts: Towards open-vocabulary part segmentation. Advances in Neural Information Processing Systems 36

  81. [89]

    Phrasecut: Language- based image segmentation in the wild, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Wu, C., Lin, Z., Cohen, S., Bui, T., Maji, S., 2020. Phrasecut: Language- based image segmentation in the wild, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 10216– 10225

    2020

  82. [90]

    See say and segment: Teaching lmms to overcome false premises, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Wu, T.H., Biamby, G., Chan, D., Dunlap, L., Gupta, R., Wang, X., Gonzalez, J.E., Darrell, T., 2024. See say and segment: Teaching lmms to overcome false premises, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 13459–13469

    2024

  83. [91]

    Gsva: Generalized segmentation via multimodal large language models, in: Pro- ceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Xia, Z., Han, D., Han, Y., Pan, X., Song, S., Huang, G., 2024. Gsva: Generalized segmentation via multimodal large language models, in: Pro- ceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 3858–3869

    2024

  84. [92]

    Youtube-vos: Sequence-to-sequence video object segmentation, in: Proceedings of the European conference on computer vision (ECCV), pp

    Xu, N., Yang, L., Fan, Y., Yang, J., Yue, D., Liang, Y., Price, B., Cohen, S., Huang, T., 2018. Youtube-vos: Sequence-to-sequence video object segmentation, in: Proceedings of the European conference on computer vision (ECCV), pp. 585–601

    2018

  85. [93]

    Visa: Reasoning video object segmentation via large language models, in: European Conference on Computer Vision, Springer

    Yan, C., Wang, H., Yan, S., Jiang, X., Hu, Y., Kang, G., Xie, W., Gavves, E., 2024. Visa: Reasoning video object segmentation via large language models, in: European Conference on Computer Vision, Springer. pp. 98– 115

    2024

  86. [94]

    Panop- tic scene graph generation, in: European Conference on Computer Vision, Springer

    Yang, J., Ang, Y.Z., Guo, Z., Zhou, K., Zhang, W., Liu, Z., 2022a. Panop- tic scene graph generation, in: European Conference on Computer Vision, Springer. pp. 178–196. 47

  87. [95]

    Video instance segmentation, in: Pro- ceedings of the IEEE/CVF international conference on computer vision, pp

    Yang, L., Fan, Y., Xu, N., 2019. Video instance segmentation, in: Pro- ceedings of the IEEE/CVF international conference on computer vision, pp. 5188–5197

    2019

  88. [96]

    Depth anything v2

    Yang, L., Kang, B., Huang, Z., et al., 2024a. Depth anything v2. arXiv preprint arXiv:2406.09414

    arXiv

  89. [97]

    An improved baseline for reasoning segmentation with large language model

    Yang, S., Qu, T., Lai, X., Tian, Z., Peng, B., Liu, S., Jia, J., 2023. An improved baseline for reasoning segmentation with large language model. CoRR

    2023

  90. [98]

    Empowering segmentation ability to multi-modal large language models

    Yang, Y., Jiang, P.T., Wang, J., Zhang, H., Zhao, K., Chen, J., Li, B., 2024b. Empowering segmentation ability to multi-modal large language models. arXiv preprint arXiv:2403.14141

    arXiv

  91. [99]

    Follow the rules: reasoning for video anomaly detection with large language models, in: European Conference on Computer Vision, Springer

    Yang, Y., Lee, K., Dariush, B., Cao, Y., Lo, S.Y., 2024c. Follow the rules: reasoning for video anomaly detection with large language models, in: European Conference on Computer Vision, Springer. pp. 304–322

  92. [100]

    Lavt: Language-aware vision transformer for referring image segmenta- tion, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Yang, Z., Wang, J., Tang, Y., Chen, K., Zhao, H., Torr, P.H., 2022b. Lavt: Language-aware vision transformer for referring image segmenta- tion, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 18155–18165

  93. [101]

    Language- guided reasoning segmentation for underwater images

    Yao, M., Tam, K.M., Wang, M., Li, L., Kawakami, R., 2025. Language- guided reasoning segmentation for underwater images. Information Fusion , 103177

    2025

  94. [102]

    Vera: Explainable video anomaly de- tection via verbalized learning of vision-language models

    Ye, M., Liu, W., He, P., 2024. Vera: Explainable video anomaly de- tection via verbalized learning of vision-language models. arXiv preprint arXiv:2412.01095

    2024 arXiv

  95. [103]

    Yu, L., Poirson, P., Yang, S., Berg, A.C., Berg, T.L., 2016. Modeling con- text in referring expressions, in: Computer Vision–ECCV 2016: 14th Eu- ropean Conference, Amsterdam, The Netherlands, October 11-14, 2016, Proceedings, Part II 14, Springer. pp. 69–85

    2016

  96. [104]

    Vision-language models for vi- sion tasks: A survey

    Zhang, J., Huang, J., Jin, S., Lu, S., 2024. Vision-language models for vi- sion tasks: A survey. IEEE Transactions on Pattern Analysis and Machine Intelligence

    2024

  97. [105]

    Zheng, R., Qi, L., Chen, X., Wang, Y., Wang, K., Qiao, Y., Zhao, H.,

  98. [106]

    Scene parsing through ade20k dataset, in: Proceedings of the IEEE con- ference on computer vision and pattern recognition, pp

    Zhou, B., Zhao, H., Puig, X., Fidler, S., Barriuso, A., Torralba, A., 2017. Scene parsing through ade20k dataset, in: Proceedings of the IEEE con- ference on computer vision and pattern recognition, pp. 633–641. 48

    2017

  99. [107]

    Egoobjects: A large-scale ego- centric dataset for fine-grained object understanding, in: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp

    Zhu, C., Xiao, F., Alvarado, A., Babaei, Y., Hu, J., El-Mohri, H., Cu- latana, S., Sumbaly, R., Yan, Z., 2023a. Egoobjects: A large-scale ego- centric dataset for fine-grained object understanding, in: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp...

  100. [108]

    Tracking with human-intent reasoning

    Zhu, J., Cheng, Z.Q., He, J.Y., Li, C., Luo, B., Lu, H., Geng, Y., Xie, X., 2023b. Tracking with human-intent reasoning. arXiv preprint arXiv:2312.17448

    arXiv

  101. [109]

    Zhu, L., Chen, T., Xu, Q., Liu, X., Ji, D., Wu, H., Soh, D.W., Liu, J.,

  102. [2016]

    Generation and comprehension of unambiguous object descriptions, in: Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 11–20

  103. [2022]

    ACM computing surveys (CSUR) 54, 1–41

    Transformers in vision: A survey. ACM computing surveys (CSUR) 54, 1–41

  104. [2023]

    Machine Intelligence Research 20, 38–56

    Vlp: A survey on vision-language pre-training. Machine Intelligence Research 20, 38–56

  105. [2024]

    arXiv preprint arXiv:2407.14500

    Villa: Video reasoning segmentation with large language model. arXiv preprint arXiv:2407.14500

    arXiv

  106. [2025]

    arXiv preprint arXiv:2504.00640

    Popen: Preference-based optimization and ensemble for lvlm-based reasoning segmentation. arXiv preprint arXiv:2504.00640 . 49

    arXiv

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.