Pith. sign in

REVIEW 4 major objections 8 minor 41 references

Vision language models have difficulty recognizing virtual objects

T0 review · 4 major / 8 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read State-of-the-art vision-language models systematically fail to recognize virtual objects—objects described in a prompt but not shown in an image.

desk verdict New diagnostic for VLM scene-updating, but the main claim needs a no-virtual-object control and human baselines before it can be believed. read the letter →

arxiv 2505.10453 v1 pith:FYTHZPAN submitted 2025-05-15 cs.CV cs.AI

Tyler Tran , Sangeet Khemlani , J.G. Trafton This is my paper · ORCID
classification cs.CVcs.AI
keywords virtualobjectsvisionlanguagemodelsscenecomprehensionspatialreasoninghypotheticalmultimodalgroundingimaginationevaluationbenchmark
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 claims that state-of-the-art vision-language models cannot reliably recognize virtual objects: objects that are described in a prompt but not shown in the image. When asked to imagine a third object next to the two objects in a tabletop image and then list what is on the table, the models often fail to name all three. In 112,896 queries across three models, aggregate accuracy reached 63% for Idefics2, 57% for Llama3, and 22% for BLIP. Accuracy shifted with irrelevant prompt features: past-tense prompts outperformed present-tense ones, and adding a numerical cue such as "what three items" raised accuracy from 32% to 62%. A sympathetic reading is that this exposes a genuine limitation in how VLMs update their scene representations from linguistic information.

What carries the argument

The load-bearing object is the "virtual object" probe: a prompt such as "Imagine there is a banana next to the cupcake" paired with a two-object image, followed by a question asking what items are on the table. The probe's work is to force a VLM to merge a textual hypothetical with the visual scene and report the union; the only success criterion is listing all three objects in any order. The evaluation machinery is the TABLETEST image set of 64 objects in two-object configurations, seven prompt templates, past and present tenses, and numerical-cue variation, yielding 112,896 queries per model at temperature zero.

What would settle it

Re-run the battery with a control condition that separates content from format, for example asking "Name the objects in the image, then name the object you imagined," and measure human accuracy on the original prompts; if model accuracy approaches ceiling under the template, or if humans fall well below ceiling, the inference that VLMs lack virtual-object tracking would be substantially weakened.

Watch

Extended reading notes

Core claim

The central discovery is a systematic failure pattern: VLMs lose track of entities introduced by hypothetical language. The authors define a virtual object as an object within a scene that is described but not depicted, and argue that a system with robust scene comprehension should incorporate it and list it alongside the two depicted objects. Across seven prompt phrasings, two tenses, and numerical-cue variations, no tested model performed near ceiling; the best aggregate score was Idefics2 at 63%, BLIP fell to 22%, and on the "if" phrasing BLIP managed only 8%. The paper also shows that factors that should be irrelevant—tense and the presence of a number in the question—produced reliable differences in accuracy. The authors conclude that current VLMs do not encode scene structure in a way that supports primitive hypothetical or imaginative updating.

Load-bearing premise

The evaluation assumes the only correct answer is to name all three objects in any order, and that every prompt is unambiguously requesting that list; if some failures come from answer formatting or wording ambiguity rather than from losing track of the imagined object, the conclusion is too strong.

Editorial extensions

If this is right

  • If the claim holds, VLMs cannot be trusted for hypothetical spatial reasoning such as replacing one object with another or moving an object relative to another, because those tasks require the same scene-updating operation.
  • Irrelevant surface features of a prompt—tense and numeric wording—materially change accuracy, so current performance on such tasks is not a stable measure of visuospatial ability.
  • The numerical-cue boost from 32% to 62% suggests that models may sometimes answer by matching the count of nouns in the prompt rather than genuinely updating their representation of the scene.
  • The near-zero performance of BLIP on the "if" phrasing indicates that particular function words can essentially disable object tracking in some architectures.

Reading between the lines

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

  • A plausible extension is that similar limitations will appear in instruction-following robots or navigation agents that must act on objects they have only been told about, since those systems rely on the same kind of fused text-image representations.
  • A direct test would collect human accuracy on the same battery; the paper assumes human performance is at ceiling but never measures it, so the true gap between human and model performance remains open.
  • It would be informative to probe decoding sensitivity by running the same prompts at higher temperatures or with multiple random seeds; if accuracy varies widely, part of the apparent failure could be response-sampling noise rather than a fixed representational limit.
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

4 major / 8 minor

Summary. This paper argues that vision language models (VLMs) fail at a rudimentary imaginative task: given an image of two objects and a prompt such as 'Imagine there is a banana next to the cupcake', the model is asked to list all objects on the table, where the banana is a 'virtual object' described but not depicted. The authors construct 112,896 prompt-image queries from the TABLETEST dataset (4,032 two-object images x 7 prompt formulations x 2 tenses x 2 numerical-cue variants) and evaluate three open-weights VLMs: Idefics2, InstructBLIP-Vicuna, and Llama 3.2. Aggregate accuracies are 63%, 57%, and 22%, respectively, with significant effects of prompt wording, tense, and numerical cues. The authors conclude that VLMs have inadequate ability to process virtual objects and that this reflects a broader limitation in visuospatial scene comprehension and representation updating.

Significance. The phenomenon investigated is timely and the experimental matrix is large and systematic. The paper makes a falsifiable claim: state-of-the-art open-weights VLMs will often fail to list an object that is described but not depicted, even when the instruction explicitly asks for all objects. If the evidence were airtight, this would be a useful benchmark result for the VLM community. The paper's strengths include the use of fixed random seeds and temperature 0 for replicability, the variation of prompt formulations, and the demonstration that superficial lexical factors (tense, numerical cues) change performance by up to 30 percentage points. However, the central attribution of the failures to virtual objects is currently under-supported: there is no control condition without a virtual object, no human baseline, no chance baseline, and no description of the scoring procedure. These omissions are load-bearing and should be addressed in revision.

major comments (4)
  1. [Section 2; Section 3] The claim that models 'systematically lost track of the virtual objects' (Section 3, last paragraph) requires a control condition in which the same images are presented without a virtual-object instruction. Without such a control, the reported aggregate accuracies (63%, 57%, 22%) conflate at least three failure modes: failing to name a visible object, failing to add the virtual object to the response, and failing to produce the required list format. The selection criterion in Section 2 ('capable of out-of-the-box, single-shot identification of the 64 objects in TABLETEST') is asserted but not quantified on the actual 2-object stimulus set. Please add a no-virtual-object control condition (e.g., 'What items are on the table?') on the same images and report per-model visible-object identification accuracy.
  2. [Section 2, paragraph 4] The success criterion is 'list all three objects in any order', but the scoring procedure is never specified. The reader does not know how responses were normalized, whether exact string matching was used, whether synonyms, extra words, or multi-sentence answers were accepted, or whether any human review was conducted. Because every accuracy number in the paper depends on this decision, the scoring pipeline (including any regex, LLM-as-judge, or manual annotation protocol) must be described in detail, and the response data should be released to allow independent verification.
  3. [Section 3, Figures 2 and 3] The assertion 'Humanlike performance estimated at ceiling (accuracy = 1.0)' is not backed by any human data, and no chance baseline is provided. Without human judgments, the premise that the correct answer is 'trivial' and that the prompt phrasings are unambiguous is an untested assumption; without a chance model, the low accuracies could be explained by a generic response-length or list-completion tendency of the decoders. Please run a small human study on a random subset of trials and report a chance baseline (e.g., uniform sampling from the 64-object vocabulary, or a model-free heuristic). These data are needed to support the claim that the failures are specific to virtual-object processing.
  4. [Section 3, second paragraph] The statistical analyses are under-specified. The text says the tests were performed on 'mean accuracies for the different objects' but does not state how many images contributed to each object's mean, how zero-answer or malformed responses were counted, or whether the Friedman test treated the three models as repeated measurements over the same 64 objects. Moreover, the tense (51% vs. 44%) and numerical-cue (62% vs. 32%) effects are reported only in aggregate; per-model results, effect sizes, and confidence intervals are needed to assess whether the effects are driven primarily by one model (e.g., BLIP at 22% overall).
minor comments (8)
  1. [Abstract] In the abstract, 'F or example' should be 'For example'.
  2. [Section 2, third paragraph] The text reads 'Architecture sthat matched'; this should be 'Architectures that matched'.
  3. [References] Reference [35] is cited for Llama 3.2 (11B), but the cited work is Touvron et al. (2023), the Llama 2 paper; either update the citation to the Llama 3 model card or correct the model name in the text.
  4. [Table 1] The numerical-cue conditions mentioned in the text ('what three items are on the table?') are not shown in Table 1; include an example of a numeric-cue prompt to make the manipulation concrete.
  5. [Figures 2 and 3] The captions are hard to parse (e.g., 'bars depict histograms across those objects, as organized by whether the object served as the leftmost object in images'); please clarify the figure design and the meaning of the y-axis labels and the percentages above each panel.
  6. [Section 3, second paragraph] The phrase 'Friedman test, chi-squared = 120.91 p < .001' is missing a comma between the statistic and the p-value; also report degrees of freedom for the chi-squared statistics.
  7. [Section 3, third paragraph] The claim that 'none of the factors we tested should have affected whether a VLM can detect virtual objects' seems to conflict with the hypothesis in Section 2 (item 3) that numerical cues should boost performance; clarify whether the factors are expected to affect detection or only the response format.
  8. [Section 2] The paper does not state whether trials in which the virtual object's name matches one of the depicted objects were excluded; such trials could create ambiguity in scoring and should be handled explicitly.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation: the evaluation measures external VLMs against a fixed benchmark; the self-cited TABLETEST dataset supplies stimuli, not the conclusion.

full rationale

This paper is an empirical benchmark, not a derivation chain. The claimed inadequacy of VLMs at virtual-object recognition is read off the models' actual responses to 112,896 queries, so the result is not forced by the authors' definitions or by any fitted parameter. The methodologically relevant self-citation is [22], which provides the TABLETEST image set used as stimuli; that citation does not supply the model accuracies, and the VLMs are external systems whose outputs are independently observable. The scoring criterion (list all three objects) is the paper's operationalization of success, but it is not an equation that entails the measured 63%, 57%, and 22% accuracies. Concerns about the absence of a no-virtual-object control and the asserted human ceiling are methodological threats to construct validity, not circular reductions of predictions into inputs. Accordingly, no circular step can be quoted or exhibited, and the appropriate finding is no significant circularity.

Assumptions & free parameters 0 free parameters · 3 assumptions · 1 invented entities

The central claim depends on three domain assumptions: the scoring criterion is a valid measure of virtual object tracking; human performance is at ceiling; and the selected models can reliably identify the dataset's objects. No free parameters are fitted; the 'virtual object' construct is a new conceptual label introduced by the paper.

assumptions (3)
  • domain assumption The success criterion for the benchmark is that the model lists all three objects in any order.
    The paper defines correctness this way in Section 2 but never validates that this is the unique natural response for the task; a model could produce a correct answer in a different format and be scored wrong.
  • domain assumption Human performance on the task is at ceiling.
    Section 3 estimates human performance at 1.0 without collecting human data on the actual prompt set.
  • domain assumption The three VLMs can reliably identify the 64 TABLETEST objects out of the box.
    This is the selection criterion in Section 2, but no verification results or accuracy numbers on object identification are provided.
invented entities (1)
  • virtual object
    purpose: A benchmark construct: an object described in the prompt but not depicted in the image, used to test whether VLMs update their scene representation.
    The paper introduces this category as a test for scene comprehension, but it is a descriptive construct rather than an independently observable entity; there is no external falsifiable handle.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Vision language models have difficulty recognizing virtual objects." pith.science (2026). https://pith.science/paper/FYTHZPAN

@misc{pith2026250510453,
  author       = {Pith},
  title        = {Pith review of: Vision language models have difficulty recognizing virtual objects},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/FYTHZPAN}},
  note         = {Machine review of arXiv:2505.10453}
}
read the original abstract

Vision language models (VLMs) are AI systems paired with both language and vision encoders to process multimodal input. They are capable of performing complex semantic tasks such as automatic captioning, but it remains an open question about how well they comprehend the visuospatial properties of scenes depicted in the images they process. We argue that descriptions of virtual objects -- objects that are not visually represented in an image -- can help test scene comprehension in these AI systems. For example, an image that depicts a person standing under a tree can be paired with the following prompt: imagine that a kite is stuck in the tree. VLMs that comprehend the scene should update their representations and reason sensibly about the spatial relations between all three objects. We describe systematic evaluations of state-of-the-art VLMs and show that their ability to process virtual objects is inadequate.

Figures

Figures reproduced from arXiv: 2505.10453 by the authors.

Figure 1
Figure 1. Examples from the TABLETEST dataset, which includes 64 individual objects (a) in various 2-object (b) and 3-object con￾figurations (c). [38] and some researchers have investigated how they can be used to generate imaginary scenes [41] and configura￾tions of objects [32]. These models are trained on webscale image caption corpora to extract visuospatial information from out-of-distribution images, yielding possible h… view at source ↗
Figure 2
Figure 2. Proportions of accuracy from evaluations of 7 separate prompts concerning virtual object recognition in 2-object images from [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Proportions of accuracy from evaluations of [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

41 extracted references · 27 canonical work pages

  1. [1]

    The imaginative mind.Human brain map- ping, 37(11):4197–4211, 2016

    Anna Abraham. The imaginative mind.Human brain map- ping, 37(11):4197–4211, 2016. 1

    work page 2016

  2. [2]

    Mapping the imaginative mind: Charting new paths forward.Current Directions in Psychological Science, 30(1):82–89, 2021

    Jessica R Andrews-Hanna and Matthew D Grilli. Mapping the imaginative mind: Charting new paths forward.Current Directions in Psychological Science, 30(1):82–89, 2021. 1 4

    work page 2021

  3. [3]

    Humans predict liquid dynamics using probabilistic simulation

    Christopher J Bates, Iker Yildirim, Joshua B Tenenbaum, and Peter W Battaglia. Humans predict liquid dynamics using probabilistic simulation. InProceedings of the Annual Meet- ing of the Cognitive Science Society, 2015. 1

    work page 2015

  4. [4]

    Non- commitment in mental imagery.Cognition, 238:105498,

    Eric J Bigelow, John P McCoy, and Tomer D Ullman. Non- commitment in mental imagery.Cognition, 238:105498,

    work page

  5. [5]

    Infusing perception with imagination.Per- ceptual imagination and perceptual memory, pages 133– 160, 2018

    Derek H Brown. Infusing perception with imagination.Per- ceptual imagination and perceptual memory, pages 133– 160, 2018. 1

    work page 2018

  6. [6]

    Spatialbot: Precise spatial understanding with vision language models

    Wenxiao Cai, Iaroslav Ponomarenko, Jianhao Yuan, Xiaoqi Li, Wankou Yang, Hao Dong, and Bo Zhao. Spatialbot: Precise spatial understanding with vision language models. arXiv preprint arXiv:2406.13642, 2024. 1

    arXiv 2024

  7. [7]

    The artist as neuroscientist.Nature, 434 (7031):301–307, 2005

    Patrick Cavanagh. The artist as neuroscientist.Nature, 434 (7031):301–307, 2005. 4

    work page 2005

  8. [8]

    Spatialvlm: Endow- ing vision-language models with spatial reasoning capabili- ties

    Boyuan Chen, Zhuo Xu, Sean Kirmani, Brain Ichter, Dorsa Sadigh, Leonidas Guibas, and Fei Xia. Spatialvlm: Endow- ing vision-language models with spatial reasoning capabili- ties. InProceedings of the IEEE/CVF Conference on Com- puter Vision and Pattern Recognition, pages 14455–14465,

Show all 41 references
  1. [9]

    Large language models are visual reasoning coordinators.Ad- vances in Neural Information Processing Systems, 36, 2024

    Liangyu Chen, Bo Li, Sheng Shen, Jingkang Yang, Chun- yuan Li, Kurt Keutzer, Trevor Darrell, and Ziwei Liu. Large language models are visual reasoning coordinators.Ad- vances in Neural Information Processing Systems, 36, 2024

    2024

  2. [10]

    Spatial- rgpt: Grounded spatial reasoning in vision language model

    An-Chieh Cheng, Hongxu Yin, Yang Fu, Qiushan Guo, Rui- han Yang, Jan Kautz, Xiaolong Wang, and Sifei Liu. Spatial- rgpt: Grounded spatial reasoning in vision language model. arXiv preprint arXiv:2406.01584, 2024. 1

    2024 arXiv

  3. [11]

    What makes mental modeling difficult? normative data for the multidimensional relational reasoning task.Frontiers in psychology, 12:668256, 2021

    Robert A Cortes, Adam B Weinberger, Griffin A Colaizzi, Grace F Porter, Emily L Dyke, Holly O Keaton, Dakota L Walker, and Adam E Green. What makes mental modeling difficult? normative data for the multidimensional relational reasoning task.Frontiers in psychology, 12:668256, 2021. 4

    2021

  4. [12]

    A survey on multimodal large lan- guage models for autonomous driving

    Can Cui, Yunsheng Ma, Xu Cao, Wenqian Ye, Yang Zhou, Kaizhao Liang, Jintai Chen, Juanwu Lu, Zichong Yang, Kuei-Da Liao, et al. A survey on multimodal large lan- guage models for autonomous driving. InProceedings of the IEEE/CVF Winter Conference on Applications of Computer Vis...

    2024

  5. [13]

    Objaverse: A universe of annotated 3d objects

    Matt Deitke, Dustin Schwenk, Jordi Salvador, Luca Weihs, Oscar Michel, Eli VanderBilt, Ludwig Schmidt, Kiana Ehsani, Aniruddha Kembhavi, and Ali Farhadi. Objaverse: A universe of annotated 3d objects. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Reco...

    2023

  6. [14]

    Psychology press, 2014

    Ronald A Finke.Creative imagery: Discoveries and inven- tions in visualization. Psychology press, 2014. 1

    2014

  7. [15]

    Scene-llm: Extending language model for 3d visual understanding and reasoning.arXiv preprint arXiv:2403.11401, 2024

    Rao Fu, Jingyu Liu, Xilun Chen, Yixin Nie, and Wen- han Xiong. Scene-llm: Extending language model for 3d visual understanding and reasoning.arXiv preprint arXiv:2403.11401, 2024. 1

    2024 arXiv

  8. [16]

    Exploring the frontier of vision- language models: A survey of current methodologies and future directions.arXiv preprint arXiv:2404.07214, 2024

    Akash Ghosh, Arkadeep Acharya, Sriparna Saha, Vinija Jain, and Aman Chadha. Exploring the frontier of vision- language models: A survey of current methodologies and future directions.arXiv preprint arXiv:2404.07214, 2024. 1

    2024

  9. [17]

    Mental animation: Inferring motion from static displays of mechanical systems.Journal of experi- mental psychology: learning, memory, and cognition, 18(5): 1084, 1992

    Mary Hegarty. Mental animation: Inferring motion from static displays of mechanical systems.Journal of experi- mental psychology: learning, memory, and cognition, 18(5): 1084, 1992. 1, 4

    1992

  10. [18]

    Components of spatial intelligence

    Mary Hegarty. Components of spatial intelligence. InPsy- chology of learning and motivation, pages 265–297. Else- vier, 2010. 4

    2010

  11. [19]

    Correctness comparison of chatgpt-4, gemini, claude-3, and copilot for spatial tasks.Transactions in GIS, 2024

    Hartwig H Hochmair, Levente Juh ´asz, and Takoda Kemp. Correctness comparison of chatgpt-4, gemini, claude-3, and copilot for spatial tasks.Transactions in GIS, 2024. 1

    2024

  12. [20]

    Vcoder: Ver- satile vision encoders for multimodal large language models

    Jitesh Jain, Jianwei Yang, and Humphrey Shi. Vcoder: Ver- satile vision encoders for multimodal large language models. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 27992–28002, 2024. 1

    2024

  13. [21]

    Imagery, visualization, and think- ing.Perception and cognition at century’s end, pages 441– 467, 1998

    Philip N Johnson-Laird. Imagery, visualization, and think- ing.Perception and cognition at century’s end, pages 441– 467, 1998. 1, 4

    1998

  14. [22]

    Harrison, Wallace E

    Sangeet Khemlani, Tyler Tran, Nathaniel Gyory, An- thony M. Harrison, Wallace E. Lawson, Ravenna Thielstrom, Hunter Thompson, Taaren Singh, and J. Gregory Trafton. Vision language models are unreliable at trivial spatial cog- nition. InProceedings of the International Conferen...

    2025

  15. [23]

    Kinematic mental sim- ulations in abduction and deduction.proceedings of the na- tional academy of sciences, 110(42):16766–16771, 2013

    Sangeet Suresh Khemlani, Robert Mackiewicz, Monica Buc- ciarelli, and Philip N Johnson-Laird. Kinematic mental sim- ulations in abduction and deduction.proceedings of the na- tional academy of sciences, 110(42):16766–16771, 2013. 4

    2013

  16. [24]

    Mit Press, 2013

    Markus Knauff.Space to reason: A spatial theory of human thought. Mit Press, 2013. 4

    2013

  17. [25]

    Visual imagery can impede reasoning.Memory & cognition, 30:363–371,

    Markus Knauff and Phil N Johnson-Laird. Visual imagery can impede reasoning.Memory & cognition, 30:363–371,

  18. [26]

    tracking

    Maria Kon and Sangeet Khemlani. How spatial simulations distinguish “tracking” verbs. InProceedings of the Annual Meeting of the Cognitive Science Society, 2024. 4

    2024

  19. [27]

    What matters when building vision-language models? arXiv preprint arXiv:2405.02246, 2024

    Hugo Laurenc ¸on, L´eo Tronchon, Matthieu Cord, and Victor Sanh. What matters when building vision-language models? arXiv preprint arXiv:2405.02246, 2024. 2

    2024 arXiv

  20. [28]

    BLIP: Bootstrapping language-image pre-training for unified vision-language understanding and generation

    Junnan Li, Dongxu Li, Caiming Xiong, and Steven Hoi. BLIP: Bootstrapping language-image pre-training for unified vision-language understanding and generation. InInterna- tional conference on machine learning, pages 12888–12900. PMLR, 2022. 2

    2022

  21. [29]

    A survey of attacks on large vision- language models: Resources, advances, and future trends

    Daizong Liu, Mingyu Yang, Xiaoye Qu, Pan Zhou, Yu Cheng, and Wei Hu. A survey of attacks on large vision- language models: Resources, advances, and future trends. arXiv preprint arXiv:2407.07403, 2024. 1

    2024 arXiv

  22. [30]

    Visual spa- tial reasoning.Transactions of the Association for Computa- tional Linguistics, 11:635–651, 2023

    Fangyu Liu, Guy Emerson, and Nigel Collier. Visual spa- tial reasoning.Transactions of the Association for Computa- tional Linguistics, 11:635–651, 2023

    2023

  23. [31]

    A survey on hallucination in large vision-language models

    Hanchao Liu, Wenyuan Xue, Yifei Chen, Dapeng Chen, Xiu- tian Zhao, Ke Wang, Liping Hou, Rongjun Li, and Wei Peng. A survey on hallucination in large vision-language models. arXiv preprint arXiv:2402.00253, 2024. 1

    2024 arXiv

  24. [32]

    5 Enhancing visual reasoning with autonomous imagination in multimodal large language models.arXiv preprint arXiv:2411.18142, 2024

    Jingming Liu, Yumeng Li, Boyuan Xiao, Yichang Jian, Ziang Qin, Tianjia Shao, Yao-Xiang Ding, and Kun Zhou. 5 Enhancing visual reasoning with autonomous imagination in multimodal large language models.arXiv preprint arXiv:2411.18142, 2024. 1, 4

    2024

  25. [33]

    The human imagination: the cognitive neu- roscience of visual mental imagery.Nature reviews neuro- science, 20(10):624–634, 2019

    Joel Pearson. The human imagination: the cognitive neu- roscience of visual mental imagery.Nature reviews neuro- science, 20(10):624–634, 2019. 1

    2019

  26. [34]

    Mental rotation of three-dimensional objects.Science, 171(3972):701–703,

    Roger N Shepard and Jacqueline Metzler. Mental rotation of three-dimensional objects.Science, 171(3972):701–703,

  27. [35]

    Llama 2: Open foundation and fine-tuned chat models.arXiv preprint arXiv:2307.09288, 2023

    Hugo Touvron, Louis Martin, Kevin Stone, Peter Albert, Amjad Almahairi, Yasmine Babaei, Nikolay Bashlykov, Soumya Batra, Prajjwal Bhargava, Shruti Bhosale, et al. Llama 2: Open foundation and fine-tuned chat models.arXiv preprint arXiv:2307.09288, 2023. 2

    2023 arXiv

  28. [36]

    Visuospatial reasoning

    Barbara Tversky. Visuospatial reasoning. InHandbook of reasoning, pages 209–249. Cambridge University Press,

  29. [37]

    Learning physical parameters from dynamic scenes.Cognitive psychology, 104:57–82,

    Tomer D Ullman, Andreas Stuhlm ¨uller, Noah D Goodman, and Joshua B Tenenbaum. Learning physical parameters from dynamic scenes.Cognitive psychology, 104:57–82,

  30. [38]

    Multimodal large language models: A sur- vey

    Jiayang Wu, Wensheng Gan, Zefeng Chen, Shicheng Wan, and Philip S Yu. Multimodal large language models: A sur- vey. In2023 IEEE International Conference on Big Data (BigData), pages 2247–2256. IEEE, 2023. 1, 4

    2023

  31. [39]

    Jiannan Wu, Muyan Zhong, Sen Xing, Zeqiang Lai, Zhaoyang Liu, Zhe Chen, Wenhai Wang, Xizhou Zhu, Lewei Lu, Tong Lu, et al. Visionllm v2: An end-to-end general- ist multimodal large language model for hundreds of vision- language tasks.Advances in Neural Information Processing ...

    2024

  32. [40]

    Vision-language models for vision tasks: A survey.IEEE Transactions on Pattern Analysis and Machine Intelligence,

    Jingyi Zhang, Jiaxing Huang, Sheng Jin, and Shijian Lu. Vision-language models for vision tasks: A survey.IEEE Transactions on Pattern Analysis and Machine Intelligence,

  33. [41]

    Imaginenav: Prompting vision-language models as embod- ied navigator through scene imagination.arXiv preprint arXiv:2410.09874, 2024

    Xinxin Zhao, Wenzhe Cai, Likun Tang, and Teng Wang. Imaginenav: Prompting vision-language models as embod- ied navigator through scene imagination.arXiv preprint arXiv:2410.09874, 2024. 1, 4 6

    2024 arXiv

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.