Pith. sign in

REVIEW 3 major objections 5 minor 30 references

DreamLLM-3D: Affective Dream Reliving using Large Language Model and 3D Generative AI

T0 review · 3 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 DreamLLM-3D aims to make dream reliving affective by feeding LLM-extracted sentiment and social interaction into 3D point-cloud rendering and a responsive soundscape.

desk verdict A well-designed creative system that integrates LLM-based dream content analysis with 3D visualization, but its central claim about affective reliving rests on entirely unvalidated LLM extraction; worth a serious referee for a creative track, not evidence of efficacy. read the letter →

arxiv 2503.16439 v1 pith:6BUCQM5Q submitted 2025-02-13 cs.HC cs.AIcs.MM

Pinyao Liu , Keon Ju Lee , Alexander Steinmaurer , Claudia Picard-Deland , Michelle Carr , Alexandra Kitson This is my paper · ORCID
classification cs.HCcs.AIcs.MM
keywords dreamrelivingLLManalysisHall-VandeCastlecodingtext-to-3DgenerationaffectivecolormappingsoundscapedesignimmersiveartinstallationAI-dreamworkerhybrid
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 presents DreamLLM-3D, an immersive art installation that tries to make dream content analysis experiential rather than purely analytical. A local language model with a zero-temperature setting processes whispered dream reports as they are spoken, extracting the dream's characters and objects (entities), classifying social interactions into aggressive/friendly/sexual subclasses, and labeling the dominant emotion among five Hall–Van de Castle classes. Those extractions drive a text-to-3D diffusion model that renders each entity as a 3D point cloud, with color mapped from the detected emotion and soundscape layers blended according to emotion and social interaction. The authors argue this bridges the divide between experiential dreamwork (reliving the dream) and quantitative dream content analysis, potentially deepening personal insight and creativity. They also propose an experiential AI-Dreamworker Hybrid paradigm in which AI would observe the dreamer's behavior and adaptively guide the dreamwork process.

What carries the argument

The mechanism that carries the argument is a real-time two-stage AI pipeline. Stage one is a local zero-temperature language model (Mistral 7B) backed by an embedding model (Nomic-Embed-Text) and a cosine-similarity search library (Chroma); it outputs structured data for each whispered segment: a list of single dream entities, one dominant social-interaction subclass, and one dominant emotion class. Stage two is the text-to-3D diffusion model Point-E, which converts each entity prompt into a 3D point cloud in roughly 17 seconds on an A100 GPU; the point clouds are parsed in Unity3D for real-time rendering. The affective color mapping (five HVDC emotion classes placed on Russell's valence–arousal circumplex, each assigned a representative color) and the context-aware soundscape (a neutral base layer plus composer-designed layers for each emotion and each social-interaction class) complete the loop. This design makes every perceptual channel of the artwork a deterministic function of the LLM's structured outputs.

What would settle it

A concrete test would be to run the LLM pipeline on a corpus of whispered dream reports and compare its entity, social-interaction, and sentiment outputs against two independent human coders using the Hall–Van de Castle rules; if agreement approaches chance on any of the three modules, the color and soundscape mappings would be built on misclassified content. A second check would have raters judge whether each Point-E point cloud visibly depicts the intended entity; failures there would break the visual link between dream content and experience.

Watch

Extended reading notes

Core claim

The paper argues that a composite multimodal AI system can automate dream content analysis in real time and feed the results into an immersive, affectively colored 3D dream-reliving experience. A zero-temperature Mistral 7B language model processes each whispered dream segment by (1) segmenting the dream into single-entity prompts for the text-to-3D model, (2) classifying social interactions into HVDC subclasses—Aggression A1–A8, Friendliness F1–F7, and Sexual Interaction S1–S5—via cosine similarity between Nomic-Embed-Text embeddings of the segment and embedded definitions of each subclass, and (3) labeling the dominant dream emotion among five HVDC classes: Anger, Apprehension, Sadness, Confusion, and Happiness. Point-E renders each extracted entity as a 3D point cloud, the detected sentiment selects a color grounded in affective color psychology and Russell's Circumplex model, and the sentiment plus social-interaction classes select soundscape layers to blend over a neutral base. On this basis, the authors propose that the system is a first step toward an experiential AI-Dreamworker Hybrid in which a future AI would observe the dreamer's behavior during reliving and adaptively guide the dreamwork process.

Load-bearing premise

The entire reliving experience rests on the assumption that the language model's classification of dream entities, social interactions, and emotions is accurate enough to match what actually happened in the dream, and the paper does not report any comparison of these automated tags against human-coded dream analysis.

Editorial extensions

If this is right

  • A dreamer can relive a dream in a shared 3D space where characters and objects appear as point clouds whose color encodes the dominant emotion detected from the whisper.
  • Complex dream scenes are reconstructed entity by entity across whispering sessions, avoiding Point-E's difficulty with multi-entity prompts.
  • The soundscape changes dynamically with the dream's content: emotion layers swap as sentiment shifts, and an aggression layer blends in when social interaction is classified as aggressive.
  • If the system works as claimed, dreamwork and entry-level dream analysis become more accessible and affordable because manual HVDC coding is no longer required, while the dreamer remains the active interpretive agent.
  • The proposed AI-Dreamworker Hybrid would extend the pipeline from one-way rendering to adaptive guidance, with a future AI observing the dreamer's behavior and prompting interactions with dream entities during reliving.

Reading between the lines

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

  • A testable extension the paper does not report is a validation study: scoring the LLM's entity, interaction, and emotion tags against human HVDC coders would establish the fidelity of the entire closed loop; the current contribution rests on the assumption that the tags are reliable.
  • Because the social-interaction classifier works by embedding definitions of HVDC subclasses and matching by cosine similarity, the same pipeline could be adapted to other coding schemes or to new custom categories by swapping those definitions, a property the paper does not exploit.
  • Deployed longitudinally, the whisper stream would become a quantitative time series of emotions and social-interaction classes across nights; correlating those traces with waking-life events would operationalize the continuity hypothesis in a way the single-session installation does not.
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

3 major / 5 minor

Summary. The paper presents DreamLLM-3D, a multimodal system for immersive dream reliving. Whispered dream reports are processed by a local Mistral 7B LLM pipeline that extracts dream entities, classifies social interactions into HVDC aggressive/friendly/sexual subclasses, and classifies emotion into the five HVDC emotion classes. The extracted entities become prompts for Point-E text-to-3D generation; the emotion output drives a color mapping; and both emotion and social interaction outputs drive a layered soundscape. The paper also proposes an experiential AI-dreamworker hybrid paradigm and discusses ethical implications. No empirical evaluation is reported; the authors state in Section 3.1 that future work will evaluate the system.

Significance. If validated, the system would be a useful bridge between automated dream content analysis and experiential dreamwork, and it appears to be the first integration of on-the-fly LLM analysis with text-to-3D affective visualization. The design choices are transparent and grounded: HVDC-based categories, Russell Circumplex-informed color mapping, and per-class sound layers are all explicit. The ethical discussion is thoughtful and goes beyond token treatment. The paper does not commit clear logical errors, and the system architecture is coherent. However, the central functional claim is currently unsubstantiated: there is no benchmark of the LLM extraction, no user study, no error analysis, and no code or data release. The contribution is therefore strongest when read as a system and installation description plus design proposal, rather than as a validated demonstration.

major comments (3)
  1. [2.1, 3.1] The paper's central functional claim is that the system enables automated dream content analysis for immersive dream-reliving, yet no evaluation of the LLM pipeline is presented. There is no comparison to human HVDC coding, no accuracy, precision/recall, or inter-rater agreement statistics, no sample input/output traces, and no error analysis. The paper itself defers this explicitly in Section 3.1, stating that the authors 'plan to adapt the experience into a longitudinal re-experiencing and analysis tool and further evaluate its potential.' Because every downstream component (Point-E prompts, color mapping, soundscape layers) is a deterministic function of the LLM outputs, the unmeasured extraction accuracy is load-bearing. Please add an evaluation against human-coded dream reports, or at minimum a set of worked examples with an error analysis, and release the prompts and data needed to reproduce the pipeline; alternatively, revise the abstract and Section 3.1 to present the system as a design proposal rather than as a validated capability.
  2. [2.1] The social interaction classification module is specified incompletely. The text says that the definitions of each class and subclass were 'used as embeddings' with cosine similarity search in Chroma, but it does not specify the actual LLM prompts, the matching decision rule, the similarity threshold, or any example of the dominant-subclass output. Since this module determines which soundscape layer is blended on top of the emotion layer, the missing procedure is load-bearing for both reproducibility and correctness. Please provide the full prompting strategy, the threshold or decision rule used to select one subclass, and one or more worked examples showing how a dream snippet is mapped to, say, A3 or F2.
  3. [3.1] The experiential claims in the abstract and Section 3.1—that the system 'could potentially facilitate a more emotionally engaging dream-reliving experience, enhancing personal insights and creativity'—are unsupported by any user data. No qualitative or quantitative user study, no measure of emotional engagement or insight, no installation session logs, and no comparison with a non-affective baseline are reported. The hedged 'could potentially' is appropriate, but the same section also asserts that the system 'enables automated dream content analysis' and 'implements an automated dream sentiment and social interaction analysis.' Please include an evaluation of the experience or explicitly partition the paper's contributions into implemented system versus speculative future benefits.
minor comments (5)
  1. [2.1] The speech recognition component is not named or specified; for reproducibility, state which ASR model is used and how whispered, possibly multilingual input is handled.
  2. [2.2] The in-text citations 'Madden et al., Bartram et al.' are incomplete: the reference list gives details for Bartram et al. but not for Madden et al., and no year is given in the text for either.
  3. [2.3] The 'music composer' is not identified and the composition procedure for the neutral and per-class sound layers is not described; if the soundscape is a bespoke artistic component, that should be stated explicitly, and if it is intended to be reproducible, more detail is needed.
  4. [3.3] The ethical discussion does not state whether the installation was deployed with human participants and, if so, whether informed consent or ethics approval was obtained; this should be clarified given the intimate nature of whispered dream reports.
  5. [References] Several references are incomplete or inconsistently formatted, for example 'fuse. ONIRICA ()' and the entry for Kelly Bulkeley, which lacks a full citation; please normalize all entries to the venue's reference style.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the system's LLM extraction, 3D rendering, and affective mapping are design choices built on externally pretrained components, with no fitted parameter renamed as a prediction.

full rationale

The paper is a system/design proposal, not a derivation of predicted quantities. The LLM pipeline extracts dream entities, social interactions, and sentiment; the text-to-3D model renders entities; and color and sound layers are mapped from the extracted classes. These are implementation choices: the color map is explicitly grounded in cited color-emotion studies, the soundscape is a composer-designed layering scheme, and both are described as design decisions rather than outputs derived from first principles. No equation or fitted parameter is introduced, and no result is predicted that is equivalent by construction to the inputs. The only self-citations, e.g., Liu et al. 2024a describing a prior text-to-3D dream-reliving system, appear in the related-work motivation and are not used as load-bearing evidence for the current system's accuracy or necessity. The absence of an empirical evaluation of the LLM's extraction accuracy is a validity risk, not circularity: an unvalidated pipeline built on external pretrained models does not reduce to its own inputs. Therefore the circularity score is 0.

Assumptions & free parameters 2 free parameters · 5 assumptions · 0 invented entities

The central claim rests on the validity of the HVDC framework and on the accuracy of the LLM extraction, both assumed without validation. The color and sound mappings are hand-chosen design parameters. No new physical or conceptual entities are introduced beyond the proposed AI-Dreamworker Hybrid, which is a paradigm rather than an entity.

free parameters (2)
  • color-emotion mapping = Happiness->light blue/gold, Anger->dark red, Apprehension->purple, Sadness->dark blue, Confusion->white glow
    Hand-chosen by the authors based on affective color psychology and Russell's circumplex model; the mapping is a design choice that shapes the visualization output but is not derived from data in this paper.
  • HVDC subclass embedding definitions = Aggression A1-A8, Friendliness F1-F7, Sexual S1-S5
    The authors use the textual definitions of each subclass as embeddings for social interaction classification. The specific embedding construction is not detailed, and the choice of these definitions influences classification outcomes.
assumptions (5)
  • domain assumption Hall and Van de Castle scheme is a valid framework for dream content analysis
    The system builds on HVDC categories to define what entities, emotions, and social interactions to extract (Section 2).
  • domain assumption The LLM can accurately perform dream content extraction from whispered speech
    The entire pipeline depends on Mistral 7B correctly identifying entities, sentiment, and social interactions from short, noisy inputs, but no accuracy data is provided (Section 2.1).
  • domain assumption Color-emotion associations are cross-culturally valid
    The affective color mapping relies on cross-cultural studies of color and emotion, and on Russell's circumplex model, to assign meaning (Section 2.2).
  • domain assumption The continuity hypothesis and Social Simulation Theory hold
    The motivation for using dream content for personal insight draws on the continuity hypothesis and SST (Introduction, Section 3.1).
  • domain assumption Point-E generates relevant 3D point clouds from simple prompts
    The system uses Point-E as the text-to-3D generator and designs the prompts accordingly, assuming it produces recognizable entities (Section 2.1).

how reviews work

0 comments
Cite this review

Pith. "Pith review of DreamLLM-3D: Affective Dream Reliving using Large Language Model and 3D Generative AI." pith.science (2026). https://pith.science/paper/6BUCQM5Q

@misc{pith2026250316439,
  author       = {Pith},
  title        = {Pith review of: DreamLLM-3D: Affective Dream Reliving using Large Language Model and 3D Generative AI},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/6BUCQM5Q}},
  note         = {Machine review of arXiv:2503.16439}
}
read the original abstract

We present DreamLLM-3D, a composite multimodal AI system behind an immersive art installation for dream re-experiencing. It enables automated dream content analysis for immersive dream-reliving, by integrating a Large Language Model (LLM) with text-to-3D Generative AI. The LLM processes voiced dream reports to identify key dream entities (characters and objects), social interaction, and dream sentiment. The extracted entities are visualized as dynamic 3D point clouds, with emotional data influencing the color and soundscapes of the virtual dream environment. Additionally, we propose an experiential AI-Dreamworker Hybrid paradigm. Our system and paradigm could potentially facilitate a more emotionally engaging dream-reliving experience, enhancing personal insights and creativity.

Figures

Figures reproduced from arXiv: 2503.16439 by the authors.

Figure 1
Figure 1. During the dream-reliving experience, the immersant whispers their dream into the system. [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. A composite multimodal AI system performing real-time dream content analysis for [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Example of a graphic score for soundscape composition illustrating the integration of [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

30 extracted references · 22 canonical work pages

  1. [1]

    Bartram, A

    L. Bartram, A. Patra, and M. Stone. Affective color in visualization. In Proceedings of the 2017 CHI Conference on Human Factors in Computing Systems , CHI '17, pages 1364--1374. Association for Computing Machinery. ISBN 978-1-4503-4655-9. doi:10.1145/3025453.3026041. URL https://doi.org/10.1145/3025453.3026041

    work page arXiv 2017

  2. [2]

    Automatic Scoring of Dream Reports' Emotional Content with Large Language Models

    L. Bertolini, V. Elce, A. Michalak, G. Bernardi, and J. Weeds. Automatic scoring of dream reports' emotional content with large language models, 2023. URL http://arxiv.org/abs/2302.14828

    work page Pith review arXiv 2023

  3. [3]

    Blagrove, L

    M. Blagrove, L. Farmer, and E. Williams. The relationship of nightmare frequency and nightmare distress to well-being. Journal of Sleep Research, 13 0 (2): 0 129--136, June 2004. ISSN 0962-1105. doi:10.1111/j.1365-2869.2004.00394.x

    work page arXiv 2004

  4. [4]

    Canet Sola and V

    M. Canet Sola and V. Guljajeva. Dream Painter : Exploring creative possibilities of AI -aided speech-to-image synthesis in the interactive art context. Proc. ACM Comput. Graph. Interact. Tech., 5 0 (4): 0 33:1--33:11, Sept. 2022. doi:10.1145/3533386. URL https://dl.acm.org/doi/10.1145/3533386

    doi:10.1145/3533386 2022

  5. [5]

    G. Cortal. Sequence-to-sequence language models for character and emotion detection in dream narratives. 2024. doi:10.48550/ARXIV.2403.15486. URL https://arxiv.org/abs/2403.15486. Publisher: arXiv Version Number: 1

    work page Pith review arXiv doi:10.48550/arxiv.2403.15486 2024

  6. [6]

    G. W. Domhoff. The scientific study of dreams: Neural networks, cognitive development, and content analysis . The scientific study of dreams: Neural networks, cognitive development, and content analysis. American Psychological Association, Washington, DC, US, 2003. ISBN 978-1-55798-935-2. doi:10.1037/10463-000. Pages: ix, 209

    work page doi:10.1037/10463-000 2003

  7. [7]

    L. Ellis. Diving deep: three experiential approaches to working with dreams and nightmares. Person-Centered & Experiential Psychotherapies, Oct. 2023. ISSN 1477-9757. URL https://www.tandfonline.com/doi/abs/10.1080/14779757.2023.2169841. Publisher: Routledge

    work page arXiv 2023

  8. [8]

    L. A. Ellis. Common factors leading to a universal approach to dreamwork: A qualitative analysis. Dreaming, 29 0 (1): 0 22--39, 2019. ISSN 1573-3351. doi:10.1037/drm0000098. Place: US Publisher: Educational Publishing Foundation

    work page doi:10.1037/drm0000098 2019

Show all 30 references
  1. [9]

    Fogli, L

    A. Fogli, L. Maria Aiello, and D. Quercia. Our dreams, our selves: automatic analysis of dream reports. 7 0 (8): 0 192080, 2020. doi:10.1098/rsos.192080. URL https://royalsocietypublishing.org/doi/10.1098/rsos.192080. Publisher: Royal Society

    2020 doi

  2. [10]

    ONIRICA (), 2023

    fuse . ONIRICA (), 2023. URL https://www.fuseworks.it/works/onirica/

    2023

  3. [11]

    Gutman Music, P

    M. Gutman Music, P. Holur, and K. Bulkeley. Mapping dreams in a computational space: A phrase-level model for analyzing fight/flight and other typical situations in dream reports. 106: 0 103428, 2022. ISSN 10538100. doi:10.1016/j.concog.2022.103428. URL https://linkinghub.else...

    2022

  4. [12]

    C. S. Hall. A cognitive theory of dream symbols. 48 0 (2): 0 169--186, 1953. ISSN 0022-1309. doi:10.1080/00221309.1953.9920189. URL https://doi.org/10.1080/00221309.1953.9920189. Publisher: Routledge \_eprint: https://doi.org/10.1080/00221309.1953.9920189

    1953 arXiv

  5. [13]

    C. S. Hall and R. L. Van De Castle. The content analysis of dreams. The content analysis of dreams. Appleton-Century-Crofts, East Norwalk, CT, US, 1966. Pages: xiv, 320

    1966

  6. [14]

    Hill and S

    C. Hill and S. Knox. The Use Of Dreams In Modern Psychotherapy . International review of neurobiology, 92: 0 291--317, Dec. 2010. ISSN 9780123813220. doi:10.1016/S0074-7742(10)92013-8

    2010 doi

  7. [15]

    Horikawa and Y

    T. Horikawa and Y. Kamitani. Hierarchical Neural Representation of Dreamed Objects Revealed by Brain Decoding with Deep Neural Network Features . Frontiers in Computational Neuroscience, 11, Jan. 2017. ISSN 1662-5188. doi:10.3389/fncom.2017.00004. URL http://journal.frontiersi...

    2017

  8. [16]

    Horikawa, M

    T. Horikawa, M. Tamaki, Y. Miyawaki, and Y. Kamitani. Neural decoding of visual imagery during sleep. Science (New York, N.Y.), 340 0 (6132): 0 639--642, May 2013. ISSN 1095-9203. doi:10.1126/science.1234330

    2013 doi

  9. [17]

    A. Q. Jiang, A. Sablayrolles, A. Mensch, C. Bamford, D. S. Chaplot, D. d. l. Casas, F. Bressand, G. Lengyel, G. Lample, L. Saulnier, L. R. Lavaud, M.-A. Lachaux, P. Stock, T. L. Scao, T. Lavril, T. Wang, T. Lacroix, and W. E. Sayed. Mistral 7B , Oct. 2023. URL https://arxiv.or...

    2023 arXiv

  10. [18]

    K. N. Kay, T. Naselaris, R. J. Prenger, and J. L. Gallant. Identifying natural images from human brain activity. Nature, 452 0 (7185): 0 352--355, Mar. 2008. ISSN 0028-0836, 1476-4687. doi:10.1038/nature06713. URL https://www.nature.com/articles/nature06713

    2008 doi

  11. [19]

    Freud and Jung Sharing Their Dreams : An AI Revival , Feb

    Kelly Bulkeley . Freud and Jung Sharing Their Dreams : An AI Revival , Feb. 2024. URL https://www.psychologytoday.com/ca/blog/dreaming-in-the-digital-age/202402/freud-and-jung-sharing-their-dreams-an-ai-revival

    2024

  12. [20]

    P. Liu, A. Kitson, C. Picard-Deland, M. Carr, S. Liu, R. Lc, and C. Zhu-Tian. Virtual Dream Reliving : Exploring Generative AI in Immersive Environment for Dream Re -experiencing. In Extended Abstracts of the 2024 CHI Conference on Human Factors in Computing Systems , CHI EA '...

    2024

  13. [21]

    S. Liu, K. Fu, X. Zeng, and R. Lc. Falling Echoes : Expressing the Act of Falling in Dreams Through Generative AI . In 29th International Symposium on Electronic Art ( ISEA ) , 2024 b . URL https://scholars.cityu.edu.hk/en/publications/falling-echoes(2efd7390-0306-49a6-994b-3e...

    2024

  14. [22]

    T. J. Madden, K. Hewett, and M. S. Roth. Managing images in different cultures: A cross-national study of color meanings and preferences. 8 0 (4): 0 90--107. ISSN 1069-031X. URL https://www.jstor.org/stable/25048831. Publisher: American Marketing Association

  15. [23]

    Miyawaki, H

    Y. Miyawaki, H. Uchida, O. Yamashita, M.-a. Sato, Y. Morito, H. C. Tanabe, N. Sadato, and Y. Kamitani. Visual Image Reconstruction from Human Brain Activity using a Combination of Multiscale Local Image Decoders . Neuron, 60 0 (5): 0 915--929, Dec. 2008. ISSN 0896-6273. doi:10...

    2008 doi

  16. [24]

    Nichol, H

    A. Nichol, H. Jun, P. Dhariwal, P. Mishkin, and M. Chen. Point- E : A System for Generating 3D Point Clouds from Complex Prompts , Dec. 2022. URL https://arxiv.org/abs/2212.08751v1

    2022 arXiv

  17. [25]

    Nussbaum, J

    Z. Nussbaum, J. X. Morris, B. Duderstadt, and A. Mulyar. Nomic Embed : Training a Reproducible Long Context Text Embedder . 2024. doi:10.48550/ARXIV.2402.01613. URL https://arxiv.org/abs/2402.01613. Publisher: arXiv Version Number: 1

    2024 doi arXiv

  18. [26]

    Picard-Deland

    C. Picard-Deland. Vestibular function and lucid dreaming: the case of flying dreams. In Lucid dreaming: state of the art and current developments, page 1, Rio de Janeiro, Brazil., Oct. 2023

    2023

  19. [27]

    Plato . Apology. In B. Jowett, editor, Dialogues of Plato : With analyses and introductions, Vol 1. , pages 305--339. Charles Scribner's Sons, New York, NY, 1901. doi:10.1037/13728-009

    1901 doi

  20. [28]

    Revonsuo, J

    A. Revonsuo, J. Tuominen, and K. Valli. The Avatars in the MachineThe Avatars in the Machine : Dreaming as a Simulation of Social Reality : Dreaming as a Simulation of Social Reality . In Open MIND . Theoretical Philosophy/MIND Group – JGU Mainz, 2015. doi:10.15502/97839585703...

    2015 doi

  21. [29]

    J. A. Russell. A circumplex model of affect. Journal of Personality and Social Psychology, 39 0 (6): 0 1161--1178, 1980. ISSN 1939-1315. doi:10.1037/h0077714. Place: US Publisher: American Psychological Association

    1980 doi

  22. [30]

    Takagi and S

    Y. Takagi and S. Nishimoto. High-resolution image reconstruction with latent diffusion models from human brain activity. In 2023 IEEE / CVF Conference on Computer Vision and Pattern Recognition ( CVPR ) , pages 14453--14463, June 2023. doi:10.1109/CVPR52729.2023.01389. URL htt...

    2023

Pith tools

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