pith. machine review for the scientific record. sign in

arxiv: 2604.13242 · v2 · submitted 2026-04-14 · 💻 cs.CY · cs.AI

Recognition: unknown

On the Creativity of AI Agents

Giorgio Franceschelli, Mirco Musolesi

Authors on Pith no claims yet

Pith reviewed 2026-05-10 13:37 UTC · model grok-4.3

classification 💻 cs.CY cs.AI
keywords AI agentscreativitylarge language modelsfunctionalist perspectiveontological perspectiveartificial intelligence
0
0 comments X

The pith

LLM agents produce creative outputs but lack the personal and social processes that mark human creativity.

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

The paper divides creativity into two views for analyzing LLM-based agents. One view judges creativity by the traits of the finished output. The other view requires looking at the internal processes plus the social and personal involvement that produced it. The authors conclude that current agents reach the first standard at moderate levels but miss essential parts of the second. This distinction shapes arguments about whether such agents should be developed further or used in creative roles.

Core claim

LLM agents exhibit functionalist creativity, defined through observable characteristics of their outputs, though not at the highest levels of sophistication, yet they lack key aspects of ontological creativity that involve underlying processes along with social and personal dimensions. The paper evaluates whether it is desirable for agentic systems to achieve both forms, weighing potential benefits against risks and outlining possible development paths toward forms of artificial creativity that support human society.

What carries the argument

The functionalist-ontological split, where functionalist creativity is judged solely by output characteristics while ontological creativity requires examination of processes and social-personal dimensions.

If this is right

  • Current LLM agents can already be deployed for moderate-level creative tasks based on output quality alone.
  • Attaining ontological creativity in agents would require new design approaches beyond scaling language models.
  • Risks of agents gaining both forms of creativity include potential misalignment with human values or over-reliance by users.
  • Pathways exist to steer development so that artificial creativity augments rather than replaces human creative activity.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The split implies that tests for AI creativity should include long-term interaction studies rather than single-output evaluations.
  • It raises the question of whether adding memory or embodiment modules could address the missing ontological elements.
  • The paper's evaluation of benefits and risks could extend to regulatory questions about labeling AI-generated creative work.

Load-bearing premise

That creativity in AI can be exhaustively separated into observable output traits on one side and deeper personal-social processes on the other, with current agents clearly failing the second side.

What would settle it

An empirical case of an LLM agent demonstrating creative work that includes independent personal reflection or social negotiation not reducible to pattern matching from training data.

read the original abstract

Large language models (LLMs), particularly when integrated into agentic systems, have demonstrated human- and even superhuman-level performance across multiple domains. Whether these systems can truly be considered creative, however, remains a matter of debate, as conclusions heavily depend on the definitions, evaluation methods, and specific use cases employed. In this paper, we analyse creativity along two complementary macro-level perspectives. The first is a functionalist perspective, focusing on the observable characteristics of creative outputs. The second is an ontological perspective, emphasising the underlying processes, as well as the social and personal dimensions involved in creativity. We focus on LLM agents and we argue that they exhibit functionalist creativity, albeit not at its most sophisticated levels, while they continue to lack key aspects of ontological creativity. Finally, we discuss whether it is desirable for agentic systems to attain both forms of creativity, evaluating potential benefits and risks, and proposing pathways toward artificial creativity that can enhance human society.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 2 minor

Summary. The manuscript analyzes creativity in LLM agents using two complementary perspectives: functionalist, which focuses on observable characteristics of creative outputs, and ontological, which emphasizes underlying processes along with social and personal dimensions. The authors argue that LLM agents exhibit functionalist creativity, though not at the highest levels, but lack key aspects of ontological creativity. They conclude by discussing the desirability of agents attaining both forms and proposing pathways for artificial creativity that benefits human society.

Significance. This work contributes to the ongoing debate on AI creativity by providing a dual-perspective framework. If the ontological limitations are rigorously demonstrated, it could influence how we design and evaluate agentic systems, potentially steering development towards more socially embedded creative processes. The balanced discussion of benefits and risks is a strength, offering practical considerations for the field.

major comments (1)
  1. [Ontological perspective] The ontological perspective section defines ontological creativity via high-level references to underlying processes, social embedding, and personal dimensions but supplies no explicit necessary conditions, thresholds, counter-examples, or evaluation criteria. This renders the central claim that LLM agents 'continue to lack key aspects' an interpretive stance rather than a verifiable result, as alternative framings (e.g., extended functionalist accounts incorporating simulated social roles) are not ruled out.
minor comments (2)
  1. [Abstract] The abstract is lengthy and could be condensed while preserving the core arguments and conclusions.
  2. [Introduction] Some citations to prior literature on creativity definitions could be expanded with more recent or field-specific references for completeness.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their constructive and insightful feedback, which helps clarify how to strengthen the rigor of our analysis. We respond to the single major comment below and will revise the manuscript accordingly.

read point-by-point responses

  1. Referee: [Ontological perspective] The ontological perspective section defines ontological creativity via high-level references to underlying processes, social embedding, and personal dimensions but supplies no explicit necessary conditions, thresholds, counter-examples, or evaluation criteria. This renders the central claim that LLM agents 'continue to lack key aspects' an interpretive stance rather than a verifiable result, as alternative framings (e.g., extended functionalist accounts incorporating simulated social roles) are not ruled out.

    Authors: We appreciate the referee highlighting the need for greater explicitness in the ontological section. Our framing draws from established literature (Boden on process creativity, Csikszentmihalyi on social systems, and accounts of personal embodiment), but we agree it was presented at a high level without enumerated necessary conditions or direct counter-examples. In the revised manuscript we will add a dedicated subsection that states three necessary conditions for ontological creativity: (1) intrinsic motivation arising from an agent's own persistent goals rather than prompt-driven optimization, (2) genuine social embedding through reciprocal, accountable interactions within a cultural community that can confer or withhold status, and (3) personal dimensions including embodied experience and autobiographical continuity. We will supply counter-examples contrasting LLM agents with human creators who satisfy these conditions. We will also explicitly address alternative framings by arguing that simulated social roles remain functionalist because they lack ontological grounding: they produce outputs that can be evaluated externally but do not participate in the actual social ontology that validates creativity. These additions will convert the central claim into a more verifiable, criterion-based argument while preserving the paper's core distinction. revision: yes

Circularity Check

0 steps flagged

No significant circularity in the paper's analytical framework

full rationale

The paper introduces functionalist and ontological perspectives on creativity as complementary macro-level views drawn from prior literature, then applies them to LLM agents by contrasting observable output characteristics against the absence of underlying processes, social embedding, and personal dimensions. This evaluation rests on independently established properties of LLMs rather than redefining the perspectives to force the conclusion, fitting parameters to data, or invoking self-citations as load-bearing justification for the distinction itself. No equations, predictions, or self-referential chains reduce the central claim to its inputs by construction; the derivation is self-contained as an interpretive analysis open to external scrutiny.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on the assumption that creativity admits a clean functionalist/ontological partition and that this partition is sufficient to evaluate current AI agents. No free parameters or invented entities are introduced.

axioms (1)
  • domain assumption Creativity can be usefully analyzed via two complementary macro-level perspectives: functionalist (observable output characteristics) and ontological (underlying processes plus social and personal dimensions).
    This partition is invoked at the start of the analysis and structures the entire argument about LLM agents.

pith-pipeline@v0.9.0 · 5455 in / 1216 out tokens · 65661 ms · 2026-05-10T13:37:09.785398+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

118 extracted references · 19 canonical work pages · 7 internal anchors

  1. [1]

    AbuMusab

    S. AbuMusab. Generative AI and human labor: who is replaceable?AI & SOCIETY, 39(6):3051–3053, 2024

    2024

  2. [2]

    Acemoglu

    D. Acemoglu. The simple macroeconomics of AI.Economic Policy, 40(121):13–58, 2025

    2025

  3. [3]

    Acemoglu and P

    D. Acemoglu and P . Restrepo. Artificial intelligence, automation, and work. InThe Economics of Artificial Intelligence: An Agenda, pages 197–236. University of Chicago Press, 2018

    2018

  4. [4]

    Acemoglu, D

    D. Acemoglu, D. Kong, and A. Ozdaglar. AI, human cognition and knowledge collapse. Technical report, National Bureau of Economic Research, 2026

    2026

  5. [5]

    M. Ahn, A. Brohan, N. Brown, Y. Chebotar, O. Cortes, B. David, C. Finn, C. Fu, K. Gopalakrishnan, K. Hausman, A. Herzog, D. Ho, J. Hsu, J. Ibarz, B. Ichter, A. Ir- pan, E. Jang, R. J. Ruano, K. Jeffrey, S. Jesmonth, N. J. Joshi, R. Julian, D. Kalashnikov, Y. Kuang, K.-H. Lee, S. Levine, Y. Lu, L. Luu, C. Parada, P . Pastor, J. Quiambao, 14 K. Rao, J. Rett...

    2022

  6. [6]

    T. M. Amabile. The social psychology of creativity: A componential conceptualiza- tion.Journal of Personality and Social Psychology, 45(2):357–376, 1983

    1983

  7. [7]

    T. M. Amabile. Motivation and creativity: Effects of motivational orientation on creative writers.Journal of Personality and Social Psychology, 48(2):393–399, 1985

    1985

  8. [8]

    Program Synthesis with Large Language Models

    J. Austin, A. Odena, M. Nye, M. Bosma, H. Michalewski, D. Dohan, E. Jiang, C. Cai, M. Terry, Q. Le, and C. Sutton. Program synthesis with large language models, 2021. arXiv:2108.07732 [cs.PL]

    work page internal anchor Pith review Pith/arXiv arXiv 2021

  9. [9]

    J. H. Barton. Non-obviousness.IDEA: The Journal of Law and Technology, 43:475–508, 2003

    2003

  10. [10]

    Bergson.Creative evolution

    H. Bergson.Creative evolution. Henry Holt and Company, 1911

    1911

  11. [11]

    T. M. Bertilsson. The elementary forms of pragmatism: On different types of abduc- tion.European Journal of Social Theory, 7(3):371–389, 2004

    2004

  12. [12]

    F. J. Binder, J. Chua, T. Korbak, H. Sleight, J. Hughes, R. Long, E. Perez, M. Turpin, and O. Evans. Looking inward: Language models can learn about themselves by introspection. InProc. of the 13th International Conference on Learning Representations (ICLR’25), 2025

    2025

  13. [13]

    Birch.The Edge of Sentience: Risk and Precaution in Humans, Other Animals, and AI

    J. Birch.The Edge of Sentience: Risk and Precaution in Humans, Other Animals, and AI. Oxford University Press, 2024

    2024

  14. [14]

    M. A. Boden. What is creativity? InDimensions of Creativity, pages 75–117. The MIT Press, 1994

    1994

  15. [15]

    M. A. Boden.The Creative Mind: Myths and Mechanisms. Routledge, 2003

    2003

  16. [16]

    M. A. Boden. Computer Models of Creativity.AI Magazine, 30(3):23–34, 2009

    2009

  17. [17]

    On the Opportunities and Risks of Foundation Models

    R. Bommasani, D. A. Hudson, E. Adeli, R. Altman, S. Arora, S. von Arx, M. S. Bernstein, J. Bohg, A. Bosselut, E. Brunskill, E. Brynjolfsson, S. Buch, D. Card, R. Castellon, N. Chatterji, A. Chen, K. Creel, J. Quincy Davis, D. Demszky, C. Don- ahue, M. Doumbouya, E. Durmus, S. Ermon, J. Etchemendy, K. Ethayarajh, L. Fei- Fei, C. Finn, T. Gale, L. Gillespie...

    work page internal anchor Pith review arXiv 2021

  18. [18]

    Borges, N

    B. Borges, N. Foroutan, D. Bayazit, A. Sotnikova, S. Montariol, T. Nazaretsky, M. Banaei, A. Sakhaeirad, P . Servant, S. P . Neshaei, J. Frej, A. Romanou, G. Weiss, S. Mamooler, Z. Chen, S. Fan, S. Gao, M. Ismayilzada, D. Paul, P . Schwaller, S. Friedli, P . Jermann, T. K ¨aser, A. Bosselut, E. G. Consortium, and E. D. Consor- tium. Could ChatGPT get an e...

    2024

  19. [19]

    T. B. Brown, B. Mann, N. Ryder, M. Subbiah, J. Kaplan, P . Dhariwal, A. Nee- lakantan, P . Shyam, G. Sastry, A. Askell, S. Agarwal, A. Herbert-Voss, G. Krueger, T. Henighan, R. Child, A. Ramesh, D. M. Ziegler, J. Wu, C. Winter, C. Hesse, M. Chen, E. Sigler, M. Litwin, S. Gray, B. Chess, J. Clark, C. Berner, S. McCandlish, A. Radford, I. Sutskever, and D. ...

    2020

  20. [20]

    and Frith, Chris and Ji, Xu and Kanai, Ryota and Klein, Colin and Lindsay, Grace and Michel, Matthias and Mudrik, Liad and Peters, Megan A

    P . Butlin, R. Long, E. Elmoznino, Y. Bengio, J. Birch, A. Constant, G. Deane, S. M. Fleming, C. Frith, X. Ji, R. Kanai, C. Klein, G. Lindsay, M. Michel, L. Mudrik, M. A. K. Peters, E. Schwitzgebel, J. Simon, and R. VanRullen. Consciousness in arti- ficial intelligence: Insights from the science of consciousness, 2023. arXiv:2308.08708 [cs.AI]

    work page arXiv 2023

  21. [21]

    M. Chen, J. Tworek, H. Jun, Q. Yuan, H. Ponde de Oliveira Pinto, J. Kaplan, H. Edwards, Y. Burda, N. Joseph, G. Brockman, A. Ray, R. Puri, G. Krueger, M. Petrov, H. Khlaaf, G. Sastry, P . Mishkin, B. Chan, S. Gray, N. Ryder, M. Pavlov, A. Power, L. Kaiser, M. Bavarian, C. Winter, P . Tillet, F. Petroski Such, D. Cummings, M. Plappert, F. Chantzis, E. Barn...

    work page internal anchor Pith review Pith/arXiv arXiv 2021

  22. [22]

    I. M. Comsa and M. Shanahan. Does it make sense to speak of introspection in large language models?, 2025. arXiv:2506.05068 [cs.CL]

    work page arXiv 2025

  23. [23]

    A. F. Cooper and J. Grimmelmann. The files are in the computer: Copyright, mem- orization, and generative AI.Chicago-Kent Law Review, 100:141, 2025

    2025

  24. [24]

    R. S. Crutchfield. Conformity and creative thinking. InContemporary approaches to creative thinking: A symposium held at the University of Colorado, pages 120–140. Atherton Press, 1962

    1962

  25. [25]

    Csikszentmihalyi

    M. Csikszentmihalyi. Society, culture, and person: A systems view of creativity. InThe Nature of Creativity: Contemporary Psychological Perspectives, pages 325–339. Cambridge University Press, 1988

    1988

  26. [26]

    E. Davis. ChatGPT’s poetry is incompetent and banal: A discussion of (Porter and Machery, 2024), 2024.https://cs.nyu.edu/ ~davise/papers/GPT-Poetry.pdf [Accessed: 2026-04-08]

    2024

  27. [27]

    E. L. Deci and R. M. Ryan.Intrinsic Motivation and Self-Determination in Human Be- havior. Springer, 1985

    1985

  28. [28]

    E. L. Deci, R. M. Ryan, P . P . Schultz, and C. P . Niemiec. Being aware and functioning fully. InHandbook of Mindfulness: Theory, Research, and Practice. The Guilford Press, 2015

    2015

  29. [29]

    Y. Du, B. Yu, T. Liu, T. Shen, J. Chen, J. G. Rittig, K. Sun, Y. Zhang, Z. Song, B. Zhou, C. Masschelein, Y. Wang, H. Wang, H. Jia, C. Zhang, H. Zhao, M. Ester, T. Head- Gordon, C. P . Gomes, H. Sun, C. Duan, P . Schwaller, and W. Jin. Accelerating sci- entific discovery with autonomous goal-evolving agents, 2025. arXiv:2512.21782 [cs.AI]

    work page arXiv 2025

  30. [30]

    E. A. Du ´e˜nez-Guzm´an, S. Sadedin, J. X. Wang, K. R. McKee, and J. Z. Leibo. A social path to human-like artificial intelligence.Nature Machine Intelligence, 5(11): 1181–1188, 2023

    2023

  31. [31]

    U. Eco. Horns, hooves, insteps: Some hypotheses on three types of abduction. In The Sign of Three: Dupin, Holmes, Peirce, pages 198–220. Indiana University Press, 1983

    1983

  32. [32]

    Einstein and L

    A. Einstein and L. Infeld.The Evolution of Physics: The Growth of Ideas from Early Concepts to Relativity and Quanta. Cambridge University Press, 1938

    1938

  33. [33]

    A. Fan, M. Lewis, and Y. Dauphin. Hierarchical neural story generation. InProc. of the 56th Annual Meeting of the Association for Computational Linguistics (ACL’18), 2018. 17

    2018

  34. [34]

    T. Feng, T. H. Trinh, G. Bingham, D. Hwang, Y. Chervonyi, J. Jung, J. Lee, C. Pagano, S. hyun Kim, F. Pasqualotto, S. Gukov, J. N. Lee, J. Kim, K. Hou, G. Ghiasi, Y. Tay, Y. Li, C. Kuang, Y. Liu, H. Lin, E. Z. Liu, N. Nayakanti, X. Yang, H.-T. Cheng, D. Has- sabis, K. Kavukcuoglu, Q. V . Le, and T. Luong. Towards autonomous mathematics research, 2026. arX...

    work page arXiv 2026

  35. [35]

    Franceschelli and M

    G. Franceschelli and M. Musolesi. Reinforcement learning for generative AI: State of the art, opportunities and open research challenges.Journal of Artificial Intelligence Research, 79:417–446, 2024

    2024

  36. [36]

    Franceschelli and M

    G. Franceschelli and M. Musolesi. On the creativity of large language models.AI & Society, 40:3785–3795, 2025

    2025

  37. [37]

    Y. Gao, Y. Xiong, X. Gao, K. Jia, J. Pan, Y. Bi, Y. Dai, J. Sun, M. Wang, and H. Wang. Retrieval-augmented generation for large language models: A survey,

  38. [38]

    arXiv:2312.10997 [cs.CL]

    work page internal anchor Pith review Pith/arXiv arXiv

  39. [39]

    B. Gaut. The philosophy of creativity.Philosophy Compass, 5(12):1034–1046, 2010

    2010

  40. [40]

    Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities

    Gemini Team and Google. Gemini 2.5: Pushing the frontier with advanced rea- soning, multimodality, long context, and next generation agentic capabilities, 2025. arXiv:2507.06261 [cs.CL]

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  41. [41]

    J. W. Getzels. Problem-finding and the inventiveness of solutions.The Journal of Creative Behavior, 9(1):12–18, 1975

    1975

  42. [42]

    Gottlieb, P .-Y

    J. Gottlieb, P .-Y. Oudeyer, M. Lopes, and A. Baranes. Information-seeking, curiosity, and attention: Computational and neural mechanisms.Trends in Cognitive Sciences, 17(11):585–593, 2013

    2013

  43. [43]

    Single-minus gluon tree amplitudes are nonzero,

    A. Guevara, A. Lupsasca, D. Skinner, A. Strominger, and K. Weil. Single-minus gluon tree amplitudes are nonzero, 2026. arXiv:2602.12176 [hep-th]

    work page arXiv 2026

  44. [44]

    E. E. Guzik, C. Byrge, and C. Gilde. The originality of machines: AI takes the Tor- rance Test.Journal of Creativity, 33(3):100065, 2023

    2023

  45. [45]

    Hadsell, D

    R. Hadsell, D. Rao, A. A. Rusu, and R. Pascanu. Embracing change: Continual learning in deep neural networks.Trends in Cognitive Sciences, 24(12):1028–1040, 2020

    2020

  46. [46]

    E. Hoel. Bits in, bits out, 2026.https://www.theintrinsicperspective.com/p/ bits-in-bits-out[Accessed: 2026-04-08]

    2026

  47. [47]

    E. Hoel. A disproof of large language model consciousness: The necessity of con- tinual learning for consciousness, 2026. arXiv:2512.12802 [q-bio.NC]. 18

    work page arXiv 2026

  48. [48]

    Holtzman, J

    A. Holtzman, J. Buys, L. Du, M. Forbes, and Y. Choi. The curious case of neural text degeneration. InProc. of the 8th International Conference on Learning Representations (ICLR’20), 2020

    2020

  49. [49]

    Y. Hu, Q. Zhou, Q. Chen, X. Li, L. Liu, D. Zhang, A. Kachroo, T. Oz, and O. Tripp. QualityFlow: An agentic workflow for program synthesis controlled by LLM qual- ity checks, 2025. arXiv:2501.17167 [cs.SE]

    work page arXiv 2025

  50. [50]

    Y. Hu, S. Liu, Y. Yue, G. Zhang, B. Liu, F. Zhu, J. Lin, H. Guo, S. Dou, Z. Xi, S. Jin, J. Tan, Y. Yin, J. Liu, Z. Zhang, Z. Sun, Y. Zhu, H. Sun, B. Peng, Z. Cheng, X. Fan, J. Guo, X. Yu, Z. Zhou, Z. Hu, J. Huo, J. Wang, Y. Niu, Y. Wang, Z. Yin, X. Hu, Y. Liao, Q. Li, K. Wang, W. Zhou, Y. Liu, D. Cheng, Q. Zhang, T. Gui, S. Pan, Y. Zhang, P . Torr, Z. Dou...

  51. [51]

    arXiv:2512.13564 [cs.CL]

    work page internal anchor Pith review arXiv

  52. [52]

    Ismayilzada, C

    M. Ismayilzada, C. Stevenson, and L. van der Plas. Evaluating creative short story generation in humans and large language models. InProc. of the 16th International Conference on Computational Creativity (ICCC’25), 2025

    2025

  53. [53]

    A. Issak. Artistic autonomy in AI art. InProc. of NIPS’21 Machine Learning for Cre- ativity and Design Workshop, 2021

    2021

  54. [54]

    Kimi K2: Open Agentic Intelligence

    Kimi Team. Kimi K2: Open agentic intelligence, 2026. arXiv:2507.20534 [cs.LG]

    work page internal anchor Pith review arXiv 2026

  55. [55]

    D. Knuth. Claude’s cycles, 2026.https://cs.stanford.edu/ ~knuth/papers/ claude-cycles.pdf[Accessed: 2026-04-08]

    2026

  56. [56]

    Koivisto and S

    M. Koivisto and S. Grassini. Best humans still outperform artificial intelligence in a creative divergent thinking task.Scientific Reports, 13(1):13601, 2023

    2023

  57. [57]

    Kojima, S

    T. Kojima, S. S. Gu, M. Reid, Y. Matsuo, and Y. Iwasawa. Large language models are zero-shot reasoners. InProc. of the 36th International Conference on Neural Information Processing Systems (NIPS’22), 2022

    2022

  58. [58]

    T. S. Kuhn.The Structure of Scientific Revolutions. University of Chicago Press, 1962

    1962

  59. [59]

    Ladosz, L

    P . Ladosz, L. Weng, M. Kim, and H. Oh. Exploration in deep reinforcement learning: A survey.Information Fusion, 85:1–22, 2022

    2022

  60. [60]

    S. Lavelle. The machine with a human face: From artificial intelligence to artificial sentience. InAdvanced Information Systems Engineering Workshops (CAiSE’20), 2020

    2020

  61. [61]

    Lewis, E

    P . Lewis, E. Perez, A. Piktus, F. Petroni, V . Karpukhin, N. Goyal, H. K ¨uttler, M. Lewis, W.-t. Yih, T. Rockt ¨aschel, S. Riedel, and D. Kiela. Retrieval-augmented generation for knowledge-intensive NLP tasks. InAdvances in Neural Information Processing Systems (NIPS’20), 2020. 19

    2020

  62. [62]

    G. Li, H. A. Al Kader Hammoud, H. Itani, D. Khizbullin, and B. Ghanem. CAMEL: communicative agents for ”mind” exploration of large language model society. InProc. of the 37th International Conference on Neural Information Processing Systems (NIPS’23), 2023

    2023

  63. [63]

    Lin, K.-C

    Y.-C. Lin, K.-C. Chen, Z.-Y. Li, T.-H. Wu, T.-H. Wu, K.-Y. Chen, H.-y. Lee, and Y.-N. Chen. Creativity in LLM-based multi-agent systems: A survey. InProc. of the 2025 Conference on Empirical Methods in Natural Language Processing (EMNLP’25), 2025

    2025

  64. [64]

    L.-C. Lu, M. Liu, P . C. Lu, Y. Tian, S.-H. Sun, and N. Peng. Rethinking creativ- ity evaluation: A critical analysis of existing creativity evaluations. InProc. of the 19th Conference of the European Chapter of the Association for Computational Linguistics (EACL’26), 2026

    2026

  65. [65]

    C. G. Lucas, D. Sterling, and C. Kemp. Superspace extrapolation reveals inductive biases in function learning. InProc. of the 34th Annual Meeting of the Cognitive Science Society (CogSci’12), 2012

    2012

  66. [66]

    Lupsasca

    A. Lupsasca. GPT-5.2 derives a new result in theoretical physics, 2026.https:// openai.com/index/new-result-theoretical-physics/[Accessed: 2026-04-08]

    2026

  67. [67]

    D. W. MacKinnon. Creativity: A multi-faceted phenomenon. InCreativity: A discus- sion at the Nobel conference, page 17–32. North-Holland Publishing Company, 1970

    1970

  68. [68]

    Marco, L

    G. Marco, L. Rello, and J. Gonzalo. Small language models can outperform humans in short creative writing: A study comparing SLMs with humans and LLMs. InProc. of the 31st International Conference on Computational Linguistics (COLING’25), 2025

    2025

  69. [69]

    N. N. Minh, A. Baker, C. Neo, A. G. Roush, A. Kirsch, and R. Shwartz-Ziv. Turning up the heat: Min-p sampling for creative and coherent LLM outputs. InProc. of the 13th International Conference on Learning Representations (ICLR’25), 2025

    2025

  70. [70]

    Nguyen, V

    D. Nguyen, V . D. Lai, S. Yoon, R. A. Rossi, H. Zhao, R. Zhang, P . Mathur, N. Lipka, Y. Wang, T. Bui, F. Dernoncourt, and T. Zhou. Dynasaur: Large language agents beyond predefined actions. InProc. of the 2nd Conference on Language Modeling (COLM’25), 2025

    2025

  71. [71]

    Nisioti, S

    E. Nisioti, S. Risi, I. Momennejad, P .-Y. Oudeyer, and C. Moulin-Frier. Collective innovation in groups of large language models. InProc. of the 2024 Conference on Artificial Life (ALIFE’24), 2024

    2024

  72. [72]

    Alethea Power, Yuri Burda, Harri Edwards, Igor Babuschkin, and Vedant Misra

    R. Nogueira, Z. Jiang, and J. Lin. Investigating the limitations of transformers with simple arithmetic tasks, 2021. arXiv:2102.13019 [cs.CL]. 20

    work page arXiv 2021

  73. [73]

    OpenAI o3 and o4-mini system card, 2025.https://cdn.openai

    OpenAI. OpenAI o3 and o4-mini system card, 2025.https://cdn.openai. com/pdf/2221c875-02dc-4789-800b-e7758f3722c1/o3-and-o4-mini-system- card.pdf[Accessed: 2026-04-08]

    2025

  74. [74]

    T. Ord. Interpolation, extrapolation, hyperpolation: Generalising into new dimen- sions, 2024. arXiv:2409.05513 [cs.LG]

    work page arXiv 2024

  75. [75]

    Ouyang, J

    L. Ouyang, J. Wu, X. Jiang, D. Almeida, C. Wainwright, P . Mishkin, C. Zhang, S. Agarwal, K. Slama, A. Ray, J. Schulman, J. Hilton, F. Kelton, L. Miller, M. Simens, A. Askell, P . Welinder, P . F. Christiano, J. Leike, and R. Lowe. Training language models to follow instructions with human feedback. InAdvances in Neural Informa- tion Processing Systems (N...

    2022

  76. [76]

    J. S. Park, J. O’Brien, C. J. Cai, M. R. Morris, P . Liang, and M. S. Bernstein. Generative agents: Interactive simulacra of human behavior. InProc. of the 36th Annual ACM Symposium on User Interface Software and Technology (UIST’23), 2023

    2023

  77. [77]

    E. S. Paul and D. Stokes. Attributing creativity. InCreativity and Philosophy, pages 193–209. Routledge, 2018

    2018

  78. [78]

    Peeperkorn, T

    M. Peeperkorn, T. Kouwenhoven, D. Brown, and A. Jordanous. Is temperature the creativity parameter of large language models? InProc. of the 15th International Conference on Computational Creativity (ICCC’24), 2024

    2024

  79. [79]

    Toy Story, 1995

    Pixar Animation Studios. Toy Story, 1995

    1995

  80. [80]

    Plaat, M

    A. Plaat, M. van Duijn, N. van Stein, M. Preuss, P . van der Putten, and K. Joost Batenburg. Agentic large language models, a survey.Journal of Artificial Intelligence Research, 84:29:1–74, 2025

    2025

Showing first 80 references.