pith. sign in

arxiv: 2606.03030 · v1 · pith:BEC7H2FPnew · submitted 2026-06-02 · 💻 cs.GT · econ.GN· q-fin.EC

Do Matching Mechanisms Work with LLM Agents?

Yukihiro Hoshino , Ayato Kitadai , Nariaki Nishino This is my paper

Pith reviewed 2026-06-28 08:28 UTC · model grok-4.3

classification 💻 cs.GT econ.GNq-fin.EC
keywords LLM agentsmatching mechanismsmarket designstabilityefficiencytruth-tellingstrategy-proofnessdeferred acceptance
0
0 comments X

The pith

Matching mechanisms produce more stable and efficient outcomes than free negotiation when LLM agents make allocation decisions.

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

The paper tests whether standard matching mechanisms continue to deliver their predicted benefits when LLM agents act as delegated decision makers instead of humans. In controlled one-to-one matching experiments, centralized mechanisms outperform decentralized free negotiation on both stability and efficiency. LLM agents also report preferences truthfully at higher rates than human subjects in comparable Deferred Acceptance and EADA settings. Truth-telling does not track strategy-proofness uniformly, since TTC fails to elicit the highest rates despite being strategy-proof. The findings indicate that matching theory supplies a useful but incomplete template for designing institutions that LLM agents will populate.

Core claim

Across controlled one-to-one matching environments, mechanism-based markets generally outperform free negotiation in terms of stability and efficiency. LLM agents report preferences truthfully at substantially higher rates than human subjects in comparable DA and EADA environments, yet truth-telling is not uniformly aligned with formal strategy-proofness across all mechanisms.

What carries the argument

Direct experimental comparison of decentralized free-negotiation markets against centralized mechanisms (DA, EADA, TTC) when LLM agents report preferences and execute allocations in one-to-one assignment problems.

If this is right

  • Centralized mechanisms deliver higher stability and efficiency than free negotiation in LLM-agent markets.
  • LLM agents report preferences truthfully more often than human subjects in DA and EADA environments.
  • Strategy-proofness does not reliably predict the highest observed truth-telling rates across mechanisms.
  • Matching theory offers a partial but not complete guide for institution design in LLM-agent settings.

Where Pith is reading between the lines

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

  • Designers of future AI-mediated markets may need to run empirical tests rather than rely solely on theoretical strategy-proofness.
  • Observed truth-telling gaps could widen or narrow as LLM capabilities or prompting methods change.
  • The advantage of mechanisms may or may not hold in multi-sided or repeated-interaction markets not examined here.

Load-bearing premise

The specific LLM agents and controlled one-to-one environments used in the experiments represent how LLM agents would behave in real-world delegated matching markets.

What would settle it

Re-running the same protocol with different LLM models or in larger real-world matching instances and finding that free negotiation matches or exceeds mechanism performance on stability and efficiency would falsify the central result.

Figures

Figures reproduced from arXiv: 2606.03030 by Ayato Kitadai, Nariaki Nishino, Yukihiro Hoshino.

Figure 1
Figure 1. Figure 1: Passive free negotiation vs. Active free negotiation [PITH_FULL_IMAGE:figures/full_fig_p015_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Protocol of passive free negotiation markets [PITH_FULL_IMAGE:figures/full_fig_p015_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Stability among markets environments 1 2 3 4 5 Preference job school nursery Scenario DA*** DA*** DA*** DA*** DA*** DA*** DA*** DA*** DA*** DA*** DA*** DA*** DA*** DA*** DA*** DA vs free_passive: Stability (gemini) 1 2 3 4 5 Preference job school nursery Scenario DA*** DA*** DA*** DA*** DA*** DA*** DA*** DA*** DA*** DA*** DA*** DA*** DA*** DA*** DA*** DA vs free_passive: Stability (gpt) Legend free_passive… view at source ↗
Figure 4
Figure 4. Figure 4: DA vs. free negotiation markets (Stability) [PITH_FULL_IMAGE:figures/full_fig_p018_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Efficiency among matching environments 6.2 Mechanism-Based Markets By comparing the stability of matching results among different mechanism-based environments, we verify whether the mechanisms function consistently with theory. Referring to [PITH_FULL_IMAGE:figures/full_fig_p019_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: RSD vs. free negotiation markets (Efficiency) [PITH_FULL_IMAGE:figures/full_fig_p020_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: TTC vs. free negotiation markets (Efficiency) [PITH_FULL_IMAGE:figures/full_fig_p020_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: EADA vs. free negotiation markets (Efficiency) [PITH_FULL_IMAGE:figures/full_fig_p020_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: DA vs. other mechanisms (Stability) [PITH_FULL_IMAGE:figures/full_fig_p021_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: EADA vs. other mechanisms 1 2 3 4 5 Preference job school nursery Scenario ns ns TTC*** ns TTC*** ns DA** TTC*** DA** TTC*** ns ns TTC*** ns TTC*** TTC vs DA: Efficiency (gemini) 1 2 3 4 5 Preference job school nursery Scenario ns DA*** TTC*** DA*** TTC*** ns DA*** TTC*** ns ns ns ns TTC*** ns TTC*** TTC vs DA: Efficiency (gpt) Legend DA p<0.01 DA p<0.05 DA p<0.1 ns TTC p<0.1 TTC p<0.05 TTC p<0.01 1 2 3 4… view at source ↗
Figure 11
Figure 11. Figure 11: TTC vs. other mechanisms For Gemini, the truth-telling rate was basically higher in EADA, whereas for GPT, the truth-telling rate was basically higher in TTC. 6.3 Market Scenarios In this section, we verify whether differences in market scenario settings cause differences in the stability and efficiency of matching results, as well as the truth-telling rates for the mechanisms [PITH_FULL_IMAGE:figures/fu… view at source ↗
Figure 12
Figure 12. Figure 12: Truth-telling rate among matching environments [PITH_FULL_IMAGE:figures/full_fig_p023_12.png] view at source ↗
Figure 13
Figure 13. Figure 13: DA vs. EADA vs. TTC 23 [PITH_FULL_IMAGE:figures/full_fig_p023_13.png] view at source ↗
Figure 14
Figure 14. Figure 14: Market scenarios (Stability, Efficiency, Truth-telling rate) [PITH_FULL_IMAGE:figures/full_fig_p025_14.png] view at source ↗
Figure 15
Figure 15. Figure 15: Outcomes of Strategic behaviors encountered in human market design, such as participants’ bounded rationality or misunderstandings of the mechanism, will be significantly reduced in LLM agent markets. Furthermore, in this experiment, mentions of specific mechanism names and dominant strategies were observed in the agents’ thought processes even without explicit instructions in the prompts. This means that… view at source ↗
read the original abstract

This study examines whether standard matching mechanisms function as intended in LLM-agent markets, where LLM agents make allocation-related decisions as delegated decision-makers. We compare decentralized free-negotiation markets with centralized mechanism-based markets including several representative mechanisms. Across controlled one-to-one matching environments, mechanism-based markets generally outperform free negotiation in terms of stability and efficiency. We also find that LLM agents report preferences truthfully at substantially higher rates than human subjects in comparable DA and EADA environments. However, truth-telling is not uniformly aligned with formal strategy-proofness across all mechanisms: TTC, despite being strategy-proof, does not always elicit higher truth-telling than EADA. These results suggest that matching theory provides a useful but incomplete guide for designing institutions in LLM-agent markets.

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

2 major / 1 minor

Summary. This paper examines the behavior of LLM agents in one-to-one matching markets. It compares decentralized free-negotiation markets against centralized mechanisms (including DA, EADA, and TTC) in controlled environments, claiming that mechanism-based markets generally outperform free negotiation on stability and efficiency. It further claims that LLM agents report preferences truthfully at substantially higher rates than human subjects in comparable DA and EADA settings, though truth-telling rates are not uniformly higher for strategy-proof mechanisms such as TTC relative to EADA. The abstract concludes that matching theory provides a useful but incomplete guide for LLM-agent markets.

Significance. If the empirical results prove robust after addressing methodological gaps, the work would provide initial evidence on how standard matching mechanisms perform when decision-making is delegated to LLM agents. This could inform institutional design for AI-mediated markets. The paper receives credit for conducting controlled comparisons between free negotiation and multiple mechanisms and for benchmarking LLM truth-telling rates against existing human-subject data.

major comments (2)
  1. [Abstract] Abstract and experimental methods: the abstract states general performance differences and higher truth-telling rates but supplies no information on number of trials, statistical tests, specific models used, temperature settings, or controls for prompt variation. Without these details the degree to which the data support the central claims cannot be assessed.
  2. [Abstract] The central claims require that behavior observed with the authors' chosen LLM agents and one-to-one environments generalizes to other frontier models, larger instances, and preference distributions drawn from actual delegated markets. No evidence or robustness checks addressing sensitivity to model version, prompt phrasing, or market complexity are described.
minor comments (1)
  1. Clarify the exact set of mechanisms tested and the precise definitions of stability and efficiency used in the LLM-agent setting.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the detailed and constructive comments. We address each major comment below and outline the revisions we will make.

read point-by-point responses

  1. Referee: [Abstract] Abstract and experimental methods: the abstract states general performance differences and higher truth-telling rates but supplies no information on number of trials, statistical tests, specific models used, temperature settings, or controls for prompt variation. Without these details the degree to which the data support the central claims cannot be assessed.

    Authors: We agree that the abstract would benefit from greater methodological transparency. In the revised manuscript we will expand the abstract to report the number of trials per condition, the statistical tests used for comparisons, the specific LLM models and versions, the temperature settings, and the controls employed for prompt variation (including use of standardized templates). Corresponding details will also be added or clarified in the methods section. revision: yes

  2. Referee: [Abstract] The central claims require that behavior observed with the authors' chosen LLM agents and one-to-one environments generalizes to other frontier models, larger instances, and preference distributions drawn from actual delegated markets. No evidence or robustness checks addressing sensitivity to model version, prompt phrasing, or market complexity are described.

    Authors: Our study is explicitly framed as an initial investigation in controlled one-to-one settings; the manuscript does not assert broad generalization. We will revise the abstract and add a limitations section that clearly delineates the scope of the findings and notes the absence of robustness checks across model versions, prompt phrasings, and larger or real-world preference distributions. Additional experiments on these dimensions are beyond the scope of the present revision. revision: partial

Circularity Check

0 steps flagged

No circularity: empirical study with direct experimental comparisons

full rationale

The paper reports controlled experiments comparing free negotiation against mechanism-based markets (DA, EADA, TTC) using LLM agents in one-to-one matching settings. Central claims concern observed stability, efficiency, and truth-telling rates; these rest on direct measurement rather than any derivation, prediction, or first-principles result that reduces to fitted inputs or self-referential definitions. No equations, ansatzes, or uniqueness theorems are invoked whose validity depends on the present results. Self-citations to matching theory are external and non-load-bearing for the empirical findings. The work is self-contained against its stated benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The paper is an empirical test of existing mechanisms and introduces no new free parameters, axioms beyond standard matching-theory assumptions, or invented entities.

axioms (1)
  • domain assumption One-to-one matching environments with agents reporting preferences.
    The study is conducted in controlled one-to-one matching environments as stated in the abstract.

pith-pipeline@v0.9.1-grok · 5658 in / 1171 out tokens · 17451 ms · 2026-06-28T08:28:45.129903+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

211 extracted references · 87 canonical work pages

  1. [1]

    Random serial dictatorship and the core from random endowments in house allocation problems.Econometrica, 66(3):689–701, 1998

    Atila Abdulkadiro˘glu and Tayfun Sönmez. Random serial dictatorship and the core from random endowments in house allocation problems.Econometrica, 66(3):689–701, 1998

    1998

  2. [2]

    School choice: A mechanism design approach.American Economic Review, 93(3):729–747, 2003

    Atila Abdulkadiro˘glu and Tayfun Sönmez. School choice: A mechanism design approach.American Economic Review, 93(3):729–747, 2003. doi:10.1257/000282803322157061

    work page doi:10.1257/000282803322157061 2003

  3. [3]

    Pathak, and Alvin E

    Atila Abdulkadiro˘glu, Parag A. Pathak, and Alvin E. Roth. The new york city high school match.American Economic Review, 95(2):364–367, 2005. doi:10.1257/000282805774670167

    work page doi:10.1257/000282805774670167 2005

  4. [4]

    Roth, and Tayfun Sonmez

    Atila Abdulkadiro˘glu, Parag Pathak, Alvin E. Roth, and Tayfun Sonmez. Changing the boston school choice mechanism. NBER Working Papers 11965, National Bureau of Economic Research, Inc, 2006

    2006

  5. [5]

    Pathak, and Alvin E

    Atila Abdulkadiro˘glu, Parag A. Pathak, and Alvin E. Roth. Strategy-proofness versus efficiency in matching with indifferences: Redesigning the NYC high school match.American Economic Review, 99(5):1954–1978,

    1954

  6. [6]

    doi:10.1257/aer.99.5.1954

    work page doi:10.1257/aer.99.5.1954 1954

  7. [7]

    boston mechanism

    Atila Abdulkadiro ˘glu, Yeon-Koo Che, and Yosuke Yasuda. Resolving conflicting preferences in school choice: The “boston mechanism” reconsidered.American Economic Review, 101(1):399–410, 2011. doi:10.1257/aer.101.1.399

    work page doi:10.1257/aer.101.1.399 2011

  8. [8]

    Pathak, Alvin E

    Atila Abdulkadiro˘glu, Yeon-Koo Che, Parag A. Pathak, Alvin E. Roth, and Olivier Tercieux. Efficiency, justified envy, and incentives in priority-based matching.American Economic Review: Insights, 2(4):425–442, 2020. doi:10.1257/aeri.20190307

    work page doi:10.1257/aeri.20190307 2020

  9. [9]

    Arriaga, and Adam Tauman Kalai

    Gati Aher, Rosa I. Arriaga, and Adam Tauman Kalai. Using large language models to simulate multiple humans and replicate human subject studies. InProceedings of the 40th International Conference on Machine Learning, ICML’23. JMLR.org, 2023

    2023

  10. [10]

    Playing repeated games with large language models.Nature Human Behaviour, 9(7):1380–1390, 2025

    Elif Akata, Lion Schulz, Julian Coda-Forno, Seong Joon Oh, Matthias Bethge, and Eric Schulz. Playing repeated games with large language models.Nature Human Behaviour, 9(7):1380–1390, 2025. doi:10.1038/s41562-025- 02172-y

    work page doi:10.1038/s41562-025- 2025

  11. [11]

    Figueroa, Yash Kanoria, and Akshit Kumar

    Amine Allouah, Omar Besbes, Josué D. Figueroa, Yash Kanoria, and Akshit Kumar. What is your AI agent buying? evaluation, biases, model dependence, & emerging implications of agentic e-commerce. InProceedings of the ACM Web Conference 2026, pages 8697–8700, 2026. doi:10.1145/3774904.3792943

    work page doi:10.1145/3774904.3792943 2026

  12. [12]

    Modeling human decisions in coupled human and natural systems: Review of agent-based models

    Li An. Modeling human decisions in coupled human and natural systems: Review of agent-based models. Ecological Modelling, 229:25–36, 2012. doi:10.1016/j.ecolmodel.2011.07.010

    work page doi:10.1016/j.ecolmodel.2011.07.010 2012

  13. [13]

    Language models as agent models

    Jacob Andreas. Language models as agent models. In Yoav Goldberg, Zornitsa Kozareva, and Yue Zhang, editors,Findings of the Association for Computational Linguistics: EMNLP 2022, pages 5769–5779, 2022. doi:10.18653/v1/2022.findings-emnlp.423

    work page doi:10.18653/v1/2022.findings-emnlp.423 2022

  14. [14]

    Argyle, Ethan C

    Lisa P. Argyle, Ethan C. Busby, Nancy Fulda, Joshua R. Gubler, Christopher Rytting, and David Wingate. Out of one, many: Using language models to simulate human samples.Political Analysis, 31(3):337–351, 2023. doi:10.1017/pan.2023.2

    work page doi:10.1017/pan.2023.2 2023

  15. [15]

    Arrow and Gerard Debreu

    Kenneth J. Arrow and Gerard Debreu. Existence of an equilibrium for a competitive economy.Econometrica, 22 (3):265–290, 1954

    1954

  16. [16]

    Stable matching mechanisms are not obviously strategy-proof.Journal of Economic Theory, 177:405–425, 2018

    Itai Ashlagi and Yannai A Gonczarowski. Stable matching mechanisms are not obviously strategy-proof.Journal of Economic Theory, 177:405–425, 2018

    2018

  17. [17]

    The dissemination of culture: A model with local convergence and global polarization.The Journal of Conflict Resolution, 41(2):203–226, 1997

    Robert Axelrod. The dissemination of culture: A model with local convergence and global polarization.The Journal of Conflict Resolution, 41(2):203–226, 1997. 30 Do Matching Mechanisms Work with LLM Agents?A PREPRINT

    1997

  18. [18]

    A guide for newcomers to agent-based modeling in the social sciences

    Robert Axelrod and Leigh Tesfatsion. A guide for newcomers to agent-based modeling in the social sciences. Staff General Research Papers Archive 12515, Iowa State University, Department of Economics, 2006

    2006

  19. [19]

    How well can LLMs negotiate? negotiationarena platform and analysis

    Federico Bianchi, Patrick John Chia, Mert Yuksekgonul, Jacopo Tagliabue, Dan Jurafsky, and James Zou. How well can LLMs negotiate? negotiationarena platform and analysis. InForty-first International Conference on Machine Learning, 2024

    2024

  20. [20]

    Synthetic replacements for hu- man survey data? the perils of large language models.Political Analysis, 32:1–16, 2024

    James Bisbee, Joshua Clinton, Cassy Dorff, Brenton Kenkel, and Jennifer Larson. Synthetic replacements for hu- man survey data? the perils of large language models.Political Analysis, 32:1–16, 2024. doi:10.1017/pan.2024.5

    work page doi:10.1017/pan.2024.5 2024

  21. [21]

    Iterative versus standard deferred acceptance: Experimental evidence.The Economic Journal, 130(626):356–392, 2020

    Inácio Bó and Rustamdjan Hakimov. Iterative versus standard deferred acceptance: Experimental evidence.The Economic Journal, 130(626):356–392, 2020

    2020

  22. [22]

    A new solution to the random assignment problem.Journal of Economic Theory, 100(2):295–328, 2001

    Anna Bogomolnaia and Hervé Moulin. A new solution to the random assignment problem.Journal of Economic Theory, 100(2):295–328, 2001. doi:10.1006/jeth.2000.2710

    work page doi:10.1006/jeth.2000.2710 2001

  23. [23]

    Agent-based modeling: Methods and techniques for simulating human systems.Proceedings of the National Academy of Sciences of the United States of America, 99(10):7280–7287, 2002

    Eric Bonabeau. Agent-based modeling: Methods and techniques for simulating human systems.Proceedings of the National Academy of Sciences of the United States of America, 99(10):7280–7287, 2002

    2002

  24. [24]

    Citysim: Modeling urban behaviors and city dynamics with large-scale llm-driven agent simulation

    Nicolas Bougie and Narimawa Watanabe. Citysim: Modeling urban behaviors and city dynamics with large-scale llm-driven agent simulation. InProceedings of the 2025 Conference on Empirical Methods in Natural Language Processing: Industry Track, pages 215–229, 2025

    2025

  25. [25]

    Constrained school choice: An experimental study

    Caterina Calsamiglia, Guillaume Haeringer, and Flip Klijn. Constrained school choice: An experimental study. American Economic Review, 100(4):1860–1874, 2010. doi:10.1257/aer.100.4.1860

    work page doi:10.1257/aer.100.4.1860 2010

  26. [26]

    Rise of the machines: Delegating decisions to autonomous AI.Computers in Human Behavior, 134:107308, 2022

    Cindy Candrian and Anne Scherer. Rise of the machines: Delegating decisions to autonomous AI.Computers in Human Behavior, 134:107308, 2022. doi:10.1016/j.chb.2022.107308

    work page doi:10.1016/j.chb.2022.107308 2022

  27. [27]

    Specializing large language models to simulate survey response distributions for global populations

    Yong Cao, Haijiang Liu, Arnav Arora, Isabelle Augenstein, Paul Röttger, and Daniel Hershcovich. Specializing large language models to simulate survey response distributions for global populations. InProceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volum...

    work page doi:10.18653/v1/2025.naacl- 2025

  28. [28]

    Guarantees for autonomy in cognitive agent architecture

    Cristiano Castelfranchi. Guarantees for autonomy in cognitive agent architecture. InProceedings of the Workshop on Agent Theories, Architectures, and Languages on Intelligent Agents, pages 56–70, 1995

    1995

  29. [29]

    School choice with consent: an experiment.The Economic Journal, 134(661):1760–1805, 2024

    Claudia Cerrone, Yoan Hermstrüwer, and Onur Kesten. School choice with consent: an experiment.The Economic Journal, 134(661):1760–1805, 2024. doi:10.1093/ej/uead120

    work page doi:10.1093/ej/uead120 2024

  30. [30]

    Put your money where your mouth is: Evaluating strategic planning and execution of LLM agents in an auction arena

    Jiangjie Chen, Siyu Yuan, Rong Ye, Bodhisattwa Prasad Majumder, and Kyle Richardson. Put your money where your mouth is: Evaluating strategic planning and execution of LLM agents in an auction arena. InNeurIPS 2024 Workshop on Open-World Agents, 2024

    2024

  31. [31]

    Chinese college admissions and school choice reforms: An experimental study

    Yan Chen and Onur Kesten. Chinese college admissions and school choice reforms: An experimental study. Games and Economic Behavior, 115:83–100, 2019. doi:10.1016/j.geb.2019.02.003

    work page doi:10.1016/j.geb.2019.02.003 2019

  32. [32]

    School choice: an experimental study.Journal of Economic Theory, 127(1): 202–231, 2006

    Yan Chen and Tayfun Sönmez. School choice: an experimental study.Journal of Economic Theory, 127(1): 202–231, 2006. doi:10.1016/j.jet.2004.10.006

    work page doi:10.1016/j.jet.2004.10.006 2006

  33. [33]

    Kirshner, Anton Ovchinnikov, Meena Andiappan, and Tracy Jenkin

    Yang Chen, Samuel N. Kirshner, Anton Ovchinnikov, Meena Andiappan, and Tracy Jenkin. A manager and an AI walk into a bar: Does chatGPT make biased decisions like we do?Manufacturing & Service Operations Management, 27(2):354–368, 2025. doi:10.1287/msom.2023.0279

    work page doi:10.1287/msom.2023.0279 2025

  34. [34]

    The emergence of economic rationality of GPT

    Yiting Chen, Tracy Xiao Liu, You Shan, and Songfa Zhong. The emergence of economic rationality of GPT. Proceedings of the National Academy of Sciences, 120(51):e2316205120, 2023. doi:10.1073/pnas.2316205120

    work page doi:10.1073/pnas.2316205120 2023

  35. [35]

    Strategic self-improvement for competitive agents in AI labour markets.arXiv preprint arXiv:2512.04988, 2025

    Christopher Chiu, Simpson Zhang, and Mihaela van der Schaar. Strategic self-improvement for competitive agents in AI labour markets.arXiv preprint arXiv:2512.04988, 2025

    arXiv 2025

  36. [36]

    Edward H. Clarke. Multipart pricing of public goods.Public Choice, 11:17–33, 1971

    1971

  37. [37]

    Coleman.Microfoundations and Macrosocial Behavior, pages 153–176

    James S. Coleman.Microfoundations and Macrosocial Behavior, pages 153–176. University of California Press, 1987

    1987

  38. [38]

    Statistical comparisons of classifiers over multiple data sets.J

    Janez Demšar. Statistical comparisons of classifiers over multiple data sets.J. Mach. Learn. Res., 7:1–30, 2006

    2006

  39. [39]

    LLMs at the bargaining table

    Yuan Deng, Vahab Mirrokni, Renato Paes Leme, Hanrui Zhang, and Song Zuo. LLMs at the bargaining table. In Agentic Markets Workshop at ICML 2024, 2024

    2024

  40. [40]

    Can AI language models replace human participants? Trends in Cognitive Sciences, 27(7):597–600, 2023

    Danica Dillion, Niket Tandon, Yuling Gu, and Kurt Gray. Can AI language models replace human participants? Trends in Cognitive Sciences, 27(7):597–600, 2023. doi:10.1016/j.tics.2023.04.008. 31 Do Matching Mechanisms Work with LLM Agents?A PREPRINT

    work page doi:10.1016/j.tics.2023.04.008 2023

  41. [41]

    Questioning the survey responses of large language models

    Ricardo Dominguez-Olmedo, Moritz Hardt, and Celestine Mendler-Dünner. Questioning the survey responses of large language models. InProceedings of the 38th International Conference on Neural Information Processing Systems, 2024

    2024

  42. [42]

    L. E. Dubins and D. A. Freedman. Machiavelli and the gale-shapley algorithm.The American Mathematical Monthly, 88(7):485–494, 1981. doi:10.1080/00029890.1981.11995301

    work page doi:10.1080/00029890.1981.11995301 1981

  43. [43]

    Adverse reactions to the use of large language models in social interactions.PNAS Nexus, 4(4):pgaf112, 2025

    Fabian Dvorak, Regina Stumpf, Sebastian Fehrler, and Urs Fischbacher. Adverse reactions to the use of large language models in social interactions.PNAS Nexus, 4(4):pgaf112, 2025. doi:10.1093/pnasnexus/pgaf112

    work page doi:10.1093/pnasnexus/pgaf112 2025

  44. [44]

    An experimental study of decentralized matching.Quantitative Economics, 16(2):497–533, 2025

    Federico Echenique, Alejandro Robinson-Cortés, and Leeat Yariv. An experimental study of decentralized matching.Quantitative Economics, 16(2):497–533, 2025. doi:10.3982/QE2316

    work page doi:10.3982/qe2316 2025

  45. [45]

    From KISS to KIDS – an ‘anti-simplistic’ modelling approach

    Bruce Edmonds and Scott Moss. From KISS to KIDS – an ‘anti-simplistic’ modelling approach. In Paul Davidsson, Brian Logan, and Keiki Takadama, editors,Multi-Agent and Multi-Agent-Based Simulation, pages 130–144, Berlin, Heidelberg, 2005. Springer Berlin Heidelberg

    2005

  46. [46]

    Epstein and Robert L

    Joshua M. Epstein and Robert L. Axtell.Growing Artificial Societies: Social Science from the Bottom Up. The MIT Press, 1996. doi:10.7551/mitpress/3374.001.0001

    work page doi:10.7551/mitpress/3374.001.0001 1996

  47. [47]

    Games of school choice under the boston mechanism.Journal of Public Economics, 90(1):215–237, 2006

    Haluk Ergin and Tayfun Sönmez. Games of school choice under the boston mechanism.Journal of Public Economics, 90(1):215–237, 2006. doi:10.1016/j.jpubeco.2005.02.002

    work page doi:10.1016/j.jpubeco.2005.02.002 2006

  48. [48]

    Featherstone and Muriel Niederle

    Clayton R. Featherstone and Muriel Niederle. Boston versus deferred acceptance in an interim setting: An experimental investigation.Games and Economic Behavior, 100:353–375, 2016. doi:10.1016/j.geb.2016.10.005

    work page doi:10.1016/j.geb.2016.10.005 2016

  49. [49]

    Simcity: Multi-agent urban development simulation with rich interactions.arXiv preprint arXiv:2510.01297, 2025

    Yeqi Feng, Yucheng Lu, Hongyu Su, and Tianxing He. Simcity: Multi-agent urban development simulation with rich interactions.arXiv preprint arXiv:2510.01297, 2025

    Pith/arXiv arXiv 2025

  50. [50]

    Gonczarowski, and Ran I

    Sara Fish, Yannai A. Gonczarowski, and Ran I. Shorrer. Algorithmic collusion by large language models.arXiv preprint arXiv:2404.00806, 2025

    arXiv 2025

  51. [51]

    Improving language model negotiation with self-play and in-context learning from AI feedback.arXiv preprint arXiv:2305.10142, 2023

    Yao Fu, Hao Peng, Tushar Khot, and Mirella Lapata. Improving language model negotiation with self-play and in-context learning from AI feedback.arXiv preprint arXiv:2305.10142, 2023

    arXiv 2023

  52. [52]

    Gale and L

    D. Gale and L. S. Shapley. College admissions and the stability of marriage.The American Mathematical Monthly, 69(1):9–15, 1962

    1962

  53. [53]

    Large language models empowered agent-based modeling and simulation: a survey and perspectives.Humanities and Social Sciences Communications, 11(1):1–24, 2024

    Chen Gao, Xiaochong Lan, Nian Li, Yuan Yuan, Jingtao Ding, Zhilun Zhou, Fengli Xu, and Yong Li. Large language models empowered agent-based modeling and simulation: a survey and perspectives.Humanities and Social Sciences Communications, 11(1):1–24, 2024. doi:10.1057/s41599-024-03611-3

    work page doi:10.1057/s41599-024-03611-3 2024

  54. [54]

    Manipulation of voting schemes: A general result.Econometrica, 41(4):587–601, 1973

    Allan Gibbard. Manipulation of voting schemes: A general result.Econometrica, 41(4):587–601, 1973

    1973

  55. [55]

    Words reveal wants: How well can simple LLM- based AI agents replicate people’s choices based on their social media posts

    Sofie Goethals, Johannes Luther, and Sandra Matz. Words reveal wants: How well can simple LLM- based AI agents replicate people’s choices based on their social media posts. InAdjunct Proceedings of the 33rd ACM Conference on User Modeling, Adaptation and Personalization, pages 126–131, 2025. doi:10.1145/3708319.3733689

    work page doi:10.1145/3708319.3733689 2025

  56. [56]

    Emergence as a construct: History and issues.Emergence, 1:49–72, 1999

    Jeffrey Goldstein. Emergence as a construct: History and issues.Emergence, 1:49–72, 1999. doi:10.1207/s15327000em0101_4

    work page doi:10.1207/s15327000em0101_4 1999

  57. [57]

    Frontiers: Can large language models capture human preferences?Marketing Science, 43(4):709–722, 2024

    Ali Goli and Amandeep Singh. Frontiers: Can large language models capture human preferences?Marketing Science, 43(4):709–722, 2024

    2024

  58. [58]

    The impact of it-enabled globalization: A structural two-sided matching model of online labor market.SSRN Electronic Journal, 2016

    Jing Gong. The impact of it-enabled globalization: A structural two-sided matching model of online labor market.SSRN Electronic Journal, 2016. doi:10.2139/ssrn.2745450

    work page doi:10.2139/ssrn.2745450 2016

  59. [59]

    Parker, Nicholas A

    Igor Grossmann, Matthew Feinberg, Dawn C. Parker, Nicholas A. Christakis, Philip E. Tetlock, and William A. Cunningham. AI and the transformation of social science research.Science, 380(6650):1108–1109, 2023. doi:10.1126/science.adi1778

    work page doi:10.1126/science.adi1778 2023

  60. [60]

    Incentives in teams.Econometrica, 41(4):617–631, 1973

    Theodore Groves. Incentives in teams.Econometrica, 41(4):617–631, 1973

    1973

  61. [61]

    The effectiveness of top-down advice in strategy-proof mechanisms: A field experiment.European Economic Review, 101:505–511, 2018

    Pablo Guillen and Rustamdjan Hakimov. The effectiveness of top-down advice in strategy-proof mechanisms: A field experiment.European Economic Review, 101:505–511, 2018. doi:10.1016/j.euroecorev.2017.10.020

    work page doi:10.1016/j.euroecorev.2017.10.020 2018

  62. [62]

    Strategy-proofness in experimental matching markets.Experimental Economics, 24(2):650–668, 2021

    Pablo Guillen and Róbert F Veszteg. Strategy-proofness in experimental matching markets.Experimental Economics, 24(2):650–668, 2021. doi:10.1007/s10683-020-09665-9

    work page doi:10.1007/s10683-020-09665-9 2021

  63. [63]

    Embodied LLM agents learn to cooperate in organized teams.IEEE Transactions on Computational Social Systems, 2026

    Xudong Guo, Kaixuan Huang, Jiale Liu, Wenhui Fan, Natalia Vélez, Qingyun Wu, Huazheng Wang, Thomas L Griffiths, and Mengdi Wang. Embodied LLM agents learn to cooperate in organized teams.IEEE Transactions on Computational Social Systems, 2026. 32 Do Matching Mechanisms Work with LLM Agents?A PREPRINT

    2026

  64. [64]

    The equitable top trading cycles mechanism for school choice.Interna- tional Economic Review, 59(4):2219–2258, 2018

    Rustamdjan Hakimov and Onur Kesten. The equitable top trading cycles mechanism for school choice.Interna- tional Economic Review, 59(4):2219–2258, 2018. doi:10.1111/iere.12335

    work page doi:10.1111/iere.12335 2018

  65. [65]

    A survey of agent-based modeling practices (January 1998 to July 2008).Journal of Artificial Societies and Social Simulation, 12(4):9, 2009

    Brian Heath, Raymond Hill, and Frank Ciarallo. A survey of agent-based modeling practices (January 1998 to July 2008).Journal of Artificial Societies and Social Simulation, 12(4):9, 2009

    1998

  66. [66]

    Hodges Jr and Erich L

    Joseph L. Hodges Jr and Erich L. Lehmann. Estimates of Location Based on Rank Tests.The Annals of Mathematical Statistics, 34(2):598–611, 1963. doi:10.1214/aoms/1177704172

    work page doi:10.1214/aoms/1177704172 1963

  67. [67]

    Friederike Holderried, Christian Stegemann-Philipps, Lea Herschbach, Julia-Astrid Moldt, Andrew Nevins, Jan Griewatz, Martin Holderried, Anne Herrmann-Werner, Teresa Festl-Wietek, and Moritz Mahling. A generative pretrained transformer (GPT)-powered chatbot as a simulated patient to practice history taking: Prospective, mixed methods study.JMIR Medical Ed...

    2024

  68. [68]

    Large language models as simulated economic agents: What can we learn from homo silicus? Working Paper 31122, National Bureau of Economic Research, 2023

    John J Horton, Apostolos Filippas, and Benjamin S Manning. Large language models as simulated economic agents: What can we learn from homo silicus? Working Paper 31122, National Bureau of Economic Research, 2023

    2023

  69. [69]

    The design of mechanisms for resource allocation.The American Economic Review, 63(2): 1–30, 1973

    Leonid Hurwicz. The design of mechanisms for resource allocation.The American Economic Review, 63(2): 1–30, 1973

    1973

  70. [70]

    Jensen and William H

    Michael C. Jensen and William H. Meckling. Theory of the firm: Managerial behavior, agency costs and ownership structure.Journal of Financial Economics, 3(4):305–360, 1976. doi:10.1016/0304-405X(76)90026- X

    work page doi:10.1016/0304-405x(76)90026- 1976

  71. [71]

    An experimental study of competitive market behavior through LLMs.arXiv preprint arXiv:2409.08357, 2024

    Jingru Jia and Zehua Yuan. An experimental study of competitive market behavior through LLMs.arXiv preprint arXiv:2409.08357, 2024

    arXiv 2024

  72. [72]

    Donald trumps in the virtual polls: Simulating and predicting public opinions in surveys using large language models.arXiv preprint arXiv:2411.01582, 2025

    Shapeng Jiang, Lijia Wei, and Chen Zhang. Donald trumps in the virtual polls: Simulating and predicting public opinions in surveys using large language models.arXiv preprint arXiv:2411.01582, 2025

    arXiv 2025

  73. [73]

    Efficient matching under distributional constraints: Theory and applications

    Yuichiro Kamada and Fuhito Kojima. Efficient matching under distributional constraints: Theory and applications. American Economic Review, 105(1):67–99, 2015. doi:10.1257/aer.20101552

    work page doi:10.1257/aer.20101552 2015

  74. [74]

    ‘Simulacrum of Stories’: Examining large language models as qualitative research participants

    Shivani Kapania, William Agnew, Motahhare Eslami, Hoda Heidari, and Sarah E Fox. ‘Simulacrum of Stories’: Examining large language models as qualitative research participants. InProceedings of the 2025 CHI Conference on Human Factors in Computing Systems, pages 1–17, 2025

    2025

  75. [75]

    The day care assignment: A dynamic matching problem.American Economic Journal: Microeconomics, 6(4):362–406, 2014

    John Kennes, Daniel Monte, and Norovsambuu Tumennasan. The day care assignment: A dynamic matching problem.American Economic Journal: Microeconomics, 6(4):362–406, 2014. doi:10.1257/mic.6.4.362

    work page doi:10.1257/mic.6.4.362 2014

  76. [76]

    School choice with consent.The Quarterly Journal of Economics, 125(3):1297–1348, 2010

    Onur Kesten. School choice with consent.The Quarterly Journal of Economics, 125(3):1297–1348, 2010

    2010

  77. [77]

    Learning to be homo economicus: Can an LLM learn preferences from choice.arXiv preprint arXiv:2401.07345, 2024

    Jeongbin Kim, Matthew Kovach, Kyu-Min Lee, Euncheol Shin, and Hector Tzavellas. Learning to be homo economicus: Can an LLM learn preferences from choice.arXiv preprint arXiv:2401.07345, 2024

    Pith/arXiv arXiv 2024

  78. [78]

    Kirshner, Yiwen Pan, Jason Xianghua Wu, and Alex Gould

    Samuel N. Kirshner, Yiwen Pan, Jason Xianghua Wu, and Alex Gould. Talking terms: Agent information in LLM supply chain bargaining.Decision Sciences, 2025. doi:10.1111/deci.70010

    work page doi:10.1111/deci.70010 2025

  79. [79]

    Can AI with high reasoning ability replicate human-like decision making in economic experiments?Group Decision and Negotiation, 34(6):1303–1326, 2025

    Ayato Kitadai, Sinndy Dayana Rico Lugo, Yudai Tsurusaki, Yusuke Fukasawa, and Nariaki Nishino. Can AI with high reasoning ability replicate human-like decision making in economic experiments?Group Decision and Negotiation, 34(6):1303–1326, 2025. doi:10.1007/s10726-025-09946-9

    work page doi:10.1007/s10726-025-09946-9 2025

  80. [80]

    School choice with asymmetric information: Priority design and the curse of acceptance.Theoretical Economics, 15(3):1095–1133, 2020

    Andrew Kloosterman and Peter Troyan. School choice with asymmetric information: Priority design and the curse of acceptance.Theoretical Economics, 15(3):1095–1133, 2020. doi:10.3982/TE3621

    work page doi:10.3982/te3621 2020

Showing first 80 references.