pith. sign in

arxiv: 2606.12057 · v1 · pith:7Y3GJURCnew · submitted 2026-06-10 · 📊 stat.AP

ChargeBD: Character-Aware Heterogeneous Agent Reasoning for Guided Engineering in Battery Development

Rui Huang , Zekun Jiang , Xingyu Niu , Yuqiang Li , Xinying Gu , Tianhang Zhou This is my paper

Pith reviewed 2026-06-27 07:54 UTC · model grok-4.3

classification 📊 stat.AP
keywords redox-flow batterieslarge language modelsmulti-agent systemsMBTI personasbattery developmentheterogeneous reasoningengineering benchmarks
0
0 comments X

The pith

MBTI-inspired persona agents adapt LLM reasoning for multi-scale redox-flow battery tasks.

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

The paper presents ChargeBD as a framework that deploys 16 agents, each built as a cognitive-bias template, to handle the range of redox-flow battery problems from molecular design through system-level decisions. Generic large language models lack the flexibility to shift between exploratory invention, rule-driven execution, detailed physical modeling, and multi-objective optimization within the same workflow. The authors derive a 500-question benchmark from an initial 50-question RFB task set and use a single base model to score the agents, producing two matrices that record capability and cognitive advantage. This structure is intended to supply guided, task-appropriate reasoning throughout battery development.

Core claim

ChargeBD is a character-aware heterogeneous-agent reasoning framework that starts from a 50-question RFB-specific task set to build a 500-question ESS-LLM Benchmark. It defines 16 MBTI-inspired persona agents as structured cognitive-bias templates and uses DeepSeek-V3-Plus to evaluate them, resulting in a persona capability matrix and a cognitive advantage matrix that together overcome the insufficient adaptability of generic LLM reasoning in innovation, execution, modeling, and trade-offs.

What carries the argument

The 16 MBTI-inspired persona agents defined as structured cognitive-bias templates, evaluated on the ESS-LLM Benchmark to produce a persona capability matrix and cognitive advantage matrix.

Load-bearing premise

That MBTI-inspired persona agents defined as structured cognitive-bias templates will produce meaningfully different and useful reasoning behaviors when applied to the multi-scale, multi-objective RFB task set.

What would settle it

If the 16 persona agents generate statistically indistinguishable reasoning traces and performance scores from one another and from a single generic LLM on the 500-question ESS-LLM Benchmark.

Figures

Figures reproduced from arXiv: 2606.12057 by Rui Huang, Tianhang Zhou, Xingyu Niu, Xinying Gu, Yuqiang Li, Zekun Jiang.

Figure 1
Figure 1. Figure 1: Multi-scale coupling mechanism of redox flow batteries. [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Hierarchical topic organization and category distribution of the ESS-LLM Benchmark. [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Constraint structure and task composition of an RFB molecular-design benchmark problem. [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Standardized evaluation control and persona prompting strategy design. [PITH_FULL_IMAGE:figures/full_fig_p008_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: ChargeBD framework for character-aware heterogeneous-agent reasoning in battery [PITH_FULL_IMAGE:figures/full_fig_p010_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Demand-driven multi-agent wake-up mechanism and dynamic architecture selection. [PITH_FULL_IMAGE:figures/full_fig_p011_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Candidate model performance and stability on the RFB task set. [PITH_FULL_IMAGE:figures/full_fig_p015_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Model generalization and persona prompting validation across energy-storage tasks. [PITH_FULL_IMAGE:figures/full_fig_p015_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Cognitive-advantage matrix of 16 MBTI-inspired persona agents. [PITH_FULL_IMAGE:figures/full_fig_p017_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Blind spots of top single-persona agents across 500 energy-storage tasks. [PITH_FULL_IMAGE:figures/full_fig_p018_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Fixed collaboration and on-demand dynamic activation performance. [PITH_FULL_IMAGE:figures/full_fig_p019_11.png] view at source ↗
read the original abstract

Redox-flow battery (RFB) research spans molecular design, electrolyte optimization, electrode and membrane materials, stack operation, system management, and safety analysis, making it a constrained, multi-scale, and multi-objective energy-storage R&D problem. Although large language models (LLMs) can support scientific knowledge integration and proposal generation, generic LLM reasoning remains insufficiently adaptive across innovation-oriented exploration, rule-based execution, mechanistic modeling, and system-level trade-offs. Here we introduce ChargeBD, a character-aware heterogeneous-agent reasoning framework for guided engineering in battery development. Starting from a 50-question RFB-specific task set, we construct a 500-question ESS-LLM Benchmark and define MBTI-inspired persona agents as structured cognitive-bias templates rather than psychometric instruments or representations of real personalities. DeepSeek-V3-Plus is selected as the shared base model, and 16 MBTI-inspired persona agents are evaluated to construct a persona capability matrix and a cognitive advantage matrix.

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 / 0 minor

Summary. The manuscript introduces ChargeBD, a character-aware heterogeneous-agent reasoning framework for redox-flow battery (RFB) development. It begins with a 50-question RFB-specific task set to construct a 500-question ESS-LLM Benchmark, defines 16 MBTI-inspired persona agents as structured cognitive-bias templates (rather than psychometric instruments), selects DeepSeek-V3-Plus as the shared base model, and evaluates the agents to construct a persona capability matrix and a cognitive advantage matrix.

Significance. If the evaluations establish that the MBTI-inspired personas produce meaningfully differentiated reasoning behaviors that improve outcomes across innovation-oriented exploration, rule-based execution, mechanistic modeling, and system-level trade-offs relative to generic LLM use, the framework could offer a practical method for injecting structured cognitive diversity into LLM-assisted scientific engineering workflows in constrained, multi-objective domains such as energy storage R&D.

major comments (2)
  1. [Abstract] Abstract: The manuscript states the construction of the benchmark and the plan to evaluate 16 persona agents in order to build the capability and advantage matrices, yet supplies no quantitative results, error analysis, baseline comparisons (e.g., against a single unconditioned DeepSeek-V3-Plus run), or statistical validation that persona differences actually improve task outcomes; this absence directly undermines the central claim that the heterogeneous framework overcomes the insufficient adaptability of generic LLM reasoning.
  2. [Section describing matrix construction] Section describing matrix construction: The persona capability matrix and cognitive advantage matrix are presented as derived from evaluations on the 500-question benchmark, but no data, divergence metrics, or evidence is provided that the MBTI-inspired cognitive-bias templates induce non-trivial behavioral differences or that any such differences yield collective gains on the multi-scale RFB task set; without this, the heterogeneous-agent premise reduces to unproven parallel sampling.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive and detailed feedback. The major comments correctly identify that the submitted manuscript describes the benchmark construction and matrix derivation but does not supply the supporting quantitative results or evidence. We address each point below and will incorporate the requested material in revision.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The manuscript states the construction of the benchmark and the plan to evaluate 16 persona agents in order to build the capability and advantage matrices, yet supplies no quantitative results, error analysis, baseline comparisons (e.g., against a single unconditioned DeepSeek-V3-Plus run), or statistical validation that persona differences actually improve task outcomes; this absence directly undermines the central claim that the heterogeneous framework overcomes the insufficient adaptability of generic LLM reasoning.

    Authors: We agree that the current manuscript version presents the framework description and states that evaluations were performed, yet omits the actual quantitative results, error analysis, baseline comparisons, and statistical validation. In the revised manuscript we will add a dedicated Results section containing the persona capability matrix, cognitive advantage matrix, performance differentials versus the unconditioned DeepSeek-V3-Plus baseline, error bars, and statistical tests confirming that the observed persona differences produce measurable gains on the multi-scale RFB tasks. This addition will directly substantiate the central claim. revision: yes

  2. Referee: [Section describing matrix construction] Section describing matrix construction: The persona capability matrix and cognitive advantage matrix are presented as derived from evaluations on the 500-question benchmark, but no data, divergence metrics, or evidence is provided that the MBTI-inspired cognitive-bias templates induce non-trivial behavioral differences or that any such differences yield collective gains on the multi-scale RFB task set; without this, the heterogeneous-agent premise reduces to unproven parallel sampling.

    Authors: The referee is correct that the manuscript asserts the matrices are derived from the 500-question evaluations but supplies neither the underlying data nor divergence metrics demonstrating non-trivial behavioral differentiation or collective gains. We will revise the matrix-construction section to include the full evaluation dataset summary, explicit divergence metrics (response variance, task-specific performance spreads), and comparative analysis showing that the MBTI-inspired templates generate differentiated reasoning trajectories whose combination yields gains beyond parallel sampling of a single model. These additions will address the concern that the heterogeneous premise remains unproven. revision: yes

Circularity Check

0 steps flagged

No significant circularity; framework is self-contained empirical construction

full rationale

The paper presents an agent-based framework without any mathematical derivations, equations, or first-principles predictions. It explicitly defines MBTI-inspired personas as cognitive-bias templates, constructs a benchmark from an initial 50-question set, and builds capability/advantage matrices directly from evaluations on that benchmark. No steps match the enumerated circularity patterns: no self-definitional reductions, no fitted inputs renamed as predictions, no load-bearing self-citations, and no imported uniqueness theorems. The central claim rests on the empirical outcomes of the defined agents rather than reducing to its own inputs by construction. This is the normal case of a self-contained proposal whose validity can be assessed externally.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

Abstract-only review yields minimal ledger entries; the primary untested premise is the domain assumption that MBTI categories supply useful cognitive biases for engineering tasks.

axioms (1)
  • domain assumption MBTI-inspired persona agents defined as structured cognitive-bias templates will yield distinct and advantageous reasoning behaviors on RFB tasks
    Stated directly in the abstract as the basis for constructing the capability and advantage matrices.

pith-pipeline@v0.9.1-grok · 5709 in / 1319 out tokens · 23204 ms · 2026-06-27T07:54:10.592704+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

40 extracted references · 21 canonical work pages · 2 internal anchors

  1. [1]

    Applications of natural language processing and large language modelsinmaterialsdiscovery[J].NPJComputationalMaterials,2025,11(1).DOI:10.1038/s41524- 025-01554-0

    Jiang X, Wang W, Tian S, et al. Applications of natural language processing and large language modelsinmaterialsdiscovery[J].NPJComputationalMaterials,2025,11(1).DOI:10.1038/s41524- 025-01554-0

    work page doi:10.1038/s41524- 2025

  2. [2]

    Zhang J, Li J, Zhao G, et al. Mining Solid-State Electrolytes from Metal–Organic Framework DatabasesthroughLargeLanguageModelsandRepresentationClustering[J].JournaloftheAmer- ican Chemical Society, 2025, 147(44): 40496-40506. 23

    2025

  3. [3]

    The application of large language models in energy storage research

    Zhong, Y.; Leng, Y.; Chen, S.; Li, P.; Zou, Z.; Liu, Y.; Wan, J. Accelerating battery research with retrieval-augmented large language models: Present and future. Energy Storage Science and Technology 2024, 13(9), 3214-3225. DOI:10.19799/j.cnki.2095-4239.2024.0604

    work page doi:10.19799/j.cnki.2095-4239.2024.0604 2024

  4. [4]

    The application of large language models in energy storage research

    Yuan, Y.; Gao, Y.; Zhang, J.; Gao, Y.; Wang, C.; Chen, X.; Zhang, Q. The application of large language models in energy storage research. Energy Storage Science and Technology 2024, 13(9), 2907-2919. DOI:10.19799/j.cnki.2095-4239.2024.0176

    work page doi:10.19799/j.cnki.2095-4239.2024.0176 2024

  5. [5]

    Zuo, W.; Zheng, H.; He, T.; Vishwanath, V.; Chan, M. K. Y.; Stevens, R. L.; Amine, K.; Xu, G.-L. Large language models for batteries.Joule2025, 9, 102037. DOI:10.1016/j.joule.2025.102037

    work page doi:10.1016/j.joule.2025.102037 2025

  6. [6]

    A generative model for inorganic materials design[J]

    Zeni C, Pinsler R, Zügner D, et al. A generative model for inorganic materials design[J]. Nature, 2025, 639(8055): 624-632

    2025

  7. [7]

    Large language models as molecu- lar design engines[J]

    Bhattacharya D, Cassady H J, Hickner M A, et al. Large language models as molecu- lar design engines[J]. Journal of Chemical Information and Modeling, 2024, 64:7086-7096. DOI:10.1021/acs.jcim.4c01396

    work page doi:10.1021/acs.jcim.4c01396 2024

  8. [8]

    Theodoris, Ling Xiao, Anant Chopra, Mark D

    Wang, H.; Fu, T.; Du, Y.; Gao, W.; Huang, K.; Liu, Z.; Chandak, P.; Liu, S.; et al. Scientific discovery in the age of artificial intelligence.Nature2023, 620, 47-60. DOI:10.1038/s41586-023- 06221-2

    work page doi:10.1038/s41586-023-

  9. [9]

    Boiko,D.A.;MacKnight,R.;Kline,B.;Gomes,G.Autonomouschemicalresearchwithlargelan- guagemodels.Nature2023,624,570-578.DOI:10.1038/s41586-023-06792-0;arXiv:2304.05332

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1038/s41586-023-06792-0;arxiv:2304.05332

  10. [10]

    Automating structure–activity analysis for electrochemical nitrogen reduction catalyst design through multi-agent collaborations[J]

    Hu X, Chen S, Chen L, et al. Automating structure–activity analysis for electrochemical nitrogen reduction catalyst design through multi-agent collaborations[J]. National Science Review, 2025, 12(11): nwaf372

    2025

  11. [11]

    ChemCrow: Augmenting large-language models with chemistry tools

    Bran, A. M.; Cox, S.; Schilter, O.; Baldassari, C.; White, A. D.; Schwaller, P. Augmenting large language models with chemistry tools.Nature Machine Intelligence2024, 6, 525-535. DOI:10.1038/s42256-024-00832-8; arXiv:2304.05376

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1038/s42256-024-00832-8

  12. [12]

    Ghafarollahi, A.; Buehler, M. J. SciAgents: Automating Scientific Discovery Through Bioin- spired Multi-Agent Intelligent Graph Reasoning.Advanced Materials2025, 37(22), e2413523. DOI:10.1002/adma.202413523; arXiv:2409.05556

    work page doi:10.1002/adma.202413523

  13. [13]

    SLM-MATRIX: a multi-agent trajectory reason- ing and verification framework for enhancing language models in materials data extraction.npj Computational Materials2025, 11, 241

    Li, X.; Huang, Z.; Quan, S.; Peng, C.; Ma, X. SLM-MATRIX: a multi-agent trajectory reason- ing and verification framework for enhancing language models in materials data extraction.npj Computational Materials2025, 11, 241. DOI:10.1038/s41524-025-01719-x

    work page doi:10.1038/s41524-025-01719-x

  14. [14]

    Agent-based multimodal information extraction for nanomaterials

    Odobesku,R.;Romanova,K.;Mirzaeva,S.;Zagorulko,O.;Sim,R.;Khakimullin,R.;Razlivina,J.; Dmitrenko, A.; Vinogradov, V. Agent-based multimodal information extraction for nanomaterials. npj Computational Materials2025, 11, 194. DOI:10.1038/s41524-025-01674-7

    work page doi:10.1038/s41524-025-01674-7

  15. [15]

    Rupprecht, S.; Gao, Q.; Karia, T.; Schweidtmann, A. M. Multi-agent systems for chemical engi- neering: a review and perspective.Current Opinion in Chemical Engineering2026, 51, 101209. DOI:10.1016/j.coche.2025.101209; arXiv:2508.07880

    work page doi:10.1016/j.coche.2025.101209 2025

  16. [16]

    D.; Tanikanti, A.; Keceli, M

    Pham, T. D.; Tanikanti, A.; Keceli, M. ChemGraph as an agentic framework for computational chemistryworkflows.Communications Chemistry2026,9,33.DOI:10.1038/s42004-025-01776-9; arXiv:2506.06363

    work page doi:10.1038/s42004-025-01776-9

  17. [17]

    Beyond chemical qa: Evaluating llm’s chemical reasoning with modular chemical operations[J]

    Hao L, Cao H, Feng B, et al. Beyond chemical qa: Evaluating llm’s chemical reasoning with modular chemical operations[J]. Advances in Neural Information Processing Systems, 2026, 38. 24

    2026

  18. [18]

    Elbeheiry, María Victoria Gil, Christina Glaubitz, Maximilian Greiner, Caroline T

    Mirza, A.; Alampara, N.; Kunchapu, S.; Rios-Garcia, M.; Emoekabu, B.; Krishnan, A.; Gupta, T.; Schilling-Wilhelmi, M.; et al. A framework for evaluating the chemical knowledge and reasoning abilities of large language models against the expertise of chemists.Nature Chemistry2025, 17, 1027-1034. DOI:10.1038/s41557-025-01815-x; arXiv:2404.01475

    work page doi:10.1038/s41557-025-01815-x

  19. [19]

    D.; Zhou, D.; Li, L.; Wang,W.MatSciBench: Benchmarkingthereasoningabilityoflargelanguagemodelsinmaterials science

    Zhang, J.; Gan, J.; Wang, X.; Jia, Z.; Gu, C.; Chen, J.; Zhu, Y.; Ma, M. D.; Zhou, D.; Li, L.; Wang,W.MatSciBench: Benchmarkingthereasoningabilityoflargelanguagemodelsinmaterials science. arXiv preprint arXiv:2510.12171, 2025

    Pith/arXiv arXiv 2025

  20. [20]

    DOI:10.1038/s41560-021-00796-8

    Sepulveda,N.A.;Jenkins,J.D.;Edington,A.;Mallapragada,D.S.;Lester,R.K.Thedesignspace for long-duration energy storage in decarbonized power systems.Nature Energy2021, 6, 506-516. DOI:10.1038/s41560-021-00796-8

    work page doi:10.1038/s41560-021-00796-8

  21. [21]

    DOI:10.1007/s10800-011- 0348-2

    Weber,A.Z.;Mench,M.M.;Meyers,J.P.;Ross,P.N.;Gostick,J.T.;Liu,Q.Redoxflowbatteries: a review.Journal of Applied Electrochemistry2011, 41, 1137-1164. DOI:10.1007/s10800-011- 0348-2

    work page doi:10.1007/s10800-011-

  22. [22]

    M.; Gallagher, K

    Darling, R. M.; Gallagher, K. G.; Kowalski, J. A.; Ha, S.; Brushett, F. R. Pathways to low-cost electrochemical energy storage: a comparison of aqueous and nonaqueous flow batteries.Energy & Environmental Science2014, 7, 3459-3477. DOI:10.1039/C4EE02158D

    work page doi:10.1039/c4ee02158d

  23. [23]

    Modelling and estimation of vanadium redox flow batteries: a review.Batteries2022, 8, 121

    Puleston, T.; Clemente, A.; Costa-Castello, R.; Serra, M. Modelling and estimation of vanadium redox flow batteries: a review.Batteries2022, 8, 121. DOI:10.3390/batteries8090121

    work page doi:10.3390/batteries8090121

  24. [24]

    Improved electrochemical performance for vanadium flow battery by optimizing the concentration of the electrolyte[J]

    Jing M, Wei Z, Su W, et al. Improved electrochemical performance for vanadium flow battery by optimizing the concentration of the electrolyte[J]. Journal of Power Sources, 2016, 324: 215-223

    2016

  25. [25]

    Journal of Power Sources, 2026, 667: 239216

    ShaheenI,ChiuWH,LeeYX,etal.Heterogeneousgraphitefeltelectrodesdecoratedwithnanos- tructured graphitic carbon nitride for enhanced redox kinetics in vanadium redox flow batteries[J]. Journal of Power Sources, 2026, 667: 239216

    2026

  26. [26]

    Nafion-Based Proton Exchange Membranes for Vanadium Redox Flow Batter- ies[J]

    He S, Chai S, Li H. Nafion-Based Proton Exchange Membranes for Vanadium Redox Flow Batter- ies[J]. ChemSusChem, 2025, 18(10): e202402506

    2025

  27. [27]

    Characterization and scale-up of serpentine and interdigitated flow fields for application in commercial vanadium redox flow batteries[J]

    Gundlapalli R, Bhattarai A, Ranjan R, et al. Characterization and scale-up of serpentine and interdigitated flow fields for application in commercial vanadium redox flow batteries[J]. Journal of Power Sources, 2022, 542: 231812

    2022

  28. [28]

    Journal of Power Sources, 2021, 490: 229514

    ZouT,ShiX,YuL.Studyonenergylossof35kWallvanadiumredoxflowbatteryenergystorage system under closed-loop flow strategy[J]. Journal of Power Sources, 2021, 490: 229514

    2021

  29. [29]

    Z.; Stinis, P.; Tartakovsky, A

    He, Q. Z.; Stinis, P.; Tartakovsky, A. M. Physics-constrained deep neural network method for estimating parameters in a redox flow battery.Journal of Power Sources2022, 528, 231147. DOI:10.1016/j.jpowsour.2022.231147; arXiv:2106.11451

    work page doi:10.1016/j.jpowsour.2022.231147 2022

  30. [30]

    Physics-informed machine learning of redox flow battery based on a two-dimensional unit cell model.Journal of Power Sources2023, 584, 233548

    Chen, W.; Fu, Y.; Stinis, P. Physics-informed machine learning of redox flow battery based on a two-dimensional unit cell model.Journal of Power Sources2023, 584, 233548. DOI:10.1016/j.jpowsour.2023.233548; arXiv:2306.01010

    work page doi:10.1016/j.jpowsour.2023.233548 2023

  31. [31]

    Accelerating battery innovation: AI-powered molecular discovery[J]

    Gao Y C, Chen X, Yuan Y H, et al. Accelerating battery innovation: AI-powered molecular discovery[J]. Chemical Society Reviews, 2025, 54(21): 9630-9684

    2025

  32. [32]

    Machine learning toward electrochemical energy storage materials and devices: A review[J]

    Ma C, Yao C, Xu J, et al. Machine learning toward electrochemical energy storage materials and devices: A review[J]. Sustainable Materials and Technologies, 2026: e01990

    2026

  33. [33]

    Chemagent: Self-updating library in large language models improves chemical reasoning[J]

    Tang X, Hu T, Ye M, et al. Chemagent: Self-updating library in large language models improves chemical reasoning[J]. arXiv preprint arXiv:2501.06590, 2025. 25

    arXiv 2025

  34. [34]

    BestaM,ChandranS,GerstenbergerR,etal.PsychologicallyenhancedAIagents[J].arXivpreprint arXiv:2509.04343, 2025

    arXiv 2025

  35. [35]

    Two tales of persona in llms: A survey of role-playing andpersonalization[C]//FindingsoftheAssociationforComputationalLinguistics: EMNLP2024

    Tseng Y M, Huang Y C, Hsiao T Y, et al. Two tales of persona in llms: A survey of role-playing andpersonalization[C]//FindingsoftheAssociationforComputationalLinguistics: EMNLP2024. 2024: 16612-16631

    2024

  36. [36]

    InProceedings of ACL 2024, 1840-1873

    Wang, X.; Xiao, Y.; Huang, J.; Yuan, S.; Xu, R.; Guo, H.; Tu, Q.; Fei, Y.; Leng, Z.; Wang, W.; Chen,J.;Li,C.;Xiao,Y.InCharacter: Evaluatingpersonalityfidelityinrole-playingagentsthrough psychological interviews. InProceedings of ACL 2024, 1840-1873. DOI:10.18653/v1/2024.acl- long.102

    work page doi:10.18653/v1/2024.acl- 2024

  37. [37]

    Machine Mindset: an MBTI exploration of large language models

    Cui, J.; Lv, L.; Wen, J.; Wang, R.; Tang, J.; Tian, Y.; Yuan, L. Machine Mindset: an MBTI exploration of large language models. arXiv preprint arXiv:2312.12999, 2023

    arXiv 2023

  38. [38]

    Palo Alto: Consulting Psychologists Press, 1985

    MyersI,McCaulleyM.MBTIManual: AGuidetotheDevelopmentandUseoftheMyers-Briggs Type Indicator[M]. Palo Alto: Consulting Psychologists Press, 1985

    1985

  39. [39]

    Psychological Types[M]

    Jung C G. Psychological Types[M]. London: Routledge, 1923

    1923

  40. [40]

    Using MBTI for the success assessment of engineering teams in project-based learning[J]

    Rodríguez Montequín V, Mesa Fernández J M, Balsera J V, et al. Using MBTI for the success assessment of engineering teams in project-based learning[J]. International journal of technology and design education, 2013, 23(4): 1127-1146. 26 Supplementary Information Supplementary Note S1. Prompt templates and persona-agent definitions Theprompt-sourcearchivec...

    2013