Pith. sign in

REVIEW 4 major objections 5 minor 29 references

Negotiating Comfort: Simulating Personality-Driven LLM Agents in Shared Residential Social Networks

T0 review · 4 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read This paper claims that LLM-powered generative agents can simulate personality-driven social negotiation in a shared residential building, with positive personalities producing happier, more connected communities.

desk verdict A genuinely useful LLM-agent methodology demo whose headline statistical claims rest on single runs and prompt-steered traits; worth refereeing for the methods, not for the correlations as established. read the letter →

arxiv 2507.09657 v1 pith:IT5TLOPY submitted 2025-07-13 cs.SI cs.MA

Ann Nedime Nese Rende , Tolga Yilmaz , Özgür Ulusoy This is my paper
classification cs.SIcs.MA
keywords generativeagentslargelanguagemodelssocialnetworksimulationbuildingenergymodelingpersonalitytraitsagent-basedhappinesscentralheating
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

Generative agents are LLM chatbots asked to behave as specific people, and this paper tries to prove they can carry a social simulation. On each of 30 days, 116 virtual residents in a 34-apartment building first negotiate a temperature with their family, then their representatives negotiate with friends; every agent is prompted with personality adjectives, a heater preference, a happiness score, and the weather. The paper compares three runs in which 100%, 50%, or 0% of assigned trait adjectives are positive, and reports that the all-positive run has higher average happiness and strengthening friendships while the all-negative run shows declining friendships. A sympathetic reader would care because, if true, LLM agents could prototype social and energy decisions that would be costly or intrusive to test with real people.

What carries the argument

The load-bearing mechanism is the two-stage prompt-engineered decision loop. Each day, every agent receives a text identity card containing its nine personality adjective pairs, heater preference, happiness, and outside temperature; the LLM returns a degree suggestion and a new happiness. Family suggestions are averaged, and each Family Representative then receives friends' closeness levels and recent votes before returning a final building vote plus friendship weight updates. The statistical machinery is the Correlated Random Effects model, a panel-data estimator chosen after a Hausman test rejected random effects, which lets the authors estimate the contribution of time-invariant traits such as assertiveness and temperature preference while pooling within- and between-agent variation.

What would settle it

Repeat each of the three settings with at least ten random seeds and compare the distributions of daily average happiness and friendship weights; if the all-positive and all-negative confidence intervals overlap once seed-to-seed variation is included, the paper's central contrast is not supported.

Watch

Extended reading notes

Core claim

On its own terms, this paper establishes that LLM-powered generative agents can simulate personality-driven negotiation in a shared residential building and that the resulting trends are statistically measurable. In a 30-day, 116-agent simulation built on a real-world social network, the all-positive personality run averaged 91.01 happiness against 85.99 for the all-negative run, with friendship weight rising at 0.0298 per day in the positive run and falling at -0.0295 in the negative run. Node-level regressions on 3,480 agent-days find that each extra degree an agent suggests adds 0.2468 happiness points, that assertiveness raises happiness by 3.2413 points, and that selflessness lowers it by 2.6886 points, while warm and hot temperature preferences cost 7.7128 and 13.584 happiness points respectively because the building settles near 21-22°C. The paper's claim is that these effects make personality and preference meaningful drivers of the emergent social outcomes.

Load-bearing premise

The load-bearing assumption is that a single 30-day run per personality setting gives a representative picture, with each agent-day treated as an independent observation; if the runs are not representative or the observations are correlated, the reported differences between positive and negative personalities could be noise.

Editorial extensions

If this is right

  • All-positive trait settings end with higher average happiness, around 91.0 versus 86.0 for all-negative, and rising friendship weights, while all-negative settings show falling friendship weights.
  • The building temperature converges to 21-22°C regardless of the personality mix, so agents with warm or hot preferences systematically lose happiness in the collective decision.
  • Individual-level factors are measurable: each extra degree an agent suggests adds roughly 0.25 happiness points, assertiveness adds about 3.24, and selflessness removes about 2.69.
  • LLM-driven generative agents can generate realistic daily occupant decisions for building energy modeling where human data are scarce or privacy-sensitive.

Reading between the lines

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

  • Editorial inference: the universal 21-22°C convergence may reflect anchoring to the 'Neutral' range in the prompt's reference table as much as genuine emergent consensus; a direct test would shift or remove that reference anchor.
  • Editorial inference: because the node-level analysis uses one 30-day run and treats agent-days as independent observations, the reported p-values are likely optimistic; multiple seeds with clustered standard errors could change the significance profile.
  • Editorial inference: the same family-then-building polling design could be applied to other shared-resource negotiations, such as water, electricity, or common-space decisions, to see whether personality effects generalize across domains.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The manuscript presents an LLM-based generative-agent simulation of a shared residential building. Thirty-four households are derived from Zachary's Karate Club network; each day family members propose a heating temperature, family representatives aggregate these proposals, and representatives then vote in a building-wide poll. Agents are assigned personality adjectives from Big Five facets, temperature preferences, and evolving friendship closeness weights, and their decisions are produced by a quantized Mistral 7B model. The paper runs simulations under three trait distributions (all positive, 50% positive, all negative) and analyzes the outputs with network-level OLS regressions and node-level Correlated Random Effects models. The headline claims are that positive trait distributions correlate with higher average happiness and stronger friendships, and that assertiveness, selflessness, and temperature preferences significantly predict individual happiness and degree choices.

Significance. If the statistical claims were properly supported, the paper would make a useful methodological contribution: it demonstrates an end-to-end integration of generative agents with a social-network simulation framework (Crowd), including prompt design, JSON parsing, and multi-day data collection. The example prompts and outputs in Figures 6-9 are clear and the observed building-level temperature convergence to the neutral range is an interesting emergent pattern. However, the paper's central empirical findings rest on inference from single stochastic runs and on standard errors that ignore the data's dependence structure. The methodological demonstration is credible, but the statistical evidence for the personality-correlation claims is not yet established. The authors do not provide code or data artifacts, so reproducibility cannot be independently checked from the manuscript alone.

major comments (4)
  1. [Section 4.1, Table 2] The network-level trends in Table 2 are OLS slopes over 15 daily observations from a single simulation run per trait setting, yet the simulation is stochastic in at least the random assignment of family sizes and the LLM decoding process. With no multiple seeds, the reported p-values only summarize how well a line fits one autocorrelated trajectory; they cannot support the claim that the all-positive setting produces increasing friendship weight while the all-negative setting declines, because those differences could be run-specific. A concrete fix is to rerun each setting with several seeds, report the distribution of slopes, and use run-level cluster-robust standard errors or a mixed model with run random effects.
  2. [Section 4.2, Tables 4 and 5] The CRE models use 3480 node-day observations from one 30-day run in the 50% positive setting. The standard errors treat each node-day as independent, but residuals are correlated within agents over time (happiness carries over from previous days), within days (shared weather and a single building temperature), and between friends whose choices are fed into each other's prompts. With only one run there is no way to separate the trait effect from the particular stochastic trajectory. I would ask for cluster-robust standard errors on agent and day, or a two-way cluster, plus multiple seeds, before the significant coefficients in Tables 4 and 5 can be interpreted as evidence about personality effects.
  3. [Section 3.5, Figures 6 and 8; Section 4.2] The prompts explicitly ask the LLM to 'mention how your traits influence your decision' and list the agent's personality adjectives. The regression estimates for assertiveness and selflessness in Tables 4 and 5 therefore partly reflect the experimental manipulation, because the trait labels are provided as inputs, rather than necessarily an emergent social process. This is a validity concern rather than a fatal flaw, but the paper should either be reframed as a test of prompt-driven trait effects or supplemented with an ablation in which trait labels are withheld, or with a rule-based baseline, to show that the observed associations are not simply the LLM following the instruction to use the listed traits.
  4. [Section 4.1, Table 3] The pooled regression in Table 3 stacks 30 days from the 50% positive run with 15 days from each of the other two runs, giving 60 observations from three settings. These observations are not independent: each setting contributes a single autocorrelated time series, and the regressor 'average friendship weight' is an outcome of the same simulation dynamics as happiness. The reported p-values therefore do not provide a valid basis for the claim that friendship weight causes happiness. A multi-level model with run-level random effects, or at least Newey-West standard errors treating each run as a cluster, would be needed.
minor comments (5)
  1. [Table 1] The row label 'Alturism' should be 'Altruism'.
  2. [Table 5] The footnote contains the typo 'p =< 0.001'; it should read 'p < 0.001'.
  3. [Section 4.1] The sentence 'The mean happiness over the 15 iterations' should say '15 days' or '15 daily observations' to match the rest of the section.
  4. [Section 4] The paper does not explain why the all-positive and all-negative settings were run for only 15 days while the 50% positive setting ran for 30 days; this asymmetry should be stated, and its implications for the comparability of the slopes in Table 2 should be discussed.
  5. [Section 3.4, Eq. (1)] Because C is set to 1, the 'cost' variable is not in meaningful physical or monetary units; a brief statement that the cost is an ordinal proxy for energy use would improve interpretability.

Circularity Check

3 steps flagged · score 7.0 of 10

The headline correlations and trait-significance regressions are restatements of the prompt instructions that generated the simulated outcomes.

  1. self definitional [Section 3.5, Figure 6; Section 4.2, Table 4]
    "**Second task**: Set a new happiness for yourself (an integer from 1 to 100) based on: - The building-wise degree set yesterday: 22. - Your traits ... The results indicate that the most significant regressors are degree choice, agents' mean degree choice, assertiveness, and temperature preferences."

    The simulated happiness values are generated by an LLM explicitly instructed to 'set a new happiness ... based on ... Your traits.' Table 4 then regresses those same generated happiness values on trait dummies and reports assertiveness and selflessness as significant predictors. The regression is therefore reading back the data-generating instruction: because the trait variables were supplied as the stated determinants of happiness, their significant coefficients are enforced by the construction of the outcome variable, not discovered from an independent signal. Presenting this as an empirical 'impact' of traits on happiness is circular.

  2. self definitional [Section 3.5, Figure 6 and Figure 8; Section 4.2, Table 5]
    "**First task**: Decide the best heater degree **today** based on: - Your personality traits, - Heater preference of you and your family, - The current outside temperature, - Your current happiness level. ... Consistent with the estimates from Table 4, ... the sign of the coefficient changes for the personality traits assertiveness and selflessness."

    Degree choices are produced by an LLM whose prompt instructs it to decide 'based on' personality traits, heater preferences, outside temperature, and happiness. Table 5 regresses those same generated degree choices on trait and temperature-preference dummies and reports significant coefficients for assertiveness, selflessness, and warm/hot preferences. The 'finding' that these traits and preferences predict degree choice is a direct consequence of making those variables the stated inputs to the decision task; it cannot validate any causal claim beyond the prompt's own specification.

1 more flagged steps
  1. self definitional [Section 3.5, Figure 8; Section 4.1, Table 2 and Figures 10a-10b]
    "**Second Task**: You may choose to update your closeness levels to your friends based on: - How aligned their decisions in the last 3 days were with yours - Your personality traits ... Table 2, Figures 10a and 10b show that both metrics increase over time in the all-positive case, while they decline in the all-negative case."

    Friendship closeness is an output of the Phase 2 prompt, which explicitly asks the LLM to update closeness 'based on ... Your personality traits.' The network-level result that all-positive trait distributions increase friendship weight while all-negative distributions decrease it is therefore a readout of the same instruction, reinforced by the semantic content of the adjective pairs (cooperative/selfless vs uncooperative/selfish). The paper's claim that positive traits correlate with stronger friendships restates the prompt construction rather than an emergent, unprompted social dynamic.

full rationale

The paper is a simulation study, and it is not circular merely to observe that a generative agent's outputs vary with its inputs; that is the purpose of a simulation. However, the paper goes further and presents regressions as evidence that personality traits and temperature preferences 'have a significant impact' on happiness, degree choices, and friendship closeness. Those regressions are not independent tests of a hypothesis: the data-generating prompts explicitly instruct the LLM to set happiness based on traits, to choose degrees based on traits and preferences, and to update closeness based on traits. The regression tables recover exactly the inputs that were written into the data-generating instructions, so the headline correlations are enforced by construction rather than discovered. The self-citation of Crowd is not load-bearing here, and the single-run, unclustered standard errors are a separate statistical-correctness concern rather than the basis of this circularity finding. Overall, the central 'findings' reduce to the prompt design, giving a circularity score of 7.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The central claims rest on the assumptions that the LLM faithfully role-plays the given personality traits, that the resulting self-reported happiness is a meaningful measurement, and that the regression models can treat these autocorrelated simulation days as independent. No new physical entities are introduced; agents, happiness scores, and friendship weights are computational constructs of the simulation.

free parameters (5)
  • Cost constant C = 1
    Set to 1 in Eq. 1 for simplicity. It scales the cost metric and does not affect the happiness or friendship findings, but it is a hand-chosen model constant.
  • Strong friendship threshold = weight > 3
    Hand-defined cutoff for counting 'strong friendships' in Section 3.4. This directly defines a key network-level outcome, and different thresholds would change the trend results in Table 2.
  • Heater preference temperature ranges = Cold <18; Cool 18-20; Neutral 21-24; Warm 25-27; Hot >27
    The reference table is included in every prompt (Figures 6 and 8), so it directly shapes the LLM's degree choices and is therefore a hidden input to the regression findings.
  • Positive trait percentage = 100%, 50%, 0%
    The three experimental settings chosen by the modelers. They define the independent variable for the network-level comparison but are not fitted to data.
  • Family size distribution = n in [0,4]
    Each representative randomly gets 0 to 4 family members, which determines the network structure (116 nodes, 246 edges). The distribution is chosen by hand, not calibrated.
assumptions (4)
  • domain assumption The LLM (Mistral 7B) role-plays the assigned personality traits accurately enough that its decisions are caused by those traits.
    Section 3.5 relies on the model following the prompt instructions to act as a person with the given adjectives. If the model ignores or poorly represents traits, the regression findings do not reflect personality.
  • domain assumption Daily observations and node-day observations within a single simulation run can be treated as independent for statistical inference.
    Section 4 uses OLS and CRE models on time-series data from one run per setting without clustering or replication. The p-values assume independent errors, but the simulation outputs are autocorrelated and generated by a single stochastic process.
  • domain assumption The happiness value reported by the LLM is a meaningful measurement of agent well-being.
    Happiness is self-reported by the LLM in response to prompt definitions (Section 3.3). It is an internal construct of the simulation, not a validated psychological measure.
  • domain assumption Zachary's Karate Club network is an appropriate proxy for friendships in a residential building.
    Section 3.1 loads this classic social network to define representatives and friendship weights. Its representativeness for a residential building is not tested or justified beyond being 'widely utilized'.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Negotiating Comfort: Simulating Personality-Driven LLM Agents in Shared Residential Social Networks." pith.science (2026). https://pith.science/paper/IT5TLOPY

@misc{pith2026250709657,
  author       = {Pith},
  title        = {Pith review of: Negotiating Comfort: Simulating Personality-Driven LLM Agents in Shared Residential Social Networks},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IT5TLOPY}},
  note         = {Machine review of arXiv:2507.09657}
}
read the original abstract

We use generative agents powered by large language models (LLMs) to simulate a social network in a shared residential building, driving the temperature decisions for a central heating system. Agents, divided into Family Members and Representatives, consider personal preferences, personal traits, connections, and weather conditions. Daily simulations involve family-level consensus followed by building-wide decisions among representatives. We tested three personality traits distributions (positive, mixed, and negative) and found that positive traits correlate with higher happiness and stronger friendships. Temperature preferences, assertiveness, and selflessness have a significant impact on happiness and decisions. This work demonstrates how LLM-driven agents can help simulate nuanced human behavior where complex real-life human simulations are difficult to set.

Figures

Figures reproduced from arXiv: 2507.09657 by the authors.

Figure 1
Figure 1. Network visualization of created families where nodes are colored [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Network visualization of created families where nodes are colored [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 4
Figure 4. Inputs provided to LLM and outputs obtained in the first phase. [PITH_FULL_IMAGE:figures/full_fig_p005_4.png] view at source ↗
Figures from the paper (7 more)
Figure 3
Figure 3. Figure 3: The methods passed to Crowd framework and their execution order. [PITH_FULL_IMAGE:figures/full_fig_p005_3.png]
Figure 5
Figure 5. Figure 5: Inputs provided to LLM and outputs obtained in the second phase. [PITH_FULL_IMAGE:figures/full_fig_p005_5.png]
Figure 6
Figure 6. Figure 6: Phase 1 sample prompt. Reasoning: Given my cold preference, the outside temperature of -3 degrees, and the fact that I am uncooperative and selfish, I would like to set the heater degree as low as possible to minimize the comfort of others while still maintaining a liv…
Figure 7
Figure 7. Figure 7: Phase 1 sample output. within the instructions. These suggestions, provided by Chat￾GPT (OpenAI), improve the quality of the reasoning provided in each query’s output and show that larger models can be uti￾lized to polish the input for smaller models for these types of…
Figure 8
Figure 8. Figure 8: Phase 2 sample prompt. Reasoning: As an environmentalist and frugal individual, I aim to conserve energy and reduce costs. Given the current outside temperature of -2 degrees, I believe it's unnecessary to set the heater too high. I also consider the closeness levels o…
Figure 10
Figure 10. Figure 10: Charts generated with Crowd. (a) Average weight of edges between friends over time. (b) Number of strong friendships (weight [PITH_FULL_IMAGE:figures/full_fig_p008_10.png]
Figure 11
Figure 11. Figure 11: Example output showing the agents choosing the higher limit of [PITH_FULL_IMAGE:figures/full_fig_p010_11.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

29 extracted references · 24 canonical work pages

  1. [1]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in ":" * " " * FUNCTION f...

  2. [2]

    , author Miyashita, Y

    author Almashor, M. , author Miyashita, Y. , author West, S. , author Dong, T.V.D. , year 2024 . title Can private llm agents synthesize household energy consumption data? , in: booktitle Proceedings of the 15th ACM International Conference on Future and Sustainable Energy Systems , pp. pages 664--668

    work page 2024

  3. [3]

    , year 2023

    author Ansaldo, G. , year 2023 . title AgentSpeak: A Framework for Agent-Based Modeling with Integrated Large Language Models; Case Study: Analyzing Policy Interventions in Electric Vehicle Adoption . Master's thesis. Northeastern University

    work page 2023

  4. [4]

    , author Ji, W

    author Chen, H. , author Ji, W. , author Xu, L. , author Zhao, S. , year 2023 . title Multi-agent consensus seeking via large language models . journal arXiv preprint arXiv:2310.20151

    arXiv 2023

  5. [5]

    , author Zhang, G

    author Chen, S. , author Zhang, G. , author Xia, X. , author Chen, Y. , author Setunge, S. , author Shi, L. , year 2021 . title The impacts of occupant behavior on building energy consumption: A review . journal Sustainable Energy Technologies and Assessments volume 45 , pages 101212

    work page 2021

  6. [6]

    , author McCrae, R.R

    author Costa Jr, P.T. , author McCrae, R.R. , year 1995 . title Domains and facets: Hierarchical personality assessment using the revised neo personality inventory . journal Journal of personality assessment volume 64 , pages 21--50

    work page 1995

  7. [7]

    , author Anderson, K.S

    author Jensen, A.R. , author Anderson, K.S. , author Holmgren, W.F. , author Mikofski, M.A. , author Hansen, C.W. , author Boeman, L.J. , author Loonen, R. , year 2023 . title pvlib iotools—open-source python functions for seamless access to solar irradiance data . journal Solar Energy volume 266 , pages 112092

    work page 2023

  8. [8]

    , author Sablayrolles, A

    author Jiang, A. , author Sablayrolles, A. , author Mensch, A. , author Bamford, C. , author Chaplot, D. , author Casas, D. , author Bressand, F. , author Lengyel, G. , author Lample, G. , author Saulnier, L. , et al., year 2024 a. title Mistral 7b. arxiv 2023

    work page 2024

Show all 29 references
  1. [9]

    , author Ma, Z

    author Jiang, G. , author Ma, Z. , author Zhang, L. , author Chen, J. , year 2024 b. title Eplus-llm: A large language model-based computing platform for automated building energy modeling . journal Applied Energy volume 367 , pages 123431

    2024

  2. [10]

    , author Nweye, K

    author Kaspar, K. , author Nweye, K. , author Buscemi, G. , author Capozzoli, A. , author Nagy, Z. , author Pinto, G. , author Eicker, U. , author Ouf, M.M. , year 2024 . title Effects of occupant thermostat preferences and override behavior on residential demand response in c...

    2024

  3. [11]

    , author Wang, Z.H

    author Li, H. , author Wang, Z.H. , author Zhang, B. , year 2023 . title How social interaction induce energy-saving behaviors in buildings: Interpersonal & passive interactions vs public & active interactions . journal Energy Economics volume 118 , pages 106515

    2023

  4. [12]

    , author Zhao, H

    author Lin, J. , author Zhao, H. , author Zhang, A. , author Wu, Y. , author Ping, H. , author Chen, Q. , year 2023 . title Agentsims: An open-source sandbox for large language model evaluation . journal arXiv preprint arXiv:2308.04026

    2023 arXiv

  5. [13]

    , author Wang, Q.C

    author Liu, X. , author Wang, Q.C. , author Jian, I.Y. , author Chi, H.L. , author Yang, D. , author Chan, E.H.W. , year 2021 . title Are you an energy saver at home? the personality insights of household energy conservation behaviors based on theory of planned behavior . jour...

    2021

  6. [14]

    , author Bol \' bar, M

    author Mart \' , J. , author Bol \' bar, M. , author Lozares, C. , year 2017 . title Network cohesion and social support . journal Social networks volume 48 , pages 192--201

    2017

  7. [15]

    , year 1978

    author Mundlak, Y. , year 1978 . title On the pooling of time series and cross section data . journal Econometrica: journal of the Econometric Society , pages 69--85

    1978

  8. [16]

    title Chatgpt

    author OpenAI , . title Chatgpt . howpublished https://chatgpt.com/ . note Accessed: Jul 1, 2025

    2025

  9. [17]

    , author O'Brien, J

    author Park, J.S. , author O'Brien, J. , author Cai, C.J. , author Morris, M.R. , author Liang, P. , author Bernstein, M.S. , year 2023 . title Generative agents: Interactive simulacra of human behavior , in: booktitle Proceedings of the 36th annual acm symposium on user inter...

    2023

  10. [18]

    , author Yilmaz, T

    author Rende, A.N.N. , author Yilmaz, T. , author Ulusoy, O. , year 2025 . title Crowd: A social network simulation framework . journal IEEE Transactions on Computational Social Systems

    2025

  11. [19]

    , author Goodie, A.S

    author Selden, M. , author Goodie, A.S. , year 2018 . title Review of the effects of five factor model personality traits on network structures and perceptions of structure . journal Social Networks volume 52 , pages 81--99

    2018

  12. [20]

    , author Cuadrado, N.M

    author Takrouri, M. , author Cuadrado, N.M. , author Tak \'a c , M. , year 2025 . title Knowledge distillation from large language models for household energy modeling . journal arXiv preprint arXiv:2502.03034

    2025 arXiv

  13. [21]

    Department of Energy , year 2025

    author U.S. Department of Energy , year 2025 . title Energyplus . howpublished https://energyplus.net

    2025

  14. [22]

    , author Ma, C

    author Wang, L. , author Ma, C. , author Feng, X. , author Zhang, Z. , author Yang, H. , author Zhang, J. , author Chen, Z. , author Tang, J. , author Chen, X. , author Lin, Y. , et al., year 2024 . title A survey on large language model based autonomous agents . journal Front...

    2024

  15. [23]

    , author Ren, Y.T

    author Wang, Q.C. , author Ren, Y.T. , author Liu, X. , author Chang, R.D. , author Zuo, J. , year 2023 . title Exploring the heterogeneity in drivers of energy-saving behaviours among hotel guests: Insights from the theory of planned behaviour and personality profiles . journ...

    2023

  16. [24]

    , author Hosseinichimeh, N

    author Williams, R. , author Hosseinichimeh, N. , author Majumdar, A. , author Ghaffarzadegan, N. , year 2023 . title Epidemic modeling with generative agents . journal arXiv preprint arXiv:2307.04986

    2023 arXiv

  17. [25]

    , author Chen, H

    author Yang, M. , author Chen, H. , author Long, R. , author Yang, J. , year 2022 . title The impact of different regulation policies on promoting green consumption behavior based on social network modeling . journal Sustainable Production and Consumption volume 32 , pages 468--478

    2022

  18. [26]

    , author Tanaka, F

    author Yonekura, H. , author Tanaka, F. , author Mizumoto, T. , author Yamaguchi, H. , year 2024 . title Generating human daily activities with llm for smart home simulator agents , in: booktitle 2024 International Conference on Intelligent Environments (IE) , organization IEE...

    2024

  19. [27]

    , year 1977

    author Zachary, W.W. , year 1977 . title An information flow model for conflict and fission in small groups . journal Journal of anthropological research volume 33 , pages 452--473

    1977

  20. [28]

    , author Woo, S.E

    author Zhu, X. , author Woo, S.E. , author Porter, C. , author Brzezinski, M. , year 2013 . title Pathways to happiness: From personality to social networks and perceived support . journal Social networks volume 35 , pages 382--393

    2013

  21. [29]

    2015, , 579, A101

    Aladro, R., Martín, S., Riquelme, D., et al. 2015, , 579, A101

    2015

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.