Pith. sign in

REVIEW 5 major objections 5 minor 66 references

TravelAgent: Generative Agents in the Built Environment

T0 review · 5 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read The paper claims that embedding LLM-driven agents in 3D environments with first-person sensory inputs makes them navigate, adapt, and fail in ways that reveal how legible an urban space is.

desk verdict The platform integration is new and the limitations section is honest, but the 76% completion rate is confounded by explicit route and sensor cues, so the human-like behavior claim is not supported. read the letter →

arxiv 2412.18985 v1 pith:LRW57TPO submitted 2024-12-25 cs.AI cs.HC

Ariel Noyman , Kai Hu , Kent Larson This is my paper · ORCID
classification cs.AIcs.HC
keywords UrbanPlanningArchitecturalDesignHumanBehaviorGenerativeAgentsSimulationAgent-BasedModelingChain-of-ThoughtWayfinding
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

TravelAgent is a simulation platform that places large-language-model 'generative agents' inside a 3D environment and gives them first-person images, ray-cast distances, a discovery map, and a compass so that they must decide each step rather than follow a route. The paper's central claim is that these agents perceive, plan, and adapt in ways that resemble human pedestrian decision-making, and that the resulting logs can be read as feedback about how legible a space is. To support this, the authors ran 100 simulations of a commute-to-subway task totaling 1,898 steps and report a 76% task-completion rate, alongside spatial, term-frequency, topic, and sentiment analyses of the agents' thoughts and observations. If the claim holds, the tool gives urban designers a way to test proposed layouts for wayfinding clarity before construction, by watching simulated pedestrians think aloud.

What carries the argument

The engine is the TravelAgent platform's closed loop of perception and reasoning. A rudimentary 3D model, with semantically segmented objects, is rendered by a class-guided diffusion model (SDXL) into first-person street images; those images, along with ray-cast collision warnings, a top-down Discovery Map, an optional compass, and a compressed spatial-memory string, become the sensory inputs. Each step runs a Chain-of-Thought (CoT) reasoning process that produces an observation, a plan, a memory update, and an action chosen from a small command set ('move forward', 'turn', 'finish') with a distance or angle. This loop is what carries the argument: the agent's behavior is claimed to emerge from perceiving and remembering, not from navigation algorithms.

What would settle it

Run the same 'Subway Station' task with human participants in a matched real or virtual environment and compare their paths, decision points, and verbal descriptions with the agents' logs; if humans do not show the night-scenario failures, winter-scenario consistency, or the sentiment-to-outcome correlation the agents show, the platform's claim to simulate human-like wayfinding fails.

Watch

Extended reading notes

Core claim

In its own terms, the paper's discovery is that a Chain-of-Thought LLM agent, equipped with multimodal sensory inputs and a textual spatial memory, can carry out everyday navigation tasks in a 3D scene without explicit pathfinding, and that its verbal and spatial traces reveal where a design supports or undermines wayfinding. The headline evidence is the 'Subway Station' experiment: 100 agent runs across scenarios (base, winter, Tokyo, night, persona) produced 1,898 steps and a 76% completion rate; agents that reached the station then handled secondary tasks such as finding coffee by reasoning from landmarks, such as inferring that a plaza near the station is likely to contain cafes. The paper also argues that failed runs are not noise: clusters of 'search' actions and negative sentiment in failed paths can indicate confusing layouts, so the simulation converts design illegibility into observable agent behavior.

Load-bearing premise

The load-bearing assumption is that an LLM's verbal reasoning about generated images, labels, and a two-dimensional memory map behaves like a real pedestrian's perception and wayfinding; the paper itself warns in Section 6.2 that the agent's seemingly natural actions must be distinguished from actual human behavior and calls for extensive validation.

Editorial extensions

If this is right

  • Designers could test alternative layouts quickly and get agent logs that name the confusing element ('a building blocks the right side') rather than just aggregate counts.
  • The 76% completion rate and the distribution of decision points provide a quantitative baseline for comparing design iterations in the same scene.
  • Failed runs become design evidence: repeated search actions and negative sentiment mark locations where a layout is illegible.
  • The same platform can be re-configured for different times of day, seasons, cities, and personas by editing the prompt and regenerating images, so scenarios scale without re-training.
  • Agents that handle open-ended subtasks after reaching their goal demonstrate that the approach can model adaptation to new information, not only scripted routes.

Reading between the lines

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

  • Editorial inference: because the Discovery Map and compass provide spatial information the agent could not get from first-person vision alone, ablating these two inputs would reveal how much of the 76% completion rate is due to the LLM's reasoning versus the built-in spatial crutches.
  • Editorial inference: the diffusion-generated images are not guaranteed temporally consistent across steps, so the visual stream may be plausible but incoherent; a controlled comparison using real image sequences, or removing the ray-cast labels, would separate how much the agent relies on visuals versus textual scene descriptions.
  • Editorial inference: the platform could be calibrated against human wayfinding data before being used for design decisions, effectively tuning the agent's priors to a specific population rather than to an LLM's generic urban stereotypes.
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

5 major / 5 minor

Summary. The paper presents TravelAgent, a platform that couples generative LLM agents with 3D urban environments, diffusion-generated street imagery, ray-cast depth/collision cues, a compass, and a discovery map. Agents use Chain-of-Thought to produce observations, plans, memories, and actions. The authors report 100 simulations of a 'Train Station' navigation task with 1,898 agent steps and a 76% completion rate, and supplement this with spatial path analysis, topic modeling, and sentiment analysis. They claim the platform enables human-like decision-making and can inform urban design evaluation.

Significance. If the platform's outputs were shown to track human wayfinding behavior, TravelAgent would be a valuable low-cost tool for design evaluation, generating rich multimodal logs. The paper contributes a concrete system and a substantial dataset, and it honestly lists validation as a limitation. However, the headline quantitative result is not yet a measure of human-like behavior: success is self-declared, the route is provided in the prompt, and no baseline or uncertainty analysis is given. As it stands, the evidence supports claims about LLM prompt-following in a simulated environment, not about human spatial cognition.

major comments (5)
  1. [§4.2.2 and Appendix C (step 11)] The success metric is circular. The initiation prompt tells the agent exactly how to reach the station ('proceed down the street, then turn left, and it will be on your left'), and the simulator explicitly announces the station's distance and instructs the agent to reply 'finish' and stop. An agent that obeys the prompt and the warning will be scored as successful regardless of whether it exhibits spatial reasoning. The 76% completion rate therefore does not isolate the generative agent's contribution. Please provide a control arm (e.g., a reactive agent that simply follows the compass and ray-cast warnings) and an objective success criterion (e.g., distance to goal at simulation end).
  2. [§3.2 vs. §4.2] The descriptions of sensory inputs are contradictory. §3.2 states that inputs are designed 'without the usage of navigation algorithms or top-down maps,' yet the Discovery Map is a top-down allocentric map, and the Compass provides a bearing to the target. §4.2 states that agents 'were provided with no maps,' but §3.2 and Appendix C show they received the Discovery Map and Compass. The paper must specify which sensory conditions were used in the main experiment and remove the contradiction.
  3. [§5.5 vs. Abstract/§1.2] The behavioral results undermine the central claim of adaptation. §5.5 reports that agents 'consistently followed the main road, indicating a deterministic decision-making process' and that behavior 'may be influenced by its initial conditions, prompting reliance on prior knowledge and the navigational cues (discovery map and compass), rather than adapting to new opportunities.' This is in tension with the abstract's claim of 'human-like decision-making, behavior, and adaptation.' The manuscript should either present evidence of adaptation or qualify the claim accordingly.
  4. [§5.2] The 76% completion rate is reported without statistical uncertainty or per-condition breakdown. With 100 simulations across scenarios and personas, the paper should report confidence intervals and per-cell completion rates (e.g., Night vs. Winter) to support the qualitative claims in §5.3. As it stands, the differences discussed could be within sampling noise.
  5. [§6.2] The paper concedes that 'extensive validation is necessary' and that 'it is crucial to distinguish between the agent's seemingly natural actions and actual human behavior.' No human-subject comparison or validation against real wayfinding data is provided. Given the abstract asserts human-like behavior, this validation gap is load-bearing for the central claim and should be addressed, or the claim should be scaled back to what the evidence supports.
minor comments (5)
  1. [Abstract/§1.2] The completion rate is reported as 76% in the abstract and 'approximately 75%' in §1.2; please reconcile these numbers.
  2. [Throughout] Several typos and usage errors: 'aquatinted' (§4.2.2), 'wether' (Appendix A), 'preform' (§1.3), 'verity' (§3.2), and 'asses' (§6.1).
  3. [§3.1 and §4.2.1] References to 'Table 6.3' are confusing because the appendix table appears to be labeled Table 2; please fix the cross-references.
  4. [Figure 5 caption and §5.2] Success is described as declaring 'stop' in the Figure 5 caption but as replying 'finish' in the text and Appendix C; please use consistent terminology.
  5. [§5.2] The statement that 'reaching the goal (i.e., finding the subway station) was not the main objective' sits uneasily with the abstract's emphasis on the 76% completion rate; clarify the role of the completion rate in evaluating the platform.

Circularity Check

1 steps flagged · score 6.0 of 10

The 76% completion rate is an instruction-following score rather than independent evidence of human-like wayfinding: the route is dictated in the init prompt, and success is the agent echoing the simulator's 'finish' cue.

  1. self definitional [Section 4.2.2 (Agent Initiation); Appendix C, Step 11; Figure 5 caption; Section 4.2]
    "To reach the station, proceed down the street, then turn left, and it will be on your left. Look for a large 'Subway' sign. ... Warning! your can't move forward.There is a/an subway station in your forward in 1.41 m.Congratulations! You have finished the task! Reply 'finish' and a value of '1' to stop the experiment. ... If the agent reaches and recognizes the subway station by declaring 'stop', the path is considered successful."

    The paper's success metric is the agent declaring completion, and the simulator's warning explicitly instructs the agent to reply 'finish' at the goal. The path to that goal is supplied verbatim in the initiation prompt ('proceed down the street, then turn left...'), while compass and ray-cast labels provide exact angles and distances (e.g., 'the target is to my right at an angle of 7 degrees'; 'subway station in your forward in 1.41 m'). Hence the reported 76% task completion rate is, by construction, a measure of whether the LLM obeys the provided route and the stop cue; it does not isolate spatial reasoning, memory, or adaptation.

full rationale

The main circularity is in the evaluation protocol, not in the platform construction. The headline quantitative claim—76% task completion across 100 simulations—is operationally defined as the agent emitting a 'finish' action after the system has already told it 'You have finished the task! Reply finish', and the path to the station is dictated in the init prompt. This makes the completion rate a prompt-following score, not an independent test of human-like navigation or adaptation. I did not count the scenario-level spatial, term-frequency, or sentiment analyses as separately circular, because they are descriptive of the generated logs rather than derived from a fitted parameter; however, they inherit the validity problem: the text being analyzed was produced under instructions that already contain the route, target, and success cue. No load-bearing self-citation was found: the authors' own prior works (Noyman 2022, Grignard et al. 2018) appear only as background, and no uniqueness theorem or ansatz is imported from those works. The paper's own limitations section explicitly disclaims equivalence between agent actions and human behavior, which supports the view that the central evidence is not yet a validated prediction. Overall, the central quantitative result partially reduces by construction, but the platform, logs, and qualitative analyses retain independent descriptive content, so a score of 6 rather than 8 or 10 is appropriate.

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

The platform's outputs rest on hand-chosen simulator parameters and unvalidated assumptions that LLM text and synthetic images stand in for real human perception. The paper is honest about many of these in Section 6.2, but none are independently verified, so the ledger is dominated by domain assumptions rather than fitted constants. No new physical or theoretical entities are introduced.

free parameters (4)
  • Allowed simulation step count = not reported
    Each agent is given a number of allowed steps that bounds the task; the value is hand-chosen per experiment and directly affects the 76% completion rate.
  • Success threshold for 'finish' = distance to station not systematically reported; example warnings show 0.82 to 1.41 meters
    Task success is the agent's own declaration of 'finish' when close to the sign, a hand-defined criterion rather than an objective arrival measure.
  • Action step lengths = examples: 15, 20, 30, 40 meters
    Movement magnitudes are generated by the LLM and vary by step, so path lengths and search behavior depend on unconstrained choices.
  • Scenario matrix cell counts = not specified
    The 100-run matrix balances scenarios and personas by hand; the exact replication counts per cell are not given.
assumptions (5)
  • domain assumption LLM-generated text and actions are a scientifically meaningful proxy for human navigation behavior.
    Used throughout Sections 4 and 5 to interpret agent logs as evidence about wayfinding, legibility, and user experience; explicitly unvalidated in Section 6.2.
  • domain assumption Diffusion-generated street-level images preserve the spatial layout and object semantics of the 3D model.
    Section 3.1 says SDXL generates realistic images from semantically segmented 3D models; no consistency or fidelity check is reported, and Section 6.2 admits images 'lack consistency across steps.'
  • domain assumption Ray-cast class labels and distances are sufficient for collision-aware navigation.
    Section 3.2 replaces dense depth estimation with ray-casting; the paper does not evaluate whether this sparse signal supports human-like decisions.
  • domain assumption No additional environmental dynamics such as traffic, social interactions, or changing obstacles are needed for the claims.
    The simulator permits only simple movements and static scenes; Section 6.2 lists this as a limitation.
  • domain assumption Standard NLP tools applied to generated text capture meaningful cognitive states.
    Sections 5.4 through 5.6 interpret sentiment and topic clusters as evidence of agent confusion or environmental legibility.

how reviews work

0 comments
Cite this review

Pith. "Pith review of TravelAgent: Generative Agents in the Built Environment." pith.science (2026). https://pith.science/paper/LRW57TPO

@misc{pith2026241218985,
  author       = {Pith},
  title        = {Pith review of: TravelAgent: Generative Agents in the Built Environment},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/LRW57TPO}},
  note         = {Machine review of arXiv:2412.18985}
}
read the original abstract

Understanding human behavior in built environments is critical for designing functional, user centered urban spaces. Traditional approaches, such as manual observations, surveys, and simplified simulations, often fail to capture the complexity and dynamics of real world behavior. To address these limitations, we introduce TravelAgent, a novel simulation platform that models pedestrian navigation and activity patterns across diverse indoor and outdoor environments under varying contextual and environmental conditions. TravelAgent leverages generative agents integrated into 3D virtual environments, enabling agents to process multimodal sensory inputs and exhibit human-like decision-making, behavior, and adaptation. Through experiments, including navigation, wayfinding, and free exploration, we analyze data from 100 simulations comprising 1898 agent steps across diverse spatial layouts and agent archetypes, achieving an overall task completion rate of 76%. Using spatial, linguistic, and sentiment analyses, we show how agents perceive, adapt to, or struggle with their surroundings and assigned tasks. Our findings highlight the potential of TravelAgent as a tool for urban design, spatial cognition research, and agent-based modeling. We discuss key challenges and opportunities in deploying generative agents for the evaluation and refinement of spatial designs, proposing TravelAgent as a new paradigm for simulating and understanding human experiences in built environments.

Figures

Figures reproduced from arXiv: 2412.18985 by the authors.

Figure 1
Figure 1. A schematic representation of the TravelAgent system. (left) TAs are initialized with various parameters [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. TravelAgent interface. The web app provides an end-to-end experimentation environment for testing and [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Pedestrian-level image generation from a 3D model. The SDXL-Turbo image generation model (middle) uses [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Visual Perception in ‘Lunch Break’ Experiment. The agent’s visual perception is guided by Google Street [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 5
Figure 5. Figure 5: Spatial Analysis of ‘Train Station’ Experiment. For each scenario, we evaluate the agent’s successful and [PITH_FULL_IMAGE:figures/full_fig_p009_5.png]
Figure 6
Figure 6. Figure 6: Analysis of the 200 most frequent words in the agents’ planning streams, clustered by scenario. The dominant [PITH_FULL_IMAGE:figures/full_fig_p010_6.png]
Figure 7
Figure 7. Figure 7: Topical modeling of agent observations and plans across scenarios. Here, key terms in the agent’s observations [PITH_FULL_IMAGE:figures/full_fig_p011_7.png]
Figure 8
Figure 8. Figure 8: Sentiment analysis of the agent’s ‘thoughts’ and ‘observations’ streams. Each experiment is displayed as [PITH_FULL_IMAGE:figures/full_fig_p012_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

66 extracted references · 45 canonical work pages

  1. [1]

    Cities for People

    Jan Gehl. Cities for People. Island Press, 2010

    work page 2010

  2. [2]

    The death and life of great American cities

    Jane Jacobs. The death and life of great American cities. Random House LLC, 1961

    work page 1961

  3. [3]

    William H. Whyte. The Social Life of Small Urban Spaces. Project for Public Spaces, 1980

    work page 1980

  4. [4]

    The Social Logic of Space

    Bill Hillier and Julienne Hanson. The Social Logic of Space. Cambridge University Press, 1984

    work page 1984

  5. [5]

    Visualizing the city: communicating urban design to planners and decision-makers

    Michael Batty, David Chapman, Steve Evans, Mordechai Haklay, Stefan Kueppers, Naru Shiode, Andy Smith, and Paul M Torrens. Visualizing the city: communicating urban design to planners and decision-makers. 2000

    work page 2000

  6. [6]

    A digital future for planning: Spatial planning reimagined

    Michael Batty and Wei Yang. A digital future for planning: Spatial planning reimagined. Digital Task Force for Planning, 2022

    work page 2022

  7. [7]

    CityScope : An Urban Modeling and Simulation Platform

    Ariel Noyman. CityScope : An Urban Modeling and Simulation Platform . PhD thesis, Massachusetts Institute of Technology, Cambridge, Massachusetts, 2022

    2022

  8. [8]

    Companion to urban design

    Tridib Banerjee and Anastasia Loukaitou-Sideris. Companion to urban design. Routledge, 2011

    work page 2011

Show all 66 references
  1. [9]

    Perez, A

    P. Perez, A. Banos, and C. Pettit. Agent-based modelling for urban planning: Current limitations and future trends. In Planning Support Science for Smarter Urban Futures, pages 60--69. Springer, 2016. doi:10.1007/978-3-319-51957-9_4

    2016 doi

  2. [10]

    Agent-based computational models and generative social science

    Joshua M Epstein. Agent-based computational models and generative social science. Complexity, 4 0 (5): 0 41--60, 1999

    1999

  3. [11]

    Multiple models

    Michael Batty. Multiple models. Environment and Planning B: Urban Analytics and City Science, 48 0 (8): 0 2129--2132, 2021. ISSN 2399-8083, 2399-8091. doi:10.1177/23998083211051139

    2021 doi

  4. [12]

    On the limits of agency in agent-based models, October 2024

    Ayush Chopra, Shashank Kumar, Nurullah Giray-Kuru , Ramesh Raskar, and Arnau Quera-Bofarull . On the limits of agency in agent-based models, October 2024

    2024

  5. [13]

    A survey on deep learning for human mobility

    Massimiliano Luca, Gianni Barlacchi, Bruno Lepri, and Luca Pappalardo. A survey on deep learning for human mobility. ACM Computing Surveys (CSUR), 55 0 (1): 0 1--44, 2021

    2021

  6. [14]

    Generative agents: Interactive simulacra of human behavior

    Joon Sung Park, Joseph O'Brien, Carrie Jun Cai, Meredith Ringel Morris, Percy Liang, and Michael S Bernstein. Generative agents: Interactive simulacra of human behavior. In Proceedings of the 36th annual acm symposium on user interface software and technology, pages 1--22, 2023

    2023

  7. [15]

    Lyfe Agents : Generative agents for low-cost real-time social interactions, 2023

    Zhao Kaiya, Michelangelo Naim, Jovana Kondic, Manuel Cortes, Jiaxin Ge, Shuying Luo, Guangyu Robert Yang, and Andrew Ahn. Lyfe Agents : Generative agents for low-cost real-time social interactions, 2023

    2023

  8. [16]

    Chain-of- Thought Prompting Elicits Reasoning in Large Language Models , January 2023

    Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Brian Ichter, Fei Xia, Ed Chi, Quoc Le, and Denny Zhou. Chain-of- Thought Prompting Elicits Reasoning in Large Language Models , January 2023

    2023

  9. [17]

    Agent-based modeling: Methods and techniques for simulating human systems

    Eric Bonabeau. Agent-based modeling: Methods and techniques for simulating human systems. Proceedings of the National Academy of Sciences, 99 0 (suppl\_3): 0 7280--7287, May 2002. ISSN 0027-8424, 1091-6490. doi:10.1073/pnas.082080899

    2002 doi

  10. [18]

    Agent-based modeling in urban and architectural research: A brief literature review

    Liang Chen. Agent-based modeling in urban and architectural research: A brief literature review. Frontiers of Architectural Research, 1 0 (2): 0 166--177, 2012. ISSN 2095-2635. doi:10.1016/j.foar.2012.03.003

    2012 doi

  11. [19]

    The image of the city

    Kevin Lynch. The image of the city. MIT press, 1960

    1960

  12. [20]

    S. H. Ameli, Shima Hamidi, Andrea Garfinkel-Castro, and R. Ewing. Do better urban design qualities lead to more walking in salt lake city, utah? Journal of Urban Design, 20: 0 393 -- 410, 2015. doi:10.1080/13574809.2015.1041894

    2015

  13. [21]

    In search of visualization challenges: The development and implementation of visualization tools for supporting dialogue in urban planning processes

    Monica Billger, Liane Thuvander, and Beata Stahre W \"a stberg. In search of visualization challenges: The development and implementation of visualization tools for supporting dialogue in urban planning processes. Environment and Planning B: Urban Analytics and City Science, 4...

    2017

  14. [22]

    L. Yin. Street level urban design qualities for walkability: Combining 2d and 3d gis measures. Comput. Environ. Urban Syst., 64: 0 288--296, 2017. doi:10.1016/j.compenvurbsys.2017.04.001

    2017 doi

  15. [23]

    Qiaomu Shen, W. Zeng, Y. Ye, S. Arisona, S. Schubiger-Banz, R. Burkhard, and Huamin Qu. Streetvizor: Visual exploration of human-scale urban forms based on street views. IEEE Transactions on Visualization and Computer Graphics, 24: 0 1004--1013, 2018. doi:10.1109/TVCG.2017.2744159

    2018

  16. [24]

    Visual communication in urban planning and urban design

    Andy Smith, Martin Dodge, and Simon Doyle. Visual communication in urban planning and urban design. University College London, Centre for Advanced Spatial Analysis (CASA), 1998

    1998

  17. [25]

    3d visibility analysis in urban environment - cognition research based on vge

    Tianpeng Lin, Hui Lin, and Mingyuan Hu. 3d visibility analysis in urban environment - cognition research based on vge. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, pages 227--236, 2013. doi:10.5194/ISPRSANNALS-II-2-W1-227-2013

    2013 doi

  18. [26]

    Petrasova, Perver K

    Payam Tabrizian, A. Petrasova, Perver K. Baran, Jelena Vukomanovic, H. Mitásová, and R. Meentemeyer. High resolution viewscape modeling evaluated through immersive virtual environments. ISPRS Int. J. Geo Inf., 9: 0 445, 2020. doi:10.3390/ijgi9070445

    2020 doi

  19. [27]

    Informal interactions in 3d education: Citizenship participation and assessment of virtual urban proposals

    David Fonseca, Francesc Valls, Ernest Redondo, and Sergi Villagrasa. Informal interactions in 3d education: Citizenship participation and assessment of virtual urban proposals. Computers in Human Behavior, 55: 0 504--518, 2016

    2016

  20. [28]

    u bel, Koen Steemers, Kerstin Sailer, Lola Ben-Alon, Christoph H \

    Michal Gath-Morad, Jascha Gr \"u bel, Koen Steemers, Kerstin Sailer, Lola Ben-Alon, Christoph H \"o lscher, and Leonel Aguilar. The role of strategic visibility in shaping wayfinding behavior in multilevel buildings. Scientific Reports, 14 0 (1): 0 3735, 2024

    2024

  21. [29]

    Scale and multiple psychologies of space

    Daniel R Montello. Scale and multiple psychologies of space. Spatial information theory a theoretical basis for GIS, pages 312--321, 1993

    1993

  22. [30]

    Human wayfinding and cognitive maps

    Reginald G Golledge. Human wayfinding and cognitive maps. Wayfinding behavior: Cognitive mapping and other spatial processes, pages 5--45, 1999

    1999

  23. [31]

    Hippocampal and prefrontal processing of network topology to simulate the future

    Amir-Homayoun Javadi, Beata Emo, Luke R Howard, and et al. Hippocampal and prefrontal processing of network topology to simulate the future. Nature Communications, 8 0 (1): 0 1--11, 2017. doi:10.1038/s41467-017-00112-0

    2017 doi

  24. [32]

    The cognitive map in humans: Spatial navigation and beyond

    Russell A Epstein, Eva Z Patai, Joshua B Julian, and Hugo J Spiers. The cognitive map in humans: Spatial navigation and beyond. Nature neuroscience, 20 0 (11): 0 1504--1513, 2017. doi:10.1038/nn.4656

    2017 doi

  25. [33]

    Spatial memory: how egocentric and allocentric combine

    Neil Burgess. Spatial memory: how egocentric and allocentric combine. Trends in cognitive sciences, 10 0 (12): 0 551--557, 2006

    2006

  26. [34]

    Allocentric and egocentric spatial representations: Definitions, distinctions, and interconnections

    Roberta L Klatzky. Allocentric and egocentric spatial representations: Definitions, distinctions, and interconnections. Spatial cognition, pages 1--17, 2008

    2008

  27. [35]

    The roles of egocentric and allocentric spatial representations in locomotion and reorientation

    Weimin Mou, Timothy P McNamara, Bj "o rn Rump, and Chengli Xiao. The roles of egocentric and allocentric spatial representations in locomotion and reorientation. Journal of Experimental Psychology: Learning, Memory, and Cognition, 30 0 (6): 0 1322, 2004

    2004

  28. [36]

    The cognitive map in humans: spatial navigation and beyond

    Neil Burgess, Russell A Epstein, and Hugo J Spiers. The cognitive map in humans: spatial navigation and beyond. Nature neuroscience, 20 0 (11): 0 1504--1513, 2017

    2017

  29. [37]

    Mnemonic networks in the hippocampal formation: From spatial maps to temporal and conceptual codes

    Branka Milivojevic and Christian F Doeller. Mnemonic networks in the hippocampal formation: From spatial maps to temporal and conceptual codes. Journal of Experimental Psychology: General, 142 0 (4): 0 1231, 2013

    2013

  30. [38]

    Steven C. Bankes. Agent-based modeling: A revolution? Proceedings of the National Academy of Sciences, 99 0 (suppl\_3): 0 7199--7200, May 2002. ISSN 0027-8424, 1091-6490. doi:10.1073/pnas.072081299

    2002 doi

  31. [39]

    Powers, Neil Urquhart, Thomas Farrenkopf, and Michael Guckert

    Johannes Nguyen, Simon T. Powers, Neil Urquhart, Thomas Farrenkopf, and Michael Guckert. An overview of agent-based traffic simulators. Transportation Research Interdisciplinary Perspectives, 12: 0 100486, December 2021. ISSN 25901982. doi:10.1016/j.trip.2021.100486

    2021

  32. [40]

    Cityscope andorra: A multi-level interactive and tangible agent-based visualization

    Arnaud Grignard, N \'u ria Maci \`a , Luis Alonso Pastor, Ariel Noyman, Yan Zhang, and Kent Larson. Cityscope andorra: A multi-level interactive and tangible agent-based visualization. In AAMAS, pages 1939--1940, 2018

    1939

  33. [41]

    Verstegen

    Gabriele Filomena and Judith A. Verstegen. Modelling the effect of landmarks on pedestrian dynamics in urban environments. Computers, Environment and Urban Systems, 86: 0 101573, March 2021. ISSN 01989715. doi:10.1016/j.compenvurbsys.2020.101573

    2021

  34. [42]

    Agent-based and individual-based modeling: a practical introduction

    Steven F Railsback and Volker Grimm. Agent-based and individual-based modeling: a practical introduction. Princeton university press, 2019

    2019

  35. [43]

    On generative agents in recommendation

    An Zhang, Yuxin Chen, Leheng Sheng, Xiang Wang, and Tat-Seng Chua. On generative agents in recommendation. In Proceedings of the 47th international ACM SIGIR conference on research and development in Information Retrieval, pages 1807--1817, 2024

    2024

  36. [44]

    Modeling and generating human mobility trajectories using transformer with day encoding

    Akihiro Kobayashi, Naoto Takeda, Yudai Yamazaki, and Daisuke Kamisaka. Modeling and generating human mobility trajectories using transformer with day encoding. In Proceedings of the 1st International Workshop on the Human Mobility Prediction Challenge, pages 7--10, 2023

    2023

  37. [45]

    An agent-based learning-embedded model (abm-learning) for urban land use planning: A case study of residential land growth simulation in shenzhen, china

    Feixue Li, Zhifeng Li, Honghua Chen, Zhenjie Chen, and Manchun Li. An agent-based learning-embedded model (abm-learning) for urban land use planning: A case study of residential land growth simulation in shenzhen, china. Land Use Policy, 95: 0 104620, 2020

    2020

  38. [46]

    Touchdown: Natural language navigation and spatial reasoning in visual street environments

    Howard Chen, Alane Suhr, Dipendra Misra, Noah Snavely, and Yoav Artzi. Touchdown: Natural language navigation and spatial reasoning in visual street environments. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 12538--12547, 2019

    2019

  39. [47]

    Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context

    Machel Reid, Nikolay Savinov, Denis Teplyashin, Dmitry Lepikhin, Timothy Lillicrap, Jean-baptiste Alayrac, Radu Soricut, Angeliki Lazaridou, Orhan Firat, Julian Schrittwieser, et al. Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context. arXiv pre...

    2024 arXiv

  40. [48]

    V-irl: Grounding virtual intelligence in real life

    Jihan Yang, Runyu Ding, Ellis Brown, Xiaojuan Qi, and Saining Xie. V-irl: Grounding virtual intelligence in real life. In European Conference on Computer Vision, pages 36--55. Springer, 2025

    2025

  41. [49]

    Understanding the planning of LLM agents: A survey, February 2024

    Xu Huang, Weiwen Liu, Xiaolong Chen, Xingmei Wang, Hao Wang, Defu Lian, Yasheng Wang, Ruiming Tang, and Enhong Chen. Understanding the planning of LLM agents: A survey, February 2024

    2024

  42. [50]

    ReAct : Synergizing Reasoning and Acting in Language Models , March 2023

    Shunyu Yao, Jeffrey Zhao, Dian Yu, Nan Du, Izhak Shafran, Karthik Narasimhan, and Yuan Cao. ReAct : Synergizing Reasoning and Acting in Language Models , March 2023

    2023

  43. [51]

    Tree of Thoughts : Deliberate Problem Solving with Large Language Models

    Shunyu Yao, Dian Yu, Jeffrey Zhao, Izhak Shafran, Thomas L Griffiths, Yuan Cao, and Karthik Narasimhan. Tree of Thoughts : Deliberate Problem Solving with Large Language Models

  44. [52]

    Generative agent-based modeling: Unveiling social system dynamics through coupling mechanistic models with generative artificial intelligence

    Navid Ghaffarzadegan, Aritra Majumdar, Ross Williams, and Niyousha Hosseinichimeh. Generative agent-based modeling: Unveiling social system dynamics through coupling mechanistic models with generative artificial intelligence. arXiv preprint arXiv:2309.11456, 2023

    2023 arXiv

  45. [53]

    Memory Matters : The Need to Improve Long-Term Memory in LLM-Agents

    Kostas Hatalis, Despina Christou, Joshua Myers, Steven Jones, Keith Lambert, Adam Amos-Binks , Zohreh Dannenhauer, and Dustin Dannenhauer. Memory Matters : The Need to Improve Long-Term Memory in LLM-Agents . Proceedings of the AAAI Symposium Series, 2 0 (1): 0 277--280, Janua...

    2024 doi

  46. [54]

    GSM-Symbolic : Understanding the Limitations of Mathematical Reasoning in Large Language Models , October 2024

    Iman Mirzadeh, Keivan Alizadeh, Hooman Shahrokhi, Oncel Tuzel, Samy Bengio, and Mehrdad Farajtabar. GSM-Symbolic : Understanding the Limitations of Mathematical Reasoning in Large Language Models , October 2024

    2024

  47. [55]

    Sdxl: Improving latent diffusion models for high-resolution image synthesis

    Dustin Podell, Zion English, Kyle Lacey, Andreas Blattmann, Tim Dockhorn, Jonas M \"u ller, Joe Penna, and Robin Rombach. Sdxl: Improving latent diffusion models for high-resolution image synthesis. arXiv preprint arXiv:2307.01952, 2023

    2023 arXiv

  48. [56]

    You only look once: unified, real-time object detection (2015)

    Joseph Redmon, Santosh Divvala, Ross Girshick, and Ali Farhadi. You only look once: unified, real-time object detection (2015). arXiv preprint arXiv:1506.02640, 825, 2015

    2015 arXiv

  49. [57]

    Gpt-4 technical report

    Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, et al. Gpt-4 technical report. arXiv preprint arXiv:2303.08774, 2023

    2023 arXiv

  50. [58]

    Masked-attention mask transformer for universal image segmentation

    B Cheng, I Misra, AG Schwing, A Kirillov, and R Girdhar. Masked-attention mask transformer for universal image segmentation. arxiv 2022. arXiv preprint arXiv:2112.01527, 2021

    2022 arXiv

  51. [59]

    Spatial Configuration, Spatial Cognition and Spatial Behaviour: the role of architectural intelligibility in shaping spatial experience

    Young Ook Kim. Spatial Configuration, Spatial Cognition and Spatial Behaviour: the role of architectural intelligibility in shaping spatial experience. University of London, University College London (United Kingdom), 1999

    1999

  52. [60]

    Data mining: concepts, models, methods, and algorithms

    Mehmed Kantardzic. Data mining: concepts, models, methods, and algorithms. John Wiley & Sons, 2011

    2011

  53. [61]

    Text clustering with large language model embeddings

    Alina Petukhova, Jo \ a o P Matos-Carvalho, and Nuno Fachada. Text clustering with large language model embeddings. International Journal of Cognitive Computing in Engineering, 2024

    2024

  54. [62]

    Latent dirichlet allocation

    David M Blei, Andrew Y Ng, and Michael I Jordan. Latent dirichlet allocation. Journal of machine Learning research, 3 0 (Jan): 0 993--1022, 2003

    2003

  55. [63]

    Vader: A parsimonious rule-based model for sentiment analysis of social media text

    Clayton Hutto and Eric Gilbert. Vader: A parsimonious rule-based model for sentiment analysis of social media text. In Proceedings of the international AAAI conference on web and social media, volume 8, pages 216--225, 2014

    2014

  56. [64]

    Should chatgpt be biased? challenges and risks of bias in large language models

    Emilio Ferrara. Should chatgpt be biased? challenges and risks of bias in large language models. arXiv preprint arXiv:2304.03738, 2023

    2023 arXiv

  57. [65]

    Gender bias and stereotypes in large language models

    Hadas Kotek, Rikker Dockum, and David Sun. Gender bias and stereotypes in large language models. In Proceedings of the ACM collective intelligence conference, pages 12--24, 2023

    2023

  58. [66]

    Large language models are geographically biased

    Rohin Manvi, Samar Khanna, Marshall Burke, David Lobell, and Stefano Ermon. Large language models are geographically biased. arXiv preprint arXiv:2402.02680, 2024

    2024 arXiv

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.