pith. sign in

arxiv: 2310.07099 · v1 · submitted 2023-10-11 · 💻 cs.CY · cs.AI· cs.CR· cs.SI

ClausewitzGPT Framework: A New Frontier in Theoretical Large Language Model Enhanced Information Operations

Benjamin Kereopa-Yorke This is my paper

Pith reviewed 2026-05-24 05:54 UTC · model grok-4.3

classification 💻 cs.CY cs.AIcs.CRcs.SI
keywords ClausewitzGPTLarge Language ModelsInformation OperationsRisk QuantificationAI EthicsMilitary StrategyCyberspace
0
0 comments X

The pith

Clausewitz's 1832 principles form a mathematical model to quantify risks in LLM-driven information operations.

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

The paper presents the ClausewitzGPT framework as a synthesis of classical military strategy and modern large language model capabilities. It claims this model can measure the hazards of rapid, AI-augmented efforts to shape narratives in cyberspace. The work further argues that autonomous AI agents must embed ethical rules to keep such operations aligned with broader societal needs. A reader following the argument would conclude that historical strategy offers a practical lens for managing new technological risks in geopolitical contests.

Core claim

The ClausewitzGPT equation mathematically combines Clausewitz's tenets with LLM dynamics to quantify risks in machine-speed information operations while establishing ethical autonomous AI agents as required components for responsible use.

What carries the argument

The ClausewitzGPT equation, a mathematical formulation that translates 1832 military strategy principles into a model for assessing LLM-augmented information campaigns.

If this is right

  • Nations and other actors gain a structured way to evaluate dangers in AI-enhanced narrative operations.
  • Ethical design becomes a core requirement for any autonomous agents used in such operations.
  • Strategic planning must incorporate both technological speed and moral constraints to avoid escalation.
  • Year-on-year growth in AI information campaigns makes immediate application of the model necessary.

Where Pith is reading between the lines

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

  • The same modeling approach could be tested on non-state actor campaigns to check consistency across actor types.
  • Integration with real-time monitoring tools might allow dynamic risk scoring during ongoing operations.
  • The framework implicitly raises questions about whether similar historical lenses apply to other AI domains such as autonomous weapons.

Load-bearing premise

Principles from an 1832 military text can be converted into a functional mathematical model that applies directly to today's LLM technologies.

What would settle it

Empirical data from actual LLM-based information campaigns showing that the proposed equation does not accurately predict or bound observed risks would falsify the claim.

Figures

Figures reproduced from arXiv: 2310.07099 by Benjamin Kereopa-Yorke.

Figure 1
Figure 1. Figure 1: Information Operations Components, FM-100-6, Department of the Army, 1996 [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Information Operations in the Australian Defence Force, Captain J. Malone, 2004 In this ever-evolving panorama of information operations, Artificial Intelligence (AI) emerges as the next frontier. AI is a branch of computer science that seeks to create machines capable of mimicking human cognitive functions. It's not about merely executing tasks that a computer can already do, but rather, it's about enabli… view at source ↗
Figure 3
Figure 3. Figure 3: Chinese Information Operations, APAN, 2021 However, the "The Palgrave Handbook of Malicious Use of AI and Psychological Security" (Pashentsev, 2023), offers a broader perspective, emphasising the ubiquitous threat posed by AI-enabled disinformation campaigns. By 2020, a staggering 81 countries had witnessed the use of social media platforms for spreading computational propaganda. This surge, from just 28 c… view at source ↗
Figure 5
Figure 5. Figure 5: Information capabilities of information operations and their effects, Haig, 2020 Traditional methods, which often grapple with the lag between message creation and feedback integration, find a potential solution in LLMs. Equipped with cutting￾edge analytics, these models can recalibrate their strategies in real-time, based on audience reactions, thus ensuring that campaigns consistently hit their mark. The… view at source ↗
Figure 6
Figure 6. Figure 6: Traditional Information Operation Impact Equation This equation encapsulates the core dynamics of time-honoured campaigns, accentuating the significance of operational outreach, intrinsic message quality, and the engagement metrics. The denominator embodies external challenges such as pervasive media interferences, inherent audience resistance, and other extraneous factors [PITH_FULL_IMAGE:figures/full_fi… view at source ↗
Figure 7
Figure 7. Figure 7: AI-Augmented Information Operation Impact Equation Within this construct, real-time feedback refines the adaptive quality of the message, bolstered further by AI-driven engagement metrics. The AI spectrum introduces distinct strategic advantages, such as micro-targeting, which can dramatically curtail resistance and counter media disruptions. Building upon this, with the specific linguistic prowess of LLMs… view at source ↗
Figure 8
Figure 8. Figure 8: LLM and AI-Augmented Information Operation Impact Equation LLMs, with their unmatched capability in human-like narrative crafting and fine￾tuned individual engagement, remarkably amplify the potential efficacy of information operations. Yet, the intricate dance of strategic endeavours is not solely about efficiency. The moral compass and ethical bearings hold paramount significance. Thus, when guiding LLMs… view at source ↗
Figure 9
Figure 9. Figure 9: Ethical LLM and AI-Augmented Information Operation Impact Equation This equation underscores the importance of ethically guided operations that align with societal norms and values, ensuring both the responsible and credible use of technology. The "Nation-State Comparative Measure" equation offers a holistic perspective on the relative effectiveness of AI-augmented information operations against traditiona… view at source ↗
Figure 11
Figure 11. Figure 11: LLM-augmented Information Operation with Cascade Effects Here, δ CascadeEffects symbolises the unpredictable and potentially vast repercussions of uncontrolled LLM outputs. This equation underscores the potential pitfalls when LLMs operate devoid of checks, where unforeseen ripple effects could undermine the intended impact. Manual Ethical and Strategic Input [PITH_FULL_IMAGE:figures/full_fig_p008_11.png] view at source ↗
read the original abstract

In a digital epoch where cyberspace is the emerging nexus of geopolitical contention, the melding of information operations and Large Language Models (LLMs) heralds a paradigm shift, replete with immense opportunities and intricate challenges. As tools like the Mistral 7B LLM (Mistral, 2023) democratise access to LLM capabilities (Jin et al., 2023), a vast spectrum of actors, from sovereign nations to rogue entities (Howard et al., 2023), find themselves equipped with potent narrative-shaping instruments (Goldstein et al., 2023). This paper puts forth a framework for navigating this brave new world in the "ClausewitzGPT" equation. This novel formulation not only seeks to quantify the risks inherent in machine-speed LLM-augmented operations but also underscores the vital role of autonomous AI agents (Wang, Xie, et al., 2023). These agents, embodying ethical considerations (Hendrycks et al., 2021), emerge as indispensable components (Wang, Ma, et al., 2023), ensuring that as we race forward, we do not lose sight of moral compasses and societal imperatives. Mathematically underpinned and inspired by the timeless tenets of Clausewitz's military strategy (Clausewitz, 1832), this thesis delves into the intricate dynamics of AI-augmented information operations. With references to recent findings and research (Department of State, 2023), it highlights the staggering year-on-year growth of AI information campaigns (Evgeny Pashentsev, 2023), stressing the urgency of our current juncture. The synthesis of Enlightenment thinking, and Clausewitz's principles provides a foundational lens, emphasising the imperative of clear strategic vision, ethical considerations, and holistic understanding in the face of rapid technological advancement.

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 / 2 minor

Summary. The manuscript proposes the ClausewitzGPT Framework and associated 'ClausewitzGPT equation,' described as a mathematically underpinned synthesis of Clausewitz's 1832 principles with LLM capabilities (e.g., Mistral 7B) to quantify risks in machine-speed information operations while highlighting the role of ethical autonomous AI agents.

Significance. If a concrete, derivable mathematical model with explicit mappings, variable definitions, and validation were supplied, the work could offer a novel theoretical bridge between classical military strategy and contemporary AI-driven information campaigns, providing a structured approach to risk assessment in a rapidly evolving domain.

major comments (2)
  1. [Abstract] Abstract: The central claim that the thesis is 'mathematically underpinned' and introduces the 'ClausewitzGPT equation' for quantifying risks is unsupported, as the manuscript supplies no equations, derivations, variable definitions, or formal mappings from Clausewitzian concepts (e.g., friction, center of gravity) to LLM parameters or risk metrics.
  2. [Abstract] Abstract: The assertion that autonomous AI agents 'embodying ethical considerations' are 'indispensable components' of the framework is stated without any integration into the purported equation or any mechanism for how they mitigate the quantified risks, rendering the mitigation claim unevaluable.
minor comments (2)
  1. [Abstract] Abstract: Terminology shifts between 'framework,' 'equation,' and 'thesis' without clarification; consistent use would improve readability.
  2. [Abstract] Abstract: Several citations (e.g., Wang, Xie, et al., 2023; Department of State, 2023) are invoked but their specific role in constructing or validating the framework is not explained.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the detailed and constructive feedback. We address the major comments point by point below and will undertake a major revision to strengthen the mathematical and integrative aspects of the framework.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The central claim that the thesis is 'mathematically underpinned' and introduces the 'ClausewitzGPT equation' for quantifying risks is unsupported, as the manuscript supplies no equations, derivations, variable definitions, or formal mappings from Clausewitzian concepts (e.g., friction, center of gravity) to LLM parameters or risk metrics.

    Authors: We agree that the current manuscript presents the ClausewitzGPT equation at a conceptual level without explicit equations, derivations, variable definitions, or formal mappings. In the revised version we will add a dedicated mathematical section that defines all variables, derives the equation from Clausewitzian principles (friction, center of gravity, etc.), and provides explicit mappings to LLM parameters and risk metrics. revision: yes

  2. Referee: [Abstract] Abstract: The assertion that autonomous AI agents 'embodying ethical considerations' are 'indispensable components' of the framework is stated without any integration into the purported equation or any mechanism for how they mitigate the quantified risks, rendering the mitigation claim unevaluable.

    Authors: We acknowledge that the current text does not specify how ethical autonomous agents integrate into the equation or mitigate risks. The revision will include an explicit integration mechanism, such as ethical constraint terms or weighting functions within the ClausewitzGPT equation, together with a description of how these terms reduce the quantified risk values. revision: yes

Circularity Check

1 steps flagged

ClausewitzGPT equation reduces to self-referential definition of the framework itself with no independent derivation

specific steps
  1. self definitional [Abstract]
    "This paper puts forth a framework for navigating this brave new world in the 'ClausewitzGPT' equation. This novel formulation not only seeks to quantify the risks inherent in machine-speed LLM-augmented operations but also underscores the vital role of autonomous AI agents"

    The novel formulation is defined as the framework the paper itself puts forth, yet is simultaneously claimed to quantify the risks of the operations it addresses. The quantification step is therefore equivalent to the definition by construction, with no separate derivation, equations, or external grounding exhibited.

full rationale

The paper asserts a 'mathematically underpinned' ClausewitzGPT equation that quantifies risks in LLM-augmented operations, yet the supplied text provides no equations, variable definitions, derivations, or mappings from Clausewitzian concepts to LLM parameters. The central claim therefore collapses to defining the framework as its own quantifier. This is self-definitional circularity: the 'prediction' or result (risk quantification) is identical to the input (the framework being put forth). No external benchmarks, independent math, or falsifiable content appears. The synthesis with Clausewitz (1832) is asserted by inspiration only, without shown reduction steps.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 1 invented entities

The claim rests on the untested transfer of historical military principles to LLM information operations and on an unnamed equation whose structure is not specified.

axioms (1)
  • domain assumption Clausewitz's military strategy principles from 1832 can be directly applied to quantify risks in contemporary LLM-enhanced information operations
    Invoked throughout the abstract as the inspirational and mathematical foundation.
invented entities (1)
  • ClausewitzGPT equation no independent evidence
    purpose: To quantify risks inherent in machine-speed LLM-augmented operations
    Introduced in the abstract as a novel formulation without specification or external validation.

pith-pipeline@v0.9.0 · 5872 in / 1240 out tokens · 63285 ms · 2026-05-24T05:54:42.492779+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

55 extracted references · 55 canonical work pages · 3 internal anchors

  1. [1]

    (2023, September 27)

    AI, M. (2023, September 27). Mistral 7B. Mistral.ai. https://mistral.ai/news/announcing- mistral-7b/

    work page 2023

  2. [2]

    A., & Freeman, L

    Batarseh, F. A., & Freeman, L. (2022). AI Assurance. Academic Press

    work page 2022

  3. [3]

    Bhaso Ndzendze, & Tshilidzi Marwala. (2023). Artificial Intelligence and International Relations Theories. Springer Nature

    work page 2023

  4. [4]

    Brooks, R. (2008). Shaping strategy: the civil- military politics of strategic assessment. Princeton Princeton Univ. Press

    work page 2008

  5. [5]

    Chapple, M., & Seidl, D. (2014). Cyberwarfare: information operations in a connected world. Jones & Bartlett Learning

    work page 2014

  6. [6]

    Cheng, D. (2017). Cyber dragon : inside China’s information warfare and cyber operations. Praeger, An Imprint Of Abc-Clio

    work page 2017

  7. [7]

    Clausewitz. (1832). On War. Princeton University Press

    work page

  8. [8]

    Coeckelbergh, M. (2020). AI Ethics. The MIT Press

    work page 2020

  9. [9]

    H., & Anil Hira

    Cohn, T. H., & Anil Hira. (2021). Global political economy theory and practice. New York Routledge

    work page 2021

  10. [10]

    CSIRO. (n.d.). Responsible AI Pattern Catalogue (RAIC). Software Systems. Retrieved October 8, 2023, from https://research.csiro.au/ss/science/projects/res ponsible-ai-pattern-catalogue/

    work page 2023

  11. [11]

    C., & Press, G

    Demchak, C. C., & Press, G. (2011). Wars of disruption and resilience : cybered conflict, power, and national security. University Of Georgia Press

    work page 2011

  12. [12]

    Department of State, G. E. C. (2023, September 28). GEC Special Report: How the People’s Republic of China Seeks to Reshape the Global Information Environment. United States Department of State. https://www.state.gov/gec-special-report-how- the-peoples-republic-of-china-seeks-to- reshape-the-global-information-environment/

    work page 2023

  13. [13]

    Department of the Army. (1996). Field Manual 100-6 Information Operations. Department of the Army

    work page 1996

  14. [14]

    Doro-on, A. M. (2022). Handbook of Systems Engineering and Risk Management in Control Systems, Communication, Space Technology, Missile, Security and Defense Operations. CRC Press

    work page 2022

  15. [15]

    D., Pasquale, F., & Das, S

    Dubber, M. D., Pasquale, F., & Das, S. (2020). The Oxford Handbook of Ethics of AI. Oxford University Press, Incorporated

    work page 2020

  16. [16]

    Floridi, L., Cowls, J., Beltrametti, M., Chatila, R., Chazerand, P., Dignum, V., Luetge, C., Madelin, R., Pagallo, U., Rossi, F., Schafer, B., Valcke, P., & Vayena, E. (2018). AI4People— An Ethical Framework for a Good AI Society: Opportunities, Risks, Principles, and Recommendations. Minds and Machines, 28(4), 689–707. https://doi.org/10.1007/s11023-018-9482-5

    work page doi:10.1007/s11023-018-9482-5 2018

  17. [17]

    L., & Findley, B

    Goldstein, F. L., & Findley, B. F. (1996). Psychological Operations: Principles and Case Studies. Air University Press

    work page 1996

  18. [18]

    A., Sastry, G., Musser, M., DiResta, R., Gentzel, M., & Katerina Sedova

    Goldstein, J. A., Sastry, G., Musser, M., DiResta, R., Gentzel, M., & Katerina Sedova. (2023). Generative Language Models and Automated Influence Operations: Emerging Threats and Potential Mitigations. ArXiv (Cornell University). https://doi.org/10.48550/arxiv.2301.04246

    work page doi:10.48550/arxiv.2301.04246 2023

  19. [19]

    Google AI. (n.d.). Google Responsible AI Practices. Google AI. https://ai.google/responsibility/responsible-ai- practices/

    work page

  20. [20]

    Haig, Zsolt. (2020). Novel Interpretation of Information Operations in Today’s Changed Operational Environment. Scientific Bulletin

    work page 2020

  21. [21]

    10.2478/bsaft-2020-0013

    93-102. 10.2478/bsaft-2020-0013. 13

    work page doi:10.2478/bsaft-2020-0013 2020

  22. [22]

    Hall, G. O. (2023). Examining US-China- Russia Foreign Relations. Taylor & Francis

    work page 2023

  23. [23]

    Hendrycks, D., Mazeika, M., Zou, A., Patel, S., Zhu, C., J Fernandez Navarro, Song, D., Li, B., & Steinhardt, J. (2021). What Would Jiminy Cricket Do? Towards Agents That Behave Morally. ArXiv (Cornell University)

    work page 2021

  24. [24]

    Henry, M. (2015). Mind Games: Setting Conditions for Successful Counterinsurgency Military Information Support Operations. Pickle Partners Publishing

    work page 2015

  25. [25]

    N., Lin, F., & Viktor Tuzov

    Howard, P. N., Lin, F., & Viktor Tuzov. (2023). Computational propaganda: Concepts, methods, and challenges. Communication and the Public, 8(2), 47–53. https://doi.org/10.1177/20570473231185996

    work page doi:10.1177/20570473231185996 2023

  26. [26]

    (2023, August 31)

    Hu, B., Zhao, C., Zhang, P., Zhou, Z., Yang, Y., Xu, Z., & Liu, B. (2023, August 31). Enabling Intelligent Interactions between an Agent and an LLM: A Reinforcement Learning Approach. ArXiv.org. https://doi.org/10.48550/arXiv.2306.03604

    work page doi:10.48550/arxiv.2306.03604 2023

  27. [27]

    Jai Galliott, Jens David Ohlin, & Macintosh, D. (2020). Lethal autonomous weapons: re- examining the law and ethics of robotic warfare. Oxford University Press

    work page 2020

  28. [28]

    Jin, Y., Jang, E., Cui, J., Chung, J.-W., Yong Jae Lee, & Shin, S. (2023). DarkBERT: A Language Model for the Dark Side of the Internet. ArXiv (Cornell University). https://doi.org/10.18653/v1/2023.acl-long.415

    work page doi:10.18653/v1/2023.acl-long.415 2023

  29. [29]

    Lennart Souchon, & Springerlink (Online Service. (2020). Strategy in the 21st Century: The Continuing Relevance of Carl von Clausewitz. Springer International Publishing, Imprint Springer

    work page 2020

  30. [30]

    Lucas, G. R. (2017). Ethics and cyber warfare: the quest for responsible security in the age of digital warfare. Oxford University Press

    work page 2017

  31. [31]

    Macdonald, S. (2007). Propaganda and information warfare in the twenty-first century: altered images and deception operations. Routledge

    work page 2007

  32. [32]

    Malone, J. (2004). Information Operations in the Australian Defence Force. Www.slideshare.net. https://www.slideshare.net/jeffreymalone1/mal one-unclas-presentation

    work page 2004

  33. [33]

    Mehdi Parvizi Amineh. (2022). The China-Led Belt and Road Initiative and Its Reflections. Routledge

    work page 2022

  34. [34]

    Mikael Weissmann, Niklas Nilsson, Per Thunholm, & Björn Palmertz. (2021). Hybrid warfare: security and asymmetric conflict in international relations. I.B. Tauris

    work page 2021

  35. [35]

    Mitra, J. (2023). Conventional Military Strategy in the Third Nuclear Age. Taylor & Francis

    work page 2023

  36. [36]

    Murphy, B. (2023). Foreign Disinformation in America and the U.S. Government’s Ethical Obligations to Respond. Springer Nature

    work page 2023

  37. [37]

    Pashentsev. (2023). The Palgrave Handbook of Malicious Use of AI and Psychological Security. Springer Nature

    work page 2023

  38. [38]

    Raska, M., & Bitzinger, R. A. (2023). The AI Wave in Defence Innovation. Taylor & Francis

    work page 2023

  39. [39]

    Reinhold, T., & Niklas Schörnig. (2022). Armament, Arms Control and Artificial Intelligence. Springer Nature

    work page 2022

  40. [40]

    Richard Ned Lebow. (2007). Coercion, cooperation, and ethics in international relations. New York; London Routledge

    work page 2007

  41. [41]

    Rid, T., & Hecker, M. (2009). War 2.0: irregular warfare in the information age. Praeger Security International

    work page 2009

  42. [42]

    Scharre, P. (2023). Four Battlegrounds: Power in the Age of Artificial Intelligence. W. W. Norton & Company

    work page 2023

  43. [43]

    A., Marquardt, E., & Collins, L

    Sheehan, M. A., Marquardt, E., & Collins, L. (2021). Routledge Handbook of U.S. Counterterrorism and Irregular Warfare Operations. Routledge

    work page 2021

  44. [44]

    Thuraisingham, B. M. (2020). Can AI be for Good in the Midst of Cyber Attacks and Privacy Violations? Proceedings of the Tenth ACM Conference on Data and Application Security and Privacy. https://doi.org/10.1145/3374664.3379334 14

    work page doi:10.1145/3374664.3379334 2020

  45. [45]

    Trine Villumsen Berling, Ulrik Pram Gad, Karen Lund Petersen, & Ole Wæver. (2021). Translations of Security. Routledge

    work page 2021

  46. [46]

    Twitter. (2021). Disclosing state-linked information operations we’ve removed. Twitter. https://blog.twitter.com/en_us/topics/company/ 2021/disclosing-state-linked-information- operations-we-ve-removed

    work page 2021

  47. [47]

    Attention Is All You Need

    Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., & Polosukhin, I. (2017). Attention Is All You Need. ArXiv.org. https://arxiv.org/abs/1706.03762

    work page internal anchor Pith review Pith/arXiv arXiv 2017

  48. [48]

    Wallace, R. (2018). Carl von Clausewitz, the fog-of-war, and the AI revolution: the real world is not a game of Go. Springer

    work page 2018

  49. [49]

    (2022, March)

    Wallis, J., Zhang, A., & Niblock, I. (2022, March). Understanding global disinformation and information operations. ASPI. https://www.aspi.org.au/report/understanding_ global_disinformation_information_operations

    work page 2022

  50. [50]

    Voyager: An Open-Ended Embodied Agent with Large Language Models

    Wang, G., Xie, Y., Jiang, Y., Mandlekar, A., Xiao, C., Zhu, Y., Fan, L., & Anandkumar, A. (2023, May 25). Voyager: An Open-Ended Embodied Agent with Large Language Models. ArXiv.org. https://doi.org/10.48550/arXiv.2305.16291

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2305.16291 2023

  51. [51]

    A Survey on Large Language Model based Autonomous Agents

    Wang, L., Ma, C., Feng, X., Zhang, Z., Yang, H., Zhang, J., Chen, Z., Tang, J., Chen, X., Lin, Y., Zhao, W. X., Wei, Z., & Wen, J.-R. (2023, September 7). A Survey on Large Language Model based Autonomous Agents. ArXiv.org. https://doi.org/10.48550/arXiv.2308.11432

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2308.11432 2023

  52. [52]

    Welch, D., & Fox, J. (2012). Justifying War : Propaganda, Politics and the Modern Age. Palgrave Macmillan

    work page 2012

  53. [53]

    Strategic Studies Institute, & Army War College (U.S.)

    Williams, P., Dighton Fiddner, Army War College (U.S.). Strategic Studies Institute, & Army War College (U.S.). Press. (2016). Cyberspace : malevolent actors, criminal opportunities, and strategic competition. Strategic Studies Institute And U.S. Army War College Press, Washington, D.C

    work page 2016

  54. [54]

    Wright, R. H. (2001). Information Operations: Doctrine, Tactics, Techniques and Procedures. Military Review, 81(2), 30

    work page 2001

  55. [55]

    E., Li, L., Wu, J., Wang, T., Qiu, S., Zhang, J., Chen, J., Wu, R., Wang, S., Zhu, S., Chen, J., Zhang, W., Zhang, N., Chen, H., Cui, P., & Sachan, M

    Zhou, W., Jiang, Y. E., Li, L., Wu, J., Wang, T., Qiu, S., Zhang, J., Chen, J., Wu, R., Wang, S., Zhu, S., Chen, J., Zhang, W., Zhang, N., Chen, H., Cui, P., & Sachan, M. (2023, September 14). Agents: An Open-source Framework for Autonomous Language Agents. ArXiv.org. https://doi.org/10.48550/arXiv.2309.07870

    work page doi:10.48550/arxiv.2309.07870 2023