Pith. sign in

REVIEW 3 major objections 1 cited by

Relativistic motion lets a quantum engine beat the Carnot limit

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

Relativistic motion of the reservoirs in a three-level maser is claimed to yield a generalized Carnot bound that allows efficiency above the ordinary Carnot limit.

T0 review reviewed 2026-08-05 challenge →

load-bearing objection The abstract promises a relativistic quantum thermal machine, but the supplied full text is an unrelated LLM-detection paper, so the central claims have no supporting text at all. the 3 major comments →

arxiv 2508.14183 v1 pith:5IXYKSYG submitted 2025-08-19 quant-ph hep-th

Relativistic Quantum Thermal Machine: Harnessing Relativistic Effects to Surpass Carnot Efficiency

Tanmoy Pandit , Pritam Chattopadhyay , Kaustav Chatterjee , Varinder Singh This is my paper
classification quant-ph hep-th PACS 03.65.-w05.70.-a
keywords relativistic quantum thermodynamicsquantum thermal machinethree-level maserUnruh-DeWitt couplingDoppler effectCarnot efficiencyefficiency at finite powerquantum heat engine
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The reading

This paper claims that a three-level quantum maser coupled to thermal reservoirs can exceed the standard Carnot efficiency when one or both of those reservoirs move relativistically relative to the working medium. The motion Doppler-shifts the reservoirs' spectra, changing which transitions absorb and emit energy and redrawing the boundaries between heat-engine and refrigerator operation. The authors compute efficiency-power curves numerically and extract an analytic generalized Carnot bound that reduces to the usual Carnot value when motion is absent. They also show that positive work can be extracted even when both reservoirs have the same temperature. If correct, this establishes relativistic motion itself as a genuine thermodynamic resource.

Core claim

The central claim is that the Doppler reshaping of a moving thermal reservoir's spectrum functions as a control knob for a quantum thermal machine. Using Unruh-DeWitt type coupling—the standard pointlike interaction between an atom and a quantum field—between a three-level maser and its reservoirs, the authors show that the rates of energy exchange become speed-dependent. This speed dependence lets the device operate beyond the static Carnot bound at finite power, and it shifts the threshold between engine and refrigerator. The paper derives an analytic expression for a generalized Carnot efficiency that contains the ordinary Carnot bound as its nonrelativistic limit, and demonstrates work e

What carries the argument

The load-bearing object is the Doppler-shifted spectral density of a moving thermal reservoir as seen by the three-level maser through Unruh-DeWitt coupling. The relative motion compresses or stretches the frequency dependence of the bath's coupling, so the effective temperature and transition rates become anisotropic. This asymmetry is what allows the device to extract work from reservoirs that are at the same temperature in their own rest frames.

Load-bearing premise

The moving reservoirs are treated as ordinary thermal baths whose spectra are only Doppler-shifted, with the same Unruh-DeWitt coupling and a standard quantum master equation; if the motion produces non-thermal correlations or changes the dissipative structure beyond a spectral shift, the generalized Carnot bound would not follow.

What would settle it

Run the same three-level maser calculation with both reservoirs at exactly the same temperature, replacing the Doppler-shifted spectrum with the exact quantum state of a uniformly moving thermal bath; the paper predicts a window of positive output work, so finding zero work for every velocity would falsify the central claim.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

If this is right

  • The generalized Carnot bound becomes the standard Carnot bound in the nonrelativistic limit, so nothing here contradicts known thermodynamics at everyday speeds.
  • A quantum heat engine can deliver positive power without any temperature gradient, provided the reservoirs move fast enough.
  • Changing the relative velocity can switch the same device between heat-engine and refrigerator modes without changing the bath temperatures.
  • Relativistic motion functions as a thermodynamic resource that can be added to a static thermal machine to push it past its usual efficiency.

Where Pith is reading between the lines

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

  • If the mechanism is real, any interaction that asymmetrically reshapes a bath spectrum—e.g., coupling to a non-equilibrium field or a moving mirror—could mimic the effect without needing actual relativistic speeds.
  • One could test the zero-temperature-gradient claim by driving a superconducting qubit with two noise sources at the same nominal temperature but different Doppler shifts, and measuring the output power.
  • The generalized bound may be re-expressed as an effective speed-dependent temperature ratio, which would connect this setup to existing studies of efficiency bounds for nonthermal baths.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 0 minor

Summary. The abstract announces a study of a three-level maser quantum thermal machine with Unruh-DeWitt type coupling, Doppler-shifted reservoir spectra, numerical efficiency-power curves, an analytic generalized Carnot bound, and work extraction without a temperature gradient. The full text supplied is an entirely different manuscript: "Two Birds with One Stone: Multi-Task Detection and Attribution of LLM-Generated Text" by Rao et al. This body text contains no equations, sections, or results pertaining to Unruh-DeWitt couplings, Doppler reshaping, quantum master equations, thermodynamic bounds, or any of the abstract's claims. The submitted manuscript therefore cannot support any aspect of the announced research.

Significance. If the abstract's claims were substantiated, the result would be significant: it would challenge the universality of the Carnot bound in a relativistic setting and identify relativistic motion as a genuine thermodynamic resource. However, since none of the supporting derivation, numerical analysis, or modeling framework is present in the submitted text, the significance cannot be evaluated. The manuscript as it stands offers no verifiable content on which to base an assessment.

major comments (3)
  1. [Full text vs. Abstract] The full text is an unrelated paper on LLM-generated text detection and attribution. It contains no mention of Unruh-DeWitt couplings, Doppler reshaping, maser thermal machines, efficiency-power curves, or a generalized Carnot bound. Every load-bearing element of the abstract's central claim is absent from the body. This is not a subtle modeling concern; the manuscript internally contradicts its own abstract and cannot be refereed on its stated topic.
  2. [Abstract, numerical and analytic claims] The abstract states that the authors "numerically analyze families of efficiency-power curves" and "extract the analytic form of a generalized Carnot bound." The submitted text contains none of these: no equations, no figures, no datasets, no error analysis, and no derivation. No independent check of the claimed bound is possible from the material provided, and the claim of surpassing Carnot efficiency at finite power is therefore entirely unsupported.
  3. [Entire manuscript] Because the body text is a different paper, standard evaluation criteria—such as correctness of the master equation, validity of the Doppler-shift approximation, thermodynamic consistency, and numerical convergence—cannot be applied. This is a load-bearing failure that cannot be fixed within the scope of the present revision; the correct manuscript would have to be submitted.

Circularity Check

0 steps flagged

No circularity identified; the provided manuscript text does not contain the claimed quantum thermal machine derivation.

full rationale

The circularity pass can only operate on a derivation chain. The abstract of arXiv:2508.14183 announces a three-level maser quantum thermal machine with Unruh-DeWitt coupling, Doppler reshaped reservoir spectra, efficiency-power curves, and a generalized Carnot bound. The supplied full text, however, is an unrelated paper titled 'Two Birds with One Stone: Multi-Task Detection and Attribution of LLM-Generated Text', with no equations, sections, or numerical analysis concerning relativistic thermodynamics, Unruh-DeWitt interactions, Doppler spectral reshaping, maser heat engines, or Carnot efficiency. Since none of the claimed derivation exists in the manuscript, there is no equation that can be shown to reduce to an input, no fitted parameter that is relabeled as a prediction, and no cited prior result that carries the argument. The mismatch between abstract and body is a serious integrity/support problem, but it is not a case of circular reasoning as defined by this pass: nothing is derived, so nothing can be circular. Score 0 is therefore the appropriate circularity verdict.

Axiom & Free-Parameter Ledger

0 free parameters · 3 axioms · 0 invented entities

Only domain assumptions are listed based on the abstract. No free parameters are visible because the abstract does not describe any fitted quantities. No new physical entities are proposed.

axioms (3)
  • domain assumption The system-reservoir interaction is described by the Unruh-DeWitt coupling model.
    Stated in the abstract as the modeling choice; this is not derived.
  • domain assumption Relativistic motion only Doppler-shifts the reservoir spectra while the reservoirs remain thermal baths.
    The abstract describes Doppler reshaping as the mechanism; this presumes thermal character is preserved under motion.
  • domain assumption A master equation (implied by 'energy-exchange rates') provides a valid description of the dynamics.
    Thermal machine analyses typically rely on Lindblad or similar master equations; not stated in abstract but required.

reviewed 2026-08-05 · how reviews work

0 comments
Cite this review

Pith. "Pith review of Relativistic Quantum Thermal Machine: Harnessing Relativistic Effects to Surpass Carnot Efficiency." pith.science (2026). https://pith.science/paper/5IXYKSYG

@misc{pith2026250814183,
  author       = {Pith},
  title        = {Pith review of: Relativistic Quantum Thermal Machine: Harnessing Relativistic Effects to Surpass Carnot Efficiency},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5IXYKSYG}},
  note         = {Machine review of arXiv:2508.14183}
}
Share X Bluesky LinkedIn Reddit HN
read the original abstract

We investigate a three-level maser quantum thermal machine in which the system-reservoir interaction is modeled via Unruh-DeWitt type coupling, with one or both reservoirs undergoing relativistic motion relative to the working medium. Motion induces Doppler reshaping of the reservoir spectra, modifying energy-exchange rates and enabling operation beyond the Carnot efficiency at finite power. We numerically analyze families of efficiency-power curves and extract the analytic form of a generalized Carnot bound, which recovers the Carnot limit. In addition, Doppler reshaping alters the boundaries between heat-engine and refrigerator operation, making it possible to extract positive work even in the absence of a temperature gradient. These findings establish relativistic motion as a genuine thermodynamic resource.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Role of Asymmetry in the Performance Optimization of a Relativistic Quantum Otto Engine

    quant-ph 2026-04 unverdicted novelty 6.0

    Asymmetry in the adiabatic processes of a relativistic quantum Otto cycle allows efficiency to approach unity under sudden compression but restricts it to one-half under sudden expansion, with increasing oscillator ve...

Reference graph

Works this paper leans on

76 extracted references · 56 canonical work pages · cited by 1 Pith paper · 1 internal anchor

  1. [1]

    arXiv preprint arXiv:2303.08774 (2023)

    Achiam, J., Adler, S., Agarwal, S., Ahmad, L., Akkaya, I., Aleman, F.L., Almeida, D., Altenschmidt, J., Altman, S., Anadkat, S., et al.: Gpt-4 technical report. arXiv preprint arXiv:2303.08774 (2023)

    Pith/arXiv arXiv 2023

  2. [2]

    In: Proceedings of the second ACM conference on Online social networks

    Almishari, M., Oguz, E., Tsudik, G.: Fighting authorship linkability with crowd- sourcing. In: Proceedings of the second ACM conference on Online social networks. pp. 69–82 (2014)

    work page 2014

  3. [3]

    Anders,B.A.:Isusingchatgptcheating,plagiarism,both,neither,orforwardthink- ing? Patterns4(3) (2023)

    work page 2023

  4. [4]

    Journal of machine learning research6(11) (2005)

    Ando, R.K., Zhang, T., Bartlett, P.: A framework for learning predictive structures from multiple tasks and unlabeled data. Journal of machine learning research6(11) (2005)

    work page 2005

  5. [5]

    Available at:https://claude.ai/ (2023)

    Anthropic: Claude. Available at:https://claude.ai/ (2023)

    work page 2023

  6. [6]

    Machine learning28, 41–75 (1997)

    Caruana, R.: Multitask learning. Machine learning28, 41–75 (1997)

    work page 1997

  7. [7]

    In: Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing

    Chen, Y., Kang, H., Zhai, V., Li, L., Singh, R., Raj, B.: Token prediction as implicit classification to identify llm-generated text. In: Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing. pp. 13112–13120 (2023)

    work page 2023

  8. [8]

    In: Proceedings of the Annual Meeting of the Association for Computational Linguistics (2020)

    Conneau, A., Khandelwal, K., Goyal, N., Chaudhary, V., Wenzek, G., et al.: Un- supervised cross-lingual representation learning at scale. In: Proceedings of the Annual Meeting of the Association for Computational Linguistics (2020)

    work page 2020

  9. [9]

    Available at: https://copyleaks.com/ ai-content-detector (2023)

    Copyleaks: AI Detector. Available at: https://copyleaks.com/ ai-content-detector (2023)

    work page 2023

  10. [10]

    arXiv preprint arXiv:2009.09796 (2020)

    Crawshaw, M.: Multi-task learning with deep neural networks: A survey. arXiv preprint arXiv:2009.09796 (2020)

    Pith/arXiv arXiv 2009

  11. [11]

    arXiv preprint arXiv:2305.08005 (2023)

    Derner, E., Batistič, K.: Beyond the safeguards: exploring the security risks of chatgpt. arXiv preprint arXiv:2305.08005 (2023)

    Pith/arXiv arXiv 2023

  12. [12]

    arXiv preprint arXiv:2402.06664 (2024)

    Fang, R., Bindu, R., Gupta, A., Zhan, Q., Kang, D.: Llm agents can autonomously hack websites. arXiv preprint arXiv:2402.06664 (2024)

    Pith/arXiv arXiv 2024

  13. [13]

    arXiv preprint arXiv:2307.09793 (2023)

    Gao, S., Gao, A.K.: On the origin of llms: An evolutionary tree and graph for 15,821 large language models. arXiv preprint arXiv:2307.09793 (2023)

    Pith/arXiv arXiv 2023

  14. [14]

    Gehrmann, S., Strobelt, H., Rush, A.M.: Gltr: Statistical detection and visualiza- tion of generated text (2019)

    work page 2019

  15. [15]

    Available at:https://gemini.google.com/ (2023)

    Google: Gemini. Available at:https://gemini.google.com/ (2023)

    work page 2023

  16. [16]

    In: 2023 22nd Interna- tional Symposium INFOTEH-JAHORINA (INFOTEH)

    Grbic, D.V., Dujlovic, I.: Social engineering with chatgpt. In: 2023 22nd Interna- tional Symposium INFOTEH-JAHORINA (INFOTEH). pp. 1–5. IEEE (2023)

    work page 2023

  17. [17]

    Nature10 (2024)

    Gruda, D.: Three ways chatgpt helps me in my academic writing. Nature10 (2024)

    work page 2024

  18. [18]

    arXiv preprint arXiv:2301.07597 (2023)

    Guo, B., Zhang, X., Wang, Z., Jiang, M., Nie, J., Ding, Y., Yue, J., Wu, Y.: How close is chatgpt to human experts? comparison corpus, evaluation, and detection. arXiv preprint arXiv:2301.07597 (2023)

    Pith/arXiv arXiv 2023

  19. [19]

    In: Proceedings of the International AAAI Conference on Web and Social Media

    Hanley, H.W., Durumeric, Z.: Machine-made media: Monitoring the mobilization of machine-generated articles on misinformation and mainstream news websites. In: Proceedings of the International AAAI Conference on Web and Social Media. vol. 18, pp. 542–556 (2024)

    work page 2024

  20. [20]

    In: Proceedings of the 2024 on ACM SIGSAC Conference on Computer and Communications Security

    He, X., Shen, X., Chen, Z., Backes, M., Zhang, Y.: Mgtbench: Benchmarking machine-generated text detection. In: Proceedings of the 2024 on ACM SIGSAC Conference on Computer and Communications Security. pp. 2251–2265 (2024)

    work page 2024

  21. [21]

    In: Companion Proceedings of the The Web Conference 2018

    Horne,B.D.,Dron,W.,Khedr,S.,Adali,S.:Assessingthenewslandscape:Amulti- module toolkit for evaluating the credibility of news. In: Companion Proceedings of the The Web Conference 2018. pp. 235–238 (2018)

    work page 2018

  22. [22]

    ACM Transactions on Software Engineering and Methodology (2023)

    Hou, X., Zhao, Y., Liu, Y., Yang, Z., Wang, K., Li, L., Luo, X., Lo, D., Grundy, J., Wang, H.: Large language models for software engineering: A systematic literature review. ACM Transactions on Software Engineering and Methodology (2023)

    work page 2023

  23. [23]

    In: Web and Big Data: 4th International Joint Conference

    Hu, Z., Lee, R.K.W., Wang, L., Lim, E.p., Dai, B.: Deepstyle: User style embedding for authorship attribution of short texts. In: Web and Big Data: 4th International Joint Conference. Springer (2020)

    work page 2020

  24. [24]

    Huang, B., Chen, C., Shu, K.: Can large language models identify authorship? In: Findings of EMNLP (2024)

    work page 2024

  25. [25]

    In: Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics

    Ippolito, D., Duckworth, D., Eck, D.: Automatic detection of generated text is easiest when humans are fooled. In: Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. pp. 1808–1822 (2020)

    work page 2020

  26. [26]

    arXiv preprint arXiv:2011.01314 (2020)

    Jawahar, G., Abdul-Mageed, M., Lakshmanan, L.V.: Automatic detection of ma- chine generated text: A critical survey. arXiv preprint arXiv:2011.01314 (2020)

    Pith/arXiv arXiv 2011

  27. [27]

    In: Proceedings of naacL-HLT

    Kenton, J.D.M.W.C., Toutanova, L.K.: Bert: Pre-training of deep bidirectional transformers for language understanding. In: Proceedings of naacL-HLT. vol. 1. Minneapolis, Minnesota (2019)

    work page 2019

  28. [28]

    Kestemont, M., et al.: Overview of the author identification task at pan-2018: Cross-domain authorship attribution and style change detection (2018)

    work page 2018

  29. [29]

    CLEF (Working Notes)1609, 890–894 (2016)

    Keswani, Y., Trivedi, H., Mehta, P., Majumder, P.: Author masking through trans- lation. CLEF (Working Notes)1609, 890–894 (2016)

    work page 2016

  30. [30]

    Applied Sciences14(21), 9795 (2024)

    Krawczyk, N., Probierz, B., Kozak, J.: Towards ai-generated essay classification using numerical text representation. Applied Sciences14(21), 9795 (2024)

    work page 2024

  31. [31]

    In: 2023 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery (CyberC)

    Kumarage, T.,Liu,H.:Neural authorshipattribution:Stylometricanalysisonlarge language models. In: 2023 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery (CyberC). pp. 51–54. IEEE (2023)

    work page 2023

  32. [32]

    In: EMNLP (2021)

    Kushnareva, L., Artemova, E., Piontkovskaya, I., Cherniavskii, D., Barannikov, S., Bernstein, A., Burnaev, E., Mikhailov, V., Piontkovski, D.: Artificial text detection via examining the topology of attention maps. In: EMNLP (2021)

    work page 2021

  33. [33]

    arXiv preprint arXiv:2304.07666 (2023)

    Liu, Y., Zhang, Z., Zhang, W., Yue, S., Zhao, X., Cheng, X., Zhang, Y., Hu, H.: Ar- gugpt: evaluating, understanding and identifying argumentative essays generated by gpt models. arXiv preprint arXiv:2304.07666 (2023)

    Pith/arXiv arXiv 2023

  34. [34]

    In: Proceedings of the 2024 on ACM SIGSAC Conference on Computer and Commu- nications Security

    Liu, Z., Yao, Z., Li, F., Luo, B.: On the detectability of chatgpt content: bench- marking, methodology, and evaluation through the lens of academic writing. In: Proceedings of the 2024 on ACM SIGSAC Conference on Computer and Commu- nications Security. pp. 2236–2250 (2024)

    work page 2024

  35. [35]

    Proceedings of the ACM on Human-Computer Interaction6(CSCW2), 1–27 (2022)

    Lu, Z., Li, P., Wang, W., Yin, M.: The effects of ai-based credibility indicators on the detection and spread of misinformation under social influence. Proceedings of the ACM on Human-Computer Interaction6(CSCW2), 1–27 (2022)

    work page 2022

  36. [36]

    In: Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing

    Macko, D., Moro, R., Uchendu, A., Lucas, J., Yamashita, M., Pikuliak, M., Srba, I., Le, T., Lee, D., Simko, J., et al.: Multitude: Large-scale multilingual machine- generated text detection benchmark. In: Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing. pp. 9960–9987 (2023)

    work page 2023

  37. [37]

    Macko, D., Moro, R., Uchendu, A., Srba, I., Lucas, J.S., et al.: Authorship obfus- cation in multilingual machine-generated text detection (2024)

    work page 2024

  38. [38]

    Proceedings on Privacy Enhancing Technologies (2019)

    Mahmood, A., Ahmad, F., Shafiq, Z., Srinivasan, P., Zaffar, F.: A girl has no name: Automated authorship obfuscation using mutant-x. Proceedings on Privacy Enhancing Technologies (2019)

    work page 2019

  39. [39]

    In: International Conference on Machine Learning

    Mitchell, E., Lee, Y., Khazatsky, A., Manning, C.D., Finn, C.: Detectgpt: Zero-shot machine-generated text detection using probability curvature. In: International Conference on Machine Learning. pp. 24950–24962. PMLR (2023)

    work page 2023

  40. [40]

    Available at: https:// copyleaks.com/ai-content-detector (2023)

    Moderation, H.: AI-Generated Content Detection. Available at: https:// copyleaks.com/ai-content-detector (2023)

    work page 2023

  41. [41]

    In: Pro- ceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume

    Munir, S., Batool, B., Shafiq, Z., Srinivasan, P., Zaffar, F.: Through the looking glass: Learning to attribute synthetic text generated by language models. In: Pro- ceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume. pp. 1811–1822 (2021)

    work page 2021

  42. [42]

    arXiv preprint arXiv:2307.06435 (2023)

    Naveed, H., Khan, A.U., Qiu, S., Saqib, M., Anwar, S., Usman, M., Akhtar, N., Barnes, N., Mian, A.: A comprehensive overview of large language models. arXiv preprint arXiv:2307.06435 (2023)

    Pith/arXiv arXiv 2023

  43. [43]

    Available at:https://chatgpt.com/ (2022)

    OpenAI: ChatGPT. Available at:https://chatgpt.com/ (2022)

    work page 2022

  44. [44]

    Available at: https://openai.com/index/ chatgpt/ (2023)

    OpenAI: Introducing ChatGPT. Available at: https://openai.com/index/ chatgpt/ (2023)

    work page 2023

  45. [45]

    Journal of machine learning research21(140), 1–67 (2020)

    Raffel, C., Shazeer, N., Roberts, A., Lee, K., Narang, S., Matena, M., Zhou, Y., Li, W., Liu, P.J.: Exploring the limits of transfer learning with a unified text-to-text transformer. Journal of machine learning research21(140), 1–67 (2020)

    2020

  46. [46]

    arXiv preprint arXiv:2305.05133 (2023)

    Roy,S.S.,Naragam,K.V.,Nilizadeh,S.:Generatingphishingattacksusingchatgpt. arXiv preprint arXiv:2305.05133 (2023)

    Pith/arXiv arXiv 2023

  47. [47]

    In: Proceedings of the AAAI conference on artificial intelligence

    Ruder, S., Bingel, J., Augenstein, I., Søgaard, A.: Latent multi-task architecture learning. In: Proceedings of the AAAI conference on artificial intelligence. vol. 33, pp. 4822–4829 (2019)

    work page 2019

  48. [48]

    Sanh,V.,Debut,L.,Chaumond,J.,Wolf,T.,Face,H.:Distilbert,adistilledversion of bert: smaller, faster, cheaper and lighter (2019)

    work page 2019

  49. [49]

    Ad- vances in neural information processing systems31 (2018)

    Sener, O., Koltun, V.: Multi-task learning as multi-objective optimization. Ad- vances in neural information processing systems31 (2018)

    work page 2018

  50. [50]

    arXiv preprint arXiv:1908.09203 (2019)

    Solaiman, I., Brundage, M., Clark, J., Askell, A., Herbert-Voss, A., Wu, J., Rad- ford, A., Krueger, G., Kim, J.W., Kreps, S., et al.: Release strategies and the social impacts of language models. arXiv preprint arXiv:1908.09203 (2019)

    Pith/arXiv arXiv 1908

  51. [51]

    In: Proceedings of 5th International Joint Conference on Natural Language Processing

    Solorio, T., Pillay, S., Raghavan, S., Montes, M.: Modality specific meta features for authorship attribution in web forum posts. In: Proceedings of 5th International Joint Conference on Natural Language Processing. pp. 156–164 (2011)

    work page 2011

  52. [52]

    In: The Twelfth International Conference on Learning Representations

    Soto, R.A.R., Koch, K., Khan, A., Chen, B.Y., Bishop, M., Andrews, N.: Few-shot detection of machine-generated text using style representations. In: The Twelfth International Conference on Learning Representations

    work page

  53. [53]

    Journal of the American Society for information Science and Technology (2009)

    Stamatatos, E.: A survey of modern authorship attribution methods. Journal of the American Society for information Science and Technology (2009)

    work page 2009

  54. [54]

    In: Findings of the Association for Computational Linguistics: EMNLP 2023

    Su, J., Zhuo, T., Wang, D., Nakov, P.: Detectllm: Leveraging log rank information for zero-shot detection of machine-generated text. In: Findings of the Association for Computational Linguistics: EMNLP 2023. pp. 12395–12412 (2023)

    work page 2023

  55. [55]

    Com- munications of the ACM67(4), 50–59 (2024)

    Tang, R., Chuang, Y.N., Hu, X.: The science of detecting llm-generated text. Com- munications of the ACM67(4), 50–59 (2024)

    work page 2024

  56. [56]

    In: ACL Joint Workshop on Multiword Expressions and WordNet

    Taslimipoor, S., Rohanian, O., et al.: Cross-lingual transfer learning and multi- task learning for capturing multiword expressions. In: ACL Joint Workshop on Multiword Expressions and WordNet. pp. 155–161 (2019)

    work page 2019

  57. [57]

    Nature medicine29(8), 1930–1940 (2023)

    Thirunavukarasu, A.J., Ting, D.S.J., Elangovan, K., Gutierrez, L., Tan, T.F., Ting, D.S.W.: Large language models in medicine. Nature medicine29(8), 1930–1940 (2023)

    1930

  58. [58]

    Available at:https://gptzero.me/ (2023)

    Tian, E.: GPTZero. Available at:https://gptzero.me/ (2023)

    work page 2023

  59. [59]

    Language Resources and Evaluation (58), 713–755 (2023)

    Tiedemann, J., Aulamo, M., Bakshandaeva, D., Boggia, M., Gronroos, S.A., Niem- inen, T., Raganato, A., Scherrer, Y., Vazquez, R., Virpioja, S.: Democratizing neu- ral machine translation with OPUS-MT. Language Resources and Evaluation (58), 713–755 (2023). https://doi.org/10.1007/s10579-023-09704-w

    doi:10.1007/s10579-023-09704-w 2023

  60. [60]

    In: Proceedings of the 22nd Annual Conferenec of the European Asso- ciation for Machine Translation (EAMT)

    Tiedemann, J., Thottingal, S.: Opus-mt — building open translation services for the world. In: Proceedings of the 22nd Annual Conferenec of the European Asso- ciation for Machine Translation (EAMT). Lisbon, Portugal (2020)

    work page 2020

  61. [61]

    arXiv preprint arXiv:2302.13971 (2023)

    Touvron, H., Lavril, T., Izacard, G., Martinet, X., Lachaux, M.A., Lacroix, T., Rozière, B., Goyal, N., Hambro, E., Azhar, F., et al.: Llama: Open and efficient foundation language models. arXiv preprint arXiv:2302.13971 (2023)

    Pith/arXiv arXiv 2023

  62. [62]

    arXiv preprint arXiv:2307.09288 (2023)

    Touvron, H., Martin, L., Stone, K., Albert, P., Almahairi, A., Babaei, Y., Bash- lykov, N., Batra, S., Bhargava, P., Bhosale, S., et al.: Llama 2: Open foundation and fine-tuned chat models. arXiv preprint arXiv:2307.09288 (2023)

    Pith/arXiv arXiv 2023

  63. [63]

    arXiv preprint arXiv:2304.14106 (2023)

    Tu, S., Li, C., Yu, J., Wang, X., Hou, L., Li, J.: Chatlog: Recording and analyzing chatgpt across time. arXiv preprint arXiv:2304.14106 (2023)

    Pith/arXiv arXiv 2023

  64. [64]

    In: Proceedings of EMNLP

    Uchendu, A., Le, T., Shu, K., Lee, D.: Authorship attribution for neural text generation. In: Proceedings of EMNLP. pp. 8384–8395 (2020)

    work page 2020

  65. [65]

    In: Findings of the Association for Computational Linguistics: EMNLP 2021

    Uchendu, A., Ma, Z., Le, T., Zhang, R., Lee, D.: Turingbench: A benchmark en- vironment for turing test in the age of neural text generation. In: Findings of the Association for Computational Linguistics: EMNLP 2021. pp. 2001–2016 (2021)

    work page 2021

  66. [66]

    Venkatraman, S., Uchendu, A., Lee, D.: Gpt-who: An information density-based machine-generatedtextdetector.In:FindingsoftheAssociationforComputational Linguistics: NAACL 2024. pp. 103–115 (2024)

    work page 2024

  67. [67]

    In: BlackboxNLP (2018)

    Wang, A., Singh, A., Michael, J., et al.: GLUE: A multi-task benchmark and analysis platform for natural language understanding. In: BlackboxNLP (2018)

    work page 2018

  68. [68]

    In: Proceedings of EACL (2024)

    Wang, Y., Mansurov, J., Ivanov, P., Su, J., Shelmanov, A., Tsvigun, A., White- house, C., Afzal, O.M., Mahmoud, T., Sasaki, T., et al.: M4: Multi-generator, multi-domain, and multi-lingual black-box machine-generated text detection. In: Proceedings of EACL (2024)

    work page 2024

  69. [69]

    In: Proceedings of the Annual Meeting of the Association for Computational Linguistics (2024)

    Wang, Y., Mansurov, J., Ivanov, P., et al.: M4GT-bench: Evaluation benchmark for black-box machine-generated text detection. In: Proceedings of the Annual Meeting of the Association for Computational Linguistics (2024)

    work page 2024

  70. [70]

    Transactions on Machine Learning Research (2022)

    Wei, J., Tay, Y., Bommasani, R., Raffel, C., Zoph, B., Borgeaud, S., Yogatama, D., Bosma, M., Zhou, D., Metzler, D., et al.: Emergent abilities of large language models. Transactions on Machine Learning Research (2022)

    2022

  71. [71]

    British Journal of Educational Technology 55(1), 90–112 (2024)

    Yan, L., Sha, L., Zhao, L., Li, Y., Martinez-Maldonado, R., Chen, G., Li, X., Jin, Y., Gašević, D.: Practical and ethical challenges of large language models in education: A systematic scoping review. British Journal of Educational Technology 55(1), 90–112 (2024)

    work page 2024

  72. [72]

    100211 (2024)

    Yao, Y., Duan, J., Xu, K., Cai, Y., Sun, Z., Zhang, Y.: A survey on large language model(llm)securityandprivacy:Thegood,thebad,andtheugly.High-Confidence Computing p. 100211 (2024)

    work page 2024

  73. [73]

    Advances in Neural Information Processing Systems33, 5824–5836 (2020)

    Yu, T., Kumar, S., Gupta, A., Levine, S., Hausman, K., Finn, C.: Gradient surgery for multi-task learning. Advances in Neural Information Processing Systems33, 5824–5836 (2020)

    2020

  74. [74]

    Advances in neural information processing systems 32 (2019)

    Zellers, R., Holtzman, A., Rashkin, H., Bisk, Y., Farhadi, A., Roesner, F., Choi, Y.: Defending against neural fake news. Advances in neural information processing systems 32 (2019)

    work page 2019

  75. [75]

    arXiv preprint arXiv:2205.01068 (2022)

    Zhang, S., Roller, S., Goyal, N., et al.: Opt: Open pre-trained transformer language models. arXiv preprint arXiv:2205.01068 (2022)

    Pith/arXiv arXiv 2022

  76. [76]

    Enhancing Text Authenticity: A Novel Hybrid Approach for AI-Generated Text Detection

    Zhang, Y., Leng, Q., Zhu, M., Ding, R., Wu, Y., Song, J., Gong, Y.: Enhancing text authenticity: A novel hybrid approach for ai-generated text detection. arXiv preprint arXiv:2406.06558 (2024)

    work page internal anchor Pith review Pith/arXiv arXiv 2024

This paper was first reviewed by deepseek-v4-flash on August 5, 2026.