Fetching the paper…
Reading the bibliography…
Since ChatGPT was introduced in November 2022, embedding (nearly) unnoticeable statistical signals into text generated by large language models (LLMs), also known as watermarking, has been used as a principled approach to provable detection of LLM-generated text from its human-written counterpart.
Statistical theory of extreme values and some practical applications: A series of lectures
E. J. Gumbel · 1948
Earlier work this paper cites.
On the asymptotic efficiency of tests and estimates
R. R. Bahadur · 1960
Earlier work this paper cites.
New fast method for generating discrete random numbers with arbitrary frequency distributions
A. J. Walker · 1974
Earlier work this paper cites.
An efficient method for generating discrete random variables with general distributions
A. J. Walker · 1977
Earlier work this paper cites.
Asymptotic evaluation of certain Markov process expectations for large time. IV
M. D. Donsker and S. S. Varadhan · 1983
Earlier work this paper cites.
A learning algorithm for Boltzmann machines
D. H. Ackley, G. E. Hinton, and T. J. Sejnowski · 1985
Earlier work this paper cites.
Convex analysis and minimization algorithms I: Fundamentals
J.-B. Hiriart-Urruty and C. Lemaréchal · 1996
Earlier work this paper cites.
Applied Cryptography
B. Schneier · 1996
Earlier work this paper cites.
Asymptotic statistics
A. W. Van der Vaart · 2000
Earlier work this paper cites.
Natural language watermarking: Design, analysis, and a proof-of-concept implementation
M. J. Atallah, V. Raskin, M. Crogan, C. Hempelmann, F. Kerschbaum, D. Mohamed, and S. Naik · 2001
Earlier work this paper cites.
Watermarking security: Theory and practice
F. Cayre, C. Fontaine, and T. Furon · 2005
Earlier work this paper cites.
Cryptography: Theory and practice
D. R. Stinson · 2005
Earlier work this paper cites.
Nonuniform random variate generation
L. Devroye · 2006
Earlier work this paper cites.
The hiding virtues of ambiguity: Quantifiably resilient watermarking of natural language text through synonym substitutions
U. Topkara, M. Topkara, and M. J. Atallah · 2006
Earlier work this paper cites.
Digital watermarking and steganography
I. Cox, M. Miller, J. Bloom, J. Fridrich, and T. Kalker · 2007
Earlier work this paper cites.
Introduction to Modern Cryptography
J. Katz and Y. Lindell · 2008
Earlier work this paper cites.
Large deviations techniques and applications
A. Dembo and O. Zeitouni · 2009
Earlier work this paper cites.
Understanding cryptography: A textbook for students and practitioners
C. Paar and J. Pelzl · 2009
Earlier work this paper cites.
Security theory and attack analysis for text watermarking
X. Zhou, W. Zhao, Z. Wang, and L. Pan · 2009
Earlier work this paper cites.
Perturb-and-map random fields: Using discrete optimization to learn and sample from energy models
G. Papandreou and A. L. Yuille · 2011
Earlier work this paper cites.
On the partition function and random maximum a-posteriori perturbations
T. Hazan and T. Jaakkola · 2012
Earlier work this paper cites.
Random utility theory for social choice
H. A. Soufiani, D. C. Parkes, and L. Xia · 2012
Earlier work this paper cites.
Probability: Theory and Examples (Edition 4.1)
R. Durrett · 2013
Earlier work this paper cites.
A* sampling
C. J. Maddison, D. Tarlow, and T. Minka · 2014
Earlier work this paper cites.
PCG: A family of simple fast space-efficient statistically good algorithms for random number generation
M. E. O’neill · 2014
Earlier work this paper cites.
Categorical reparameterization with Gumbel-Softmax
E. Jang, S. Gu, and B. Poole · 2016
Earlier work this paper cites.
Human behavior and the principle of least effort: An introduction to human ecology
G. K. Zipf · 2016
Earlier work this paper cites.
Attention is all you need
A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, Ł. Kaiser, and I. Polosukhin · 2017
Earlier work this paper cites.
Concentration inequalities for randomly permuted sums
M. Albert · 2019
Cited alongside, same era.
GLTR: Statistical detection and visualization of generated text
S. Gehrmann, H. Strobelt, and A. Rush · 2019
Cited alongside, same era.
Convexification of Permutation-invariant Sets
J. Kim, M. Tawarmalani, and J.-P. P. Richar · 2019
Cited alongside, same era.
Language models are unsupervised multitask learners
A. Radford, J. Wu, R. Child, D. Luan, D. Amodei, I. Sutskever, et al · 2019
Cited alongside, same era.
Disinformation’s spread: Bots, trolls and all of us
K. Starbird · 2019
Cited alongside, same era.
Defending against neural fake news
R. Zellers, A. Holtzman, H. Rashkin, Y. Bisk, A. Farhadi, F. Roesner, and Y. Choi · 2019
Cited alongside, same era.
On the reliability of watermarks for large language models
J. Kirchenbauer, J. Geiping, Y. Wen, M. Shu, K. Saifullah, K. Kong, K. Fernando, A. Saha, M. Goldblum, and T. Goldstein · 2023
Later among the works it cites.
Robust distortion-free watermarks for language models
R. Kuditipudi, J. Thickstun, T. Hashimoto, and P. Liang · 2023
Later among the works it cites.
GPT detectors are biased against non-native english writers
W. Liang, M. Yuksekgonul, Y. Mao, E. Wu, and J. Zou · 2023
Later among the works it cites.
Large language models challenge the future of higher education
S. Milano, J. A. McGrane, and S. Leonelli · 2023
Later among the works it cites.
Detectgpt: Zero-shot machine-generated text detection using probability curvature
E. Mitchell, Y. Lee, A. Khazatsky, C. D. Manning, and C. Finn · 2023
alphaXiv searches the wider corpus for related work and actual follow-ups.
alphaXiv is searching for related work…
Language models are few-shot learners
T. Brown, B. Mann, N. Ryder, M. Subbiah, J. D. Kaplan, P. Dhariwal, A. Neelakantan, P. Shyam, G. Sastry, A. Askell, et al · 2020
Cited alongside, same era.
Array programming with NumPy
C. R. Harris, K. J. Millman, S. J. Van Der Walt, R. Gommers, P. Virtanen, D. Cournapeau, E. Wieser, J. Taylor, S. Berg, N. J. Smith, et al · 2020
Cited alongside, same era.
Near-optimal algorithms for minimax optimization
T. Lin, C. Jin, and M. I. Jordan · 2020
Cited alongside, same era.
Exploring the limits of transfer learning with a unified text-to-text transformer
C. Raffel, N. Shazeer, A. Roberts, K. Lee, S. Narang, M. Matena, Y. Zhou, W. Li, and P. J. Liu · 2020
Cited alongside, same era.
An intensive introduction to cryptography, lectures notes for Harvard CS 127
B. Barak · 2021
Cited alongside, same era.
Ethical and social risks of harm from language models
L. Weidinger, J. Mellor, M. Rauh, C. Griffin, J. Uesato, P.-S. Huang, M. Cheng, M. Glaese, B. Balle, A. Kasirzadeh, et al · 2021
Cited alongside, same era.
Later among the works it cites.
ChatGPT: Optimizing language models for dialogue
OpenAI · 2023
Later among the works it cites.
Mark my words: Analyzing and evaluating language model watermarks
J. Piet, C. Sitawarin, V. Fang, N. Mu, and D. Wagner · 2023
Later among the works it cites.
Robust speech recognition via large-scale weak supervision
A. Radford, J. W. Kim, T. Xu, G. Brockman, C. McLeavey, and I. Sutskever · 2023
Later among the works it cites.
Can AI-generated text be reliably detected?
V. S. Sadasivan, A. Kumar, S. Balasubramanian, W. Wang, and S. Feizi · 2023
Later among the works it cites.
The curse of recursion: Training on generated data makes models forget
I. Shumailov, Z. Shumaylov, Y. Zhao, Y. Gal, N. Papernot, and R. Anderson · 2023
Later among the works it cites.
LLaMA: Open and efficient foundation language models
H. Touvron, T. Lavril, G. Izacard, X. Martinet, M.-A. Lachaux, T. Lacroix, B. Rozière, N. Goyal, E. Hambro, F. Azhar, et al · 2023
Later among the works it cites.
Testing of detection tools for AI-generated text
D. Weber-Wulff, A. Anohina-Naumeca, S. Bjelobaba, T. Foltỳnek, J. Guerrero-Dib, O. Popoola, P · 2023
Later among the works it cites.
DiPmark: A stealthy, efficient and resilient watermark for large language models
Y. Wu, Z. Hu, H. Zhang, and H. Huang · 2023
Later among the works it cites.
Sheared LLaMA: Accelerating language model pre-training via structured pruning
M. Xia, T. Gao, Z. Zeng, and D. Chen · 2023
Later among the works it cites.
DNA-GPT: Divergent n-gram analysis for training-free detection of GPT-generated text
X. Yang, W. Cheng, L. Petzold, W. Y. Wang, and H. Chen · 2023
Later among the works it cites.
ZeroGPT: Trusted GPT-4, ChatGPT and AI detector tool by ZeroGPT
ZeroGPT · 2023
Later among the works it cites.
Watermarks in the sand: Impossibility of strong watermarking for generative models
H. Zhang, B. L. Edelman, D. Francati, D. Venturi, G. Ateniese, and B. Barak · 2023
Later among the works it cites.
Towards better statistical understanding of watermarking LLMs
Z. Cai, S. Liu, H. Wang, H. Zhong, and X. Li · 2024
Closest in time.
Under the surface: Tracking the artifactuality of LLM-generated data
D. Das, K. De Langis, A. Martin, J. Kim, M. Lee, Z. M. Kim, S. Hayati, R. Owan, B. Hu, R. Parkar, et al · 2024
Closest in time.
WaterMax: breaking the LLM watermark detectability-robustness-quality trade-off
E. Giboulot and F. Teddy · 2024
Closest in time.
Paraphrasing evades detectors of AI-generated text, but retrieval is an effective defense
K. Krishna, Y. Song, M. Karpinska, J. Wieting, and M. Iyyer · 2024
Closest in time.
Robust detection of watermarks for large language models under human edits
X. Li, F. Ruan, H. Wang, Q. Long, and W. J. Su · 2024
Closest in time.
Adaptive text watermark for large language models
Y. Liu and Y. Bu · 2024
Closest in time.
Intrinsic dimension estimation for robust detection of AI-generated texts
E. Tulchinskii, K. Kuznetsov, L. Kushnareva, D. Cherniavskii, S. Nikolenko, E. Burnaev, S. Barannikov, and I. Piontkovskaya · 2024
Closest in time.
Debiasing watermarks for large language models via maximal coupling
Y. Xie, X. Li, T. Mallick, W. J. Su, and R. Zhang · 2024
Closest in time.
Provable robust watermarking for AI-generated text
X. Zhao, P. V. Ananth, L. Li, and Y.-X. Wang · 2024
Closest in time.
Permute-and-Flip: An optimally robust and watermarkable decoder for LLMs
X. Zhao, L. Li, and Y.-X. Wang · 2024
Closest in time.