id
string
sources
list
title
string
abstract
string
authors
list
categories
list
fields_of_study
list
published_date
timestamp[s]
url
string
pdf_url
string
arxiv_id
string
doi
string
citation_count
int64
influential_citation_count
int64
has_code
bool
code_url
string
venue
string
quality_score
float64
778030e3203f938a87b9c05a13afd15aa5c9078c49693d86abac7e19ed422dd5
[ "arxiv", "semantic_scholar" ]
CoheMark: A Novel Sentence-Level Watermark for Enhanced Text Quality
Watermarking technology is a method used to trace the usage of content generated by large language models. Sentence-level watermarking aids in preserving the semantic integrity within individual sentences while maintaining greater robustness. However, many existing sentence-level watermarking techniques depend on arbit...
[ "Junyan Zhang", "Shuliang Liu", "Aiwei Liu", "Yubo Gao", "Jungang Li", "Xiaojie Gu", "Xuming Hu" ]
[ "cs.CL" ]
[ "Computer Science" ]
2025-04-24T00:00:00
https://arxiv.org/abs/2504.17309
https://arxiv.org/pdf/2504.17309v1
2504.17309
10.48550/arXiv.2504.17309
17
0
false
null
arXiv.org
0.3138
7b15a966dc1b665f48b2edebb8b0b050830e6a73e22ca407e4b5f9791f179e20
[ "arxiv", "semantic_scholar" ]
Tracing Thought: Using Chain-of-Thought Reasoning to Identify the LLM Behind AI-Generated Text
In recent years, the detection of AI-generated text has become a critical area of research due to concerns about academic integrity, misinformation, and ethical AI deployment. This paper presents COT Fine-tuned, a novel framework for detecting AI-generated text and identifying the specific language model. responsible f...
[ "Shifali Agrahari", "Sanasam Ranbir Singh" ]
[ "cs.CL", "cs.AI" ]
[ "Computer Science" ]
2025-04-23T00:00:00
https://arxiv.org/abs/2504.16913
https://arxiv.org/pdf/2504.16913v1
2504.16913
10.48550/arXiv.2504.16913
4
0
false
null
arXiv.org
0.1747
2f7a97ca606d6c2d46039a513ef7be2c03f3294135933623f50f6a460ad90ef9
[ "arxiv", "semantic_scholar" ]
Certified Mitigation of Worst-Case LLM Copyright Infringement
The exposure of large language models (LLMs) to copyrighted material during pre-training raises concerns about unintentional copyright infringement post deployment. This has driven the development of "copyright takedown" methods, post-training approaches aimed at preventing models from generating content substantially ...
[ "Jingyu Zhang", "Jiacan Yu", "Marc Marone", "Benjamin Van Durme", "Daniel Khashabi" ]
[ "cs.CL" ]
[ "Computer Science" ]
2025-04-22T00:00:00
https://arxiv.org/abs/2504.16046
https://arxiv.org/pdf/2504.16046v2
2504.16046
10.48550/arXiv.2504.16046
2
0
false
null
Conference on Empirical Methods in Natural Language Processing
0.1341
179efa13e92998216b96e3eeed219b76d226bfe16d6c5a8404c762a8cdb5c0f2
[ "arxiv", "semantic_scholar" ]
Entropy-Guided Watermarking for LLMs: A Test-Time Framework for Robust and Traceable Text Generation
The rapid development of Large Language Models (LLMs) has intensified concerns about content traceability and potential misuse. Existing watermarking schemes for sampled text often face trade-offs between maintaining text quality and ensuring robust detection against various attacks. To address these issues, we propose...
[ "Shizhan Cai", "Liang Ding", "Dacheng Tao" ]
[ "cs.CL" ]
[ "Computer Science" ]
2025-04-16T00:00:00
https://arxiv.org/abs/2504.12108
https://arxiv.org/pdf/2504.12108v1
2504.12108
10.48550/arXiv.2504.12108
1
0
false
null
arXiv.org
0.1272
c97ed61161f2ad2c3f0c4982cefaa8358ada8a37f604d91c485edb2694226f65
[ "arxiv", "semantic_scholar" ]
Watermarking Needs Input Repetition Masking
Recent advancements in Large Language Models (LLMs) raised concerns over potential misuse, such as for spreading misinformation. In response two counter measures emerged: machine learning-based detectors that predict if text is synthetic, and LLM watermarking, which subtly marks generated text for identification and at...
[ "David Khachaturov", "Robert Mullins", "Ilia Shumailov", "Sumanth Dathathri" ]
[ "cs.LG", "cs.CL", "cs.CR" ]
[ "Computer Science" ]
2025-04-16T00:00:00
https://arxiv.org/abs/2504.12229
https://arxiv.org/pdf/2504.12229v1
2504.12229
10.48550/arXiv.2504.12229
0
0
false
null
arXiv.org
0.1272
5042af3ac013f1bb90a67f19461300048d8ac92112895e23465847a0ca4bcc76
[ "arxiv", "semantic_scholar" ]
Deep Audio Watermarks are Shallow: Limitations of Post-Hoc Watermarking Techniques for Speech
In the audio modality, state-of-the-art watermarking methods leverage deep neural networks to allow the embedding of human-imperceptible signatures in generated audio. The ideal is to embed signatures that can be detected with high accuracy when the watermarked audio is altered via compression, filtering, or other tran...
[ "Patrick O'Reilly", "Zeyu Jin", "Jiaqi Su", "Bryan Pardo" ]
[ "cs.SD", "eess.AS" ]
[ "Computer Science", "Engineering" ]
2025-04-15T00:00:00
https://arxiv.org/abs/2504.10782
https://arxiv.org/pdf/2504.10782v1
2504.10782
10.48550/arXiv.2504.10782
13
3
false
null
arXiv.org
0.301
3585b2e0595849b83dec182122f4e903ab4e407291fa63d6a1fe1ece7f7402b4
[ "arxiv", "semantic_scholar" ]
PT-Mark: Invisible Watermarking for Text-to-image Diffusion Models via Semantic-aware Pivotal Tuning
Watermarking for diffusion images has drawn considerable attention due to the widespread use of text-to-image diffusion models and the increasing need for their copyright protection. Recently, advanced watermarking techniques, such as Tree Ring, integrate watermarks by embedding traceable patterns (e.g., Rings) into th...
[ "Yaopeng Wang", "Huiyu Xu", "Zhibo Wang", "Jiacheng Du", "Zhichao Li", "Yiming Li", "Qiu Wang", "Kui Ren" ]
[ "cs.CR" ]
[ "Computer Science" ]
2025-04-15T00:00:00
https://arxiv.org/abs/2504.10853
https://arxiv.org/pdf/2504.10853v2
2504.10853
10.48550/arXiv.2504.10853
2
0
false
null
IEEE Transactions on Dependable and Secure Computing
0.126
e8d204959e84a9c1683c0c7cde69969d9ea84e7e932db487c92385f7b95f770c
[ "arxiv", "semantic_scholar" ]
OpenTuringBench: An Open-Model-based Benchmark and Framework for Machine-Generated Text Detection and Attribution
Open Large Language Models (OLLMs) are increasingly leveraged in generative AI applications, posing new challenges for detecting their outputs. We propose OpenTuringBench, a new benchmark based on OLLMs, designed to train and evaluate machine-generated text detectors on the Turing Test and Authorship Attribution proble...
[ "Lucio La Cava", "Andrea Tagarelli" ]
[ "cs.CL", "cs.AI", "cs.CY", "cs.HC", "physics.soc-ph" ]
[ "Computer Science", "Physics" ]
2025-04-15T00:00:00
https://arxiv.org/abs/2504.11369
https://arxiv.org/pdf/2504.11369v1
2504.11369
10.48550/arXiv.2504.11369
6
0
false
null
Conference on Empirical Methods in Natural Language Processing
0.2113
8309d676b7fc480394f73eb321ac9928918bd06af18dda08147d1034ec9a84e1
[ "arxiv", "semantic_scholar" ]
FractalForensics: Proactive Deepfake Detection and Localization via Fractal Watermarks
Proactive Deepfake detection via robust watermarks has seen interest ever since passive Deepfake detectors encountered challenges in identifying high-quality synthetic images. However, while demonstrating reasonable detection performance, they lack localization functionality and explainability in detection results. Add...
[ "Tianyi Wang", "Harry Cheng", "Ming-Hui Liu", "Mohan Kankanhalli" ]
[ "cs.CV" ]
[ "Computer Science" ]
2025-04-13T00:00:00
https://arxiv.org/abs/2504.09451
https://arxiv.org/pdf/2504.09451v2
2504.09451
10.1145/3746027.3754544
11
0
false
null
ACM Multimedia
0.2698
7b1c098849718a962e5d228a4f7d38dfa50e8273c0f35fe93b64a0111a157ab1
[ "arxiv", "semantic_scholar" ]
Defending LLM Watermarking Against Spoofing Attacks with Contrastive Representation Learning
Watermarking has emerged as a promising technique for detecting texts generated by LLMs. Current research has primarily focused on three design criteria: high quality of the watermarked text, high detectability, and robustness against removal attack. However, the security against spoofing attacks remains relatively und...
[ "Li An", "Yujian Liu", "Yepeng Liu", "Yang Zhang", "Yuheng Bu", "Shiyu Chang" ]
[ "cs.CR", "cs.CL" ]
[ "Computer Science" ]
2025-04-09T00:00:00
https://arxiv.org/abs/2504.06575
https://arxiv.org/pdf/2504.06575v2
2504.06575
10.48550/arXiv.2504.06575
10
0
true
https://github.com/UCSB-NLP-Chang/contrastive-watermark
arXiv.org
0.2603
88f04e47077a1a5556e8709b89485451a679021b0cea91ce916f027fafa2d527
[ "arxiv", "semantic_scholar" ]
DeCoMa: Detecting and Purifying Code Dataset Watermarks through Dual Channel Code Abstraction
Watermarking is a technique to help identify the source of data points, which can be used to help prevent the misuse of protected datasets. Existing methods on code watermarking, leveraging the idea from the backdoor research, embed stealthy triggers as watermarks. Despite their high resilience against dilution attacks...
[ "Yuan Xiao", "Yuchen Chen", "Shiqing Ma", "Haocheng Huang", "Chunrong Fang", "Yanwei Chen", "Weisong Sun", "Yunfeng Zhu", "Xiaofang Zhang", "Zhenyu Chen" ]
[ "cs.CR", "cs.SE" ]
[ "Computer Science" ]
2025-04-09T00:00:00
https://arxiv.org/abs/2504.07002
https://arxiv.org/pdf/2504.07002v2
2504.07002
10.1145/3728952
6
0
true
https://github.com/xiaoyuanpigo/DeCoMa
null
0.2113
acdea3d4c6d07bfaebc5ab2f43be81257a325da4c01b8f9cfaa8a8ef1413dbc5
[ "arxiv", "semantic_scholar" ]
Can you Finetune your Binoculars? Embedding Text Watermarks into the Weights of Large Language Models
The indistinguishability of AI-generated content from human text raises challenges in transparency and accountability. While several methods exist to watermark models behind APIs, embedding watermark strategies directly into model weights that are later reflected in the outputs of the model is challenging. In this stud...
[ "Fay Elhassan", "Niccolò Ajroldi", "Antonio Orvieto", "Jonas Geiping" ]
[ "cs.LG", "cs.AI" ]
[ "Computer Science" ]
2025-04-08T00:00:00
https://arxiv.org/abs/2504.06446
https://arxiv.org/pdf/2504.06446v1
2504.06446
10.48550/arXiv.2504.06446
3
0
false
null
arXiv.org
0.1505
e3c9e42c9bec62cfb22e4ae95ceb67f07e291805fd0634a96b49a7ab1413e287
[ "arxiv", "semantic_scholar" ]
Detection Limits and Statistical Separability of Tree Ring Watermarks in Rectified Flow-based Text-to-Image Generation Models
Tree-Ring Watermarking is a significant technique for authenticating AI-generated images. However, its effectiveness in rectified flow-based models remains unexplored, particularly given the inherent challenges of these models with noise latent inversion. Through extensive experimentation, we evaluated and compared the...
[ "Ved Umrajkar", "Aakash Kumar Singh" ]
[ "cs.CV", "cs.AI", "cs.CR", "cs.LG", "stat.ML" ]
[ "Computer Science", "Mathematics" ]
2025-04-04T00:00:00
https://arxiv.org/abs/2504.03850
https://arxiv.org/pdf/2504.03850v1
2504.03850
10.48550/arXiv.2504.03850
0
0
true
https://github.com/dsgiitr/flux-watermarking}{\textbf{link}}
arXiv.org
0.1753
12411a5f5d1cfce249e6a33cf00bb1583cda3f4cca00bfef47ffbf31773a7981
[ "arxiv", "semantic_scholar" ]
FontGuard: A Robust Font Watermarking Approach Leveraging Deep Font Knowledge
The proliferation of AI-generated content brings significant concerns on the forensic and security issues such as source tracing, copyright protection, etc, highlighting the need for effective watermarking technologies. Font-based text watermarking has emerged as an effective solution to embed information, which could ...
[ "Kahim Wong", "Jicheng Zhou", "Kemou Li", "Yain-Whar Si", "Xiaowei Wu", "Jiantao Zhou" ]
[ "cs.CV" ]
[ "Computer Science" ]
2025-04-04T00:00:00
https://arxiv.org/abs/2504.03128
https://arxiv.org/pdf/2504.03128v1
2504.03128
10.1109/TMM.2025.3604908
3
0
true
https://github.com/KAHIMWONG/FontGuard
IEEE transactions on multimedia
0.1753
82edbec16fa4114262b1378f062aa75da0b6560b22f3a9b8ab1b03c547166ebb
[ "arxiv", "semantic_scholar" ]
RoSMM: A Robust and Secure Multi-Modal Watermarking Framework for Diffusion Models
Current image watermarking technologies are predominantly categorized into text watermarking techniques and image steganography; however, few methods can simultaneously handle text and image-based watermark data, which limits their applicability in complex digital environments. This paper introduces an innovative multi...
[ "ZhongLi Fang", "Yu Xie", "Ping Chen" ]
[ "cs.MM" ]
[ "Computer Science" ]
2025-04-03T00:00:00
https://arxiv.org/abs/2504.02640
https://arxiv.org/pdf/2504.02640v2
2504.02640
10.48550/arXiv.2504.02640
1
0
false
null
arXiv.org
0.1123
a588235413e0feab72976d9c29357112fe3419d3e3417cb50496e2dbc73e2f7f
[ "arxiv", "semantic_scholar" ]
Watermarking for AI Content Detection: A Review on Text, Visual, and Audio Modalities
The rapid advancement of generative artificial intelligence (GenAI) has revolutionized content creation across text, visual, and audio domains, simultaneously introducing significant risks such as misinformation, identity fraud, and content manipulation. This paper presents a practical survey of watermarking techniques...
[ "Lele Cao" ]
[ "cs.CR" ]
[ "Computer Science" ]
2025-04-02T00:00:00
https://arxiv.org/abs/2504.03765
https://arxiv.org/pdf/2504.03765v1
2504.03765
10.48550/arXiv.2504.03765
7
0
false
null
arXiv.org
0.2258
13baafc28ce03dcb5a3003fcdba71a792a1eaf15a3504924cb2c25efa63755ca
[ "arxiv", "semantic_scholar" ]
Understanding the Effects of RLHF on the Quality and Detectability of LLM-Generated Texts
Large Language Models (LLMs) have demonstrated exceptional performance on a range of downstream NLP tasks by generating text that closely resembles human writing. However, the ease of achieving this similarity raises concerns from potential malicious uses at scale by bad actors, as LLM-generated text becomes increasing...
[ "Beining Xu", "Arkaitz Zubiaga" ]
[ "cs.CL", "cs.AI" ]
[ "Computer Science" ]
2025-03-23T00:00:00
https://arxiv.org/abs/2503.17965
https://arxiv.org/pdf/2503.17965v1
2503.17965
10.48550/arXiv.2503.17965
3
2
false
null
arXiv.org
0.2386
ba550fad2988da3213f757a516da11a5b591a39177cfdf6f31aeb2cbb6ddcefa
[ "arxiv", "semantic_scholar" ]
Text-Guided Image Invariant Feature Learning for Robust Image Watermarking
Ensuring robustness in image watermarking is crucial for and maintaining content integrity under diverse transformations. Recent self-supervised learning (SSL) approaches, such as DINO, have been leveraged for watermarking but primarily focus on general feature representation rather than explicitly learning invariant f...
[ "Muhammad Ahtesham", "Xin Zhong" ]
[ "cs.CV", "cs.LG", "cs.MM" ]
[ "Computer Science" ]
2025-03-18T00:00:00
https://arxiv.org/abs/2503.13805
https://arxiv.org/pdf/2503.13805v1
2503.13805
10.1109/MIPR67560.2025.00033
2
0
false
null
Conference on Multimedia Information Processing and Retrieval
0.1193
e0a46955eff0a40603425abdc87ca5dca4a5bb8f35050b8c0b05d27f2afdcd64
[ "arxiv", "semantic_scholar" ]
Your Text Encoder Can Be An Object-Level Watermarking Controller
Invisible watermarking of AI-generated images can help with copyright protection, enabling detection and identification of AI-generated media. In this work, we present a novel approach to watermark images of T2I Latent Diffusion Models (LDMs). By only fine-tuning text token embeddings $W_*$, we enable watermarking in s...
[ "Naresh Kumar Devulapally", "Mingzhen Huang", "Vishal Asnani", "Shruti Agarwal", "Siwei Lyu", "Vishnu Suresh Lokhande" ]
[ "cs.CV", "cs.CR", "cs.LG" ]
[ "Computer Science" ]
2025-03-15T00:00:00
https://arxiv.org/abs/2503.11945
https://arxiv.org/pdf/2503.11945v1
2503.11945
10.1109/ICCV51701.2025.01539
1
0
false
null
IEEE International Conference on Computer Vision
0.0905
81cd6cb49bede85aa59c39128cf773d92a87d0e38cdac74fc899a25d6d075b33
[ "arxiv", "semantic_scholar" ]
Safe-VAR: Safe Visual Autoregressive Model for Text-to-Image Generative Watermarking
With the success of autoregressive learning in large language models, it has become a dominant approach for text-to-image generation, offering high efficiency and visual quality. However, invisible watermarking for visual autoregressive (VAR) models remains underexplored, despite its importance in misuse prevention. Ex...
[ "Ziyi Wang", "Songbai Tan", "Gang Xu", "Xuerui Qiu", "Hongbin Xu", "Xin Meng", "Ming Li", "Fei Richard Yu" ]
[ "cs.MM", "cs.CV", "eess.IV" ]
[ "Computer Science", "Engineering" ]
2025-03-14T00:00:00
https://arxiv.org/abs/2503.11324
https://arxiv.org/pdf/2503.11324v1
2503.11324
10.48550/arXiv.2503.11324
3
0
false
null
arXiv.org
0.1505
1725e205d4ae5488bac5927e2231d9fe9eb29a0acd3819a81a862325281ad22b
[ "arxiv", "semantic_scholar" ]
Pathology-Aware Adaptive Watermarking for Text-Driven Medical Image Synthesis
As recent text-conditioned diffusion models have enabled the generation of high-quality images, concerns over their potential misuse have also grown. This issue is critical in the medical domain, where text-conditioned generated medical images could enable insurance fraud or falsified records, highlighting the urgent n...
[ "Chanyoung Kim", "Dayun Ju", "Jinyeong Kim", "Woojung Han", "Roberto Alcover-Couso", "Seong Jae Hwang" ]
[ "cs.CV" ]
[ "Computer Science" ]
2025-03-11T00:00:00
https://arxiv.org/abs/2503.08346
https://arxiv.org/pdf/2503.08346v2
2503.08346
10.48550/arXiv.2503.08346
1
0
false
null
International Conference on Medical Image Computing and Computer-Assisted Intervention
0.0859
c7654a0ee51f054f15dd168a4642194c9293349ebe102b36b252eb30a89be525
[ "arxiv", "semantic_scholar" ]
Fine-Grained Bias Detection in LLM: Enhancing detection mechanisms for nuanced biases
Recent advancements in Artificial Intelligence, particularly in Large Language Models (LLMs), have transformed natural language processing by improving generative capabilities. However, detecting biases embedded within these models remains a challenge. Subtle biases can propagate misinformation, influence decision-maki...
[ "Suvendu Mohanty" ]
[ "cs.CL", "cs.AI" ]
[ "Computer Science" ]
2025-03-08T00:00:00
https://arxiv.org/abs/2503.06054
https://arxiv.org/pdf/2503.06054v1
2503.06054
10.48550/arXiv.2503.06054
3
1
false
null
arXiv.org
0.1505
3152585a56521600498c3c475abd496ab75a2ffd8619a479c4685fe673ceb2f2
[ "arxiv", "semantic_scholar" ]
Mark Your LLM: Detecting the Misuse of Open-Source Large Language Models via Watermarking
As open-source large language models (LLMs) like Llama3 become more capable, it is crucial to develop watermarking techniques to detect their potential misuse. Existing watermarking methods either add watermarks during LLM inference, which is unsuitable for open-source LLMs, or primarily target classification LLMs rath...
[ "Yijie Xu", "Aiwei Liu", "Xuming Hu", "Lijie Wen", "Hui Xiong" ]
[ "cs.CL", "cs.AI", "cs.CR", "cs.LG" ]
[ "Computer Science" ]
2025-03-06T00:00:00
https://arxiv.org/abs/2503.04636
https://arxiv.org/pdf/2503.04636v2
2503.04636
10.48550/arXiv.2503.04636
16
1
true
null
arXiv.org
0.3076
60291a583b9767bcc2703b4c375f51a6d7ff0af813839449fd5ae9e359799474
[ "arxiv", "semantic_scholar" ]
Robust Data Watermarking in Language Models by Injecting Fictitious Knowledge
Data watermarking in language models injects traceable signals, such as specific token sequences or stylistic patterns, into copyrighted text, allowing copyright holders to track and verify training data ownership. Previous data watermarking techniques primarily focus on effective memorization during pretraining, while...
[ "Xinyue Cui", "Johnny Tian-Zheng Wei", "Swabha Swayamdipta", "Robin Jia" ]
[ "cs.CR", "cs.CL", "cs.LG" ]
[ "Computer Science" ]
2025-03-06T00:00:00
https://arxiv.org/abs/2503.04036
https://arxiv.org/pdf/2503.04036v3
2503.04036
10.48550/arXiv.2503.04036
13
1
false
null
Annual Meeting of the Association for Computational Linguistics
0.2865
3112ef497ced5cab3aa67aa702a57ff39884b986784b726a15e65c9fbb1e03b8
[ "arxiv", "semantic_scholar" ]
GaussianSeal: Rooting Adaptive Watermarks for 3D Gaussian Generation Model
With the advancement of AIGC technologies, the modalities generated by models have expanded from images and videos to 3D objects, leading to an increasing number of works focused on 3D Gaussian Splatting (3DGS) generative models. Existing research on copyright protection for generative models has primarily concentrated...
[ "Runyi Li", "Xuanyu Zhang", "Chuhan Tong", "Zhipei Xu", "Jian Zhang" ]
[ "cs.CV", "eess.IV" ]
[ "Computer Science", "Engineering" ]
2025-03-01T00:00:00
https://arxiv.org/abs/2503.00531
https://arxiv.org/pdf/2503.00531v2
2503.00531
10.1007/s11633-025-1588-7
6
1
false
null
Machine Intelligence Research
0.2113
4bb7556861c870f69b4105cccc0f7f3ecbdf41732041a7bffb353b2a8a016973
[ "arxiv", "semantic_scholar" ]
Marking Code Without Breaking It: Code Watermarking for Detecting LLM-Generated Code
Identifying LLM-generated code through watermarking poses a challenge in preserving functional correctness. Previous methods rely on the assumption that watermarking high-entropy tokens effectively maintains output quality. Our analysis reveals a fundamental limitation of this assumption: syntax-critical tokens such as...
[ "Jungin Kim", "Shinwoo Park", "Yo-Sub Han" ]
[ "cs.CR", "cs.AI" ]
[ "Computer Science" ]
2025-02-26T00:00:00
https://arxiv.org/abs/2502.18851
https://arxiv.org/pdf/2502.18851v4
2502.18851
10.48550/arXiv.2502.18851
10
1
true
https://github.com/inistory/STONE-watermarking
Conference of the European Chapter of the Association for Computational Linguistics
0.2603
a13cf33b09a72ab14a076555593c120fb9fcedcba1fe62f540085ea38ba044fe
[ "arxiv", "semantic_scholar" ]
Project Alexandria: Towards Freeing Scientific Knowledge from Copyright Burdens via LLMs
Paywalls, licenses and copyright rules often restrict the broad dissemination and reuse of scientific knowledge. We take the position that it is both legally and technically feasible to extract the scientific knowledge in scholarly texts. Current methods, like text embeddings, fail to reliably preserve factual content,...
[ "Christoph Schuhmann", "Gollam Rabby", "Ameya Prabhu", "Tawsif Ahmed", "Andreas Hochlehnert", "Huu Nguyen", "Nick Akinci", "Ludwig Schmidt", "Robert Kaczmarczyk", "Sören Auer", "Jenia Jitsev", "Matthias Bethge" ]
[ "cs.LG", "cs.AI", "cs.CL" ]
[ "Computer Science" ]
2025-02-26T00:00:00
https://arxiv.org/abs/2502.19413
https://arxiv.org/pdf/2502.19413v2
2502.19413
10.48550/arXiv.2502.19413
0
0
true
null
arXiv.org
0.1098
911a29ad83e3015400988af2d4450ab33fc34b66c2485708c05277511180f707
[ "arxiv", "semantic_scholar" ]
Breaking Distortion-free Watermarks in Large Language Models
In recent years, LLM watermarking has emerged as an attractive safeguard against AI-generated content, with promising applications in many real-world domains. However, there are growing concerns that the current LLM watermarking schemes are vulnerable to expert adversaries wishing to reverse-engineer the watermarking m...
[ "Shayleen Reynolds", "Hengzhi He", "Dung Daniel T. Ngo", "Saheed Obitayo", "Niccolò Dalmasso", "Guang Cheng", "Vamsi K. Potluru", "Manuela Veloso" ]
[ "cs.CR", "cs.LG" ]
[ "Computer Science" ]
2025-02-25T00:00:00
https://arxiv.org/abs/2502.18608
https://arxiv.org/pdf/2502.18608v2
2502.18608
null
0
0
false
null
null
0.0445
7c317b94e80ce106b766d8a72fe0b379d76bcf06a515e688b72bacde3591c254
[ "arxiv", "semantic_scholar" ]
KatFishNet: Detecting LLM-Generated Korean Text through Linguistic Feature Analysis
The rapid advancement of large language models (LLMs) increases the difficulty of distinguishing between human-written and LLM-generated text. Detecting LLM-generated text is crucial for upholding academic integrity, preventing plagiarism, protecting copyrights, and ensuring ethical research practices. Most prior studi...
[ "Shinwoo Park", "Shubin Kim", "Do-Kyung Kim", "Yo-Sub Han" ]
[ "cs.CL", "cs.AI" ]
[ "Computer Science" ]
2025-02-25T00:00:00
https://arxiv.org/abs/2503.00032
https://arxiv.org/pdf/2503.00032v5
2503.00032
10.18653/v1/2025.acl-long.1030
6
0
true
https://github.com/Shinwoo-Park/detecting_llm_generated_korean_text_through_linguistic_analysis
Annual Meeting of the Association for Computational Linguistics
0.2113
5d9e547dc733232d337c1572ef71f7990ad799a555b2e1cc54e966bde66130a3
[ "arxiv", "semantic_scholar" ]
Sarang at DEFACTIFY 4.0: Detecting AI-Generated Text Using Noised Data and an Ensemble of DeBERTa Models
This paper presents an effective approach to detect AI-generated text, developed for the Defactify 4.0 shared task at the fourth workshop on multimodal fact checking and hate speech detection. The task consists of two subtasks: Task-A, classifying whether a text is AI generated or human written, and Task-B, classifying...
[ "Avinash Trivedi", "Sangeetha Sivanesan" ]
[ "cs.CL", "cs.AI" ]
[ "Computer Science" ]
2025-02-24T00:00:00
https://arxiv.org/abs/2502.16857
https://arxiv.org/pdf/2502.16857v1
2502.16857
10.48550/arXiv.2502.16857
2
1
false
null
arXiv.org
0.1505
e9881846358212cd5eba830788dd58286b3afe3cd8039e07751c24c48541d819
[ "arxiv", "semantic_scholar" ]
Detecting Benchmark Contamination Through Watermarking
Benchmark contamination poses a significant challenge to the reliability of Large Language Models (LLMs) evaluations, as it is difficult to assert whether a model has been trained on a test set. We introduce a solution to this problem by watermarking benchmarks before their release. The embedding involves reformulating...
[ "Tom Sander", "Pierre Fernandez", "Saeed Mahloujifar", "Alain Durmus", "Chuan Guo" ]
[ "cs.CR", "cs.AI" ]
[ "Computer Science" ]
2025-02-24T00:00:00
https://arxiv.org/abs/2502.17259
https://arxiv.org/pdf/2502.17259v2
2502.17259
10.48550/arXiv.2502.17259
6
1
false
null
arXiv.org
0.2113
93561b61cd89722e701d8348cc33e068e383b8c040a2a6a711a502c4eebba46a
[ "arxiv", "semantic_scholar" ]
Evaluating the Robustness and Accuracy of Text Watermarking Under Real-World Cross-Lingual Manipulations
We present a study to benchmark representative watermarking methods in cross-lingual settings. The current literature mainly focuses on the evaluation of watermarking methods for the English language. However, the literature for evaluating watermarking in cross-lingual settings is scarce. This results in overlooking im...
[ "Mansour Al Ghanim", "Jiaqi Xue", "Rochana Prih Hastuti", "Mengxin Zheng", "Yan Solihin", "Qian Lou" ]
[ "cs.CL", "cs.LG" ]
[ "Computer Science" ]
2025-02-23T00:00:00
https://arxiv.org/abs/2502.16699
https://arxiv.org/pdf/2502.16699v2
2502.16699
10.18653/v1/2025.findings-emnlp.390
5
0
false
null
Conference on Empirical Methods in Natural Language Processing
0.1945
a568b9cf5592164168e15e729a43cb453cdd3886a661c817d34112a81a7cd8bd
[ "arxiv", "semantic_scholar" ]
Can Large Vision-Language Models Detect Images Copyright Infringement from GenAI?
Generative AI models, renowned for their ability to synthesize high-quality content, have sparked growing concerns over the improper generation of copyright-protected material. While recent studies have proposed various approaches to address copyright issues, the capability of large vision-language models (LVLMs) to de...
[ "Qipan Xu", "Zhenting Wang", "Xiaoxiao He", "Ligong Han", "Ruixiang Tang" ]
[ "cs.CV", "cs.AI", "cs.CL" ]
[ "Computer Science" ]
2025-02-23T00:00:00
https://arxiv.org/abs/2502.16618
https://arxiv.org/pdf/2502.16618v1
2502.16618
10.48550/arXiv.2502.16618
6
0
false
null
arXiv.org
0.2113
9ddb812dede8ec2980229b40cd762adb6776698d58fbf38e73d3e587f043cf28
[ "arxiv", "semantic_scholar" ]
Deep Learning-based Dual Watermarking for Image Copyright Protection and Authentication
Advancements in digital technologies make it easy to modify the content of digital images. Hence, ensuring digital images integrity and authenticity is necessary to protect them against various attacks that manipulate them. We present a Deep Learning (DL) based dual invisible watermarking technique for performing sourc...
[ "Sudev Kumar Padhi", "Archana Tiwari", "Sk. Subidh Ali" ]
[ "cs.CR", "cs.AI" ]
[ "Computer Science" ]
2025-02-21T00:00:00
https://arxiv.org/abs/2502.18501
https://arxiv.org/pdf/2502.18501v1
2502.18501
10.1109/TAI.2024.3485519
15
3
false
null
IEEE Transactions on Artificial Intelligence
0.301
f6a795b0eef80fdfcde154c84382a6de5f1ce07209feb9d352eec4c881b1f2ed
[ "arxiv", "semantic_scholar" ]
Can LLM Watermarks Robustly Prevent Unauthorized Knowledge Distillation?
The radioactive nature of Large Language Model (LLM) watermarking enables the detection of watermarks inherited by student models when trained on the outputs of watermarked teacher models, making it a promising tool for preventing unauthorized knowledge distillation. However, the robustness of watermark radioactivity a...
[ "Leyi Pan", "Aiwei Liu", "Shiyu Huang", "Yijian Lu", "Xuming Hu", "Lijie Wen", "Irwin King", "Philip S. Yu" ]
[ "cs.CL" ]
[ "Computer Science" ]
2025-02-17T00:00:00
https://arxiv.org/abs/2502.11598
https://arxiv.org/pdf/2502.11598v2
2502.11598
10.48550/arXiv.2502.11598
11
2
true
https://github.com/THU-BPM/Watermark-Radioactivity-Attack
Annual Meeting of the Association for Computational Linguistics
0.2698
0bcc403ba94d25a81fab5c5b246e6a1f5580365057307b62129748fcc46b6fcd
[ "arxiv", "semantic_scholar" ]
LLM-driven Knowledge Distillation for Dynamic Text-Attributed Graphs
Dynamic Text-Attributed Graphs (DyTAGs) have numerous real-world applications, e.g. social, collaboration, citation, communication, and review networks. In these networks, nodes and edges often contain text descriptions, and the graph structure can evolve over time. Future link prediction, edge classification, relation...
[ "Amit Roy", "Ning Yan", "Masood Mortazavi" ]
[ "cs.LG" ]
[ "Computer Science" ]
2025-02-15T00:00:00
https://arxiv.org/abs/2502.10914
https://arxiv.org/pdf/2502.10914v1
2502.10914
10.48550/arXiv.2502.10914
3
1
false
null
arXiv.org
0.1505
373b2bbbb09b8d5948716fad7ac271a38909364ec3f98327ea697ee80a589a34
[ "arxiv", "semantic_scholar" ]
Beyond English: Unveiling Multilingual Bias in LLM Copyright Compliance
Large Language Models (LLMs) have raised significant concerns regarding the fair use of copyright-protected content. While prior studies have examined the extent to which LLMs reproduce copyrighted materials, they have predominantly focused on English, neglecting multilingual dimensions of copyright protection. In this...
[ "Yupeng Chen", "Xiaoyu Zhang", "Yixian Huang", "Qian Xie" ]
[ "cs.CY", "cs.CL" ]
[ "Computer Science" ]
2025-02-14T00:00:00
https://arxiv.org/abs/2503.05713
https://arxiv.org/pdf/2503.05713v1
2503.05713
10.48550/arXiv.2503.05713
2
1
false
null
arXiv.org
0.1505
18f91aa434b29d7231cf814397d83a93eaf01906a1441677b5d70c95afa5d81b
[ "arxiv", "semantic_scholar" ]
Towards Watermarking of Open-Source LLMs
While watermarks for closed LLMs have matured and have been included in large-scale deployments, these methods are not applicable to open-source models, which allow users full control over the decoding process. This setting is understudied yet critical, given the rising performance of open-source models. In this work, ...
[ "Thibaud Gloaguen", "Nikola Jovanović", "Robin Staab", "Martin Vechev" ]
[ "cs.CR", "cs.LG" ]
[ "Computer Science" ]
2025-02-14T00:00:00
https://arxiv.org/abs/2502.10525
https://arxiv.org/pdf/2502.10525v1
2502.10525
10.48550/arXiv.2502.10525
13
2
true
null
arXiv.org
0.2865
1168ed5125c8d155ff3765a6ca48519fdf9df49ce1553f316b68550d89e55250
[ "arxiv", "semantic_scholar" ]
GraphiT: Efficient Node Classification on Text-Attributed Graphs with Prompt Optimized LLMs
The application of large language models (LLMs) to graph data has attracted a lot of attention recently. LLMs allow us to use deep contextual embeddings from pretrained models in text-attributed graphs, where shallow embeddings are often used for the text attributes of nodes. However, it is still challenging to efficie...
[ "Shima Khoshraftar", "Niaz Abedini", "Amir Hajian" ]
[ "cs.AI", "cs.LG" ]
[ "Computer Science" ]
2025-02-14T00:00:00
https://arxiv.org/abs/2502.10522
https://arxiv.org/pdf/2502.10522v1
2502.10522
10.1145/3701716.3717805
6
0
false
null
The Web Conference
0.2113
26c9934ca47ad9f82ff75211420c34d5a253e2fa6d25459e40c15d5e75af1865
[ "arxiv", "semantic_scholar" ]
Modification and Generated-Text Detection: Achieving Dual Detection Capabilities for the Outputs of LLM by Watermark
The development of large language models (LLMs) has raised concerns about potential misuse. One practical solution is to embed a watermark in the text, allowing ownership verification through watermark extraction. Existing methods primarily focus on defending against modification attacks, often neglecting other spoofin...
[ "Yuhang Cai", "Yaofei Wang", "Donghui Hu", "Chen Gu" ]
[ "cs.CR", "cs.AI" ]
[ "Computer Science" ]
2025-02-12T00:00:00
https://arxiv.org/abs/2502.08332
https://arxiv.org/pdf/2502.08332v2
2502.08332
10.48550/arXiv.2502.08332
0
0
false
null
arXiv.org
0.055
386c02f986c25caee7b73a6cbfa12a71ff5c1122666b3117480dc4523fd477dd
[ "arxiv", "semantic_scholar" ]
Image Watermarking of Generative Diffusion Models
Embedding watermarks into the output of generative models is essential for establishing copyright and verifiable ownership over the generated content. Emerging diffusion model watermarking methods either embed watermarks in the frequency domain or offer limited versatility of the watermark patterns in the image space, ...
[ "Yunzhuo Chen", "Jordan Vice", "Naveed Akhtar", "Nur Al Hasan Haldar", "Ajmal Mian" ]
[ "eess.IV", "cs.CR" ]
[ "Computer Science", "Engineering" ]
2025-02-12T00:00:00
https://arxiv.org/abs/2502.10465
https://arxiv.org/pdf/2502.10465v1
2502.10465
10.48550/arXiv.2502.10465
4
0
false
null
arXiv.org
0.1747
3c3207a225f2517ce2db36c33dbcf1fffea4839ff2cfab15e6868c835dacce40
[ "arxiv", "semantic_scholar" ]
Toward Copyright Integrity and Verifiability via Multi-Bit Watermarking for Intelligent Transportation Systems
Intelligent transportation systems (ITS) use advanced technologies such as artificial intelligence to significantly improve traffic flow management efficiency, and promote the intelligent development of the transportation industry. However, if the data in ITS is attacked, such as tampering or forgery, it will endanger ...
[ "Yihao Wang", "Lingxiao Li", "Yifan Tang", "Ru Zhang", "Jianyi Liu" ]
[ "cs.CR", "cs.CL" ]
[ "Computer Science" ]
2025-02-08T00:00:00
https://arxiv.org/abs/2502.05425
https://arxiv.org/pdf/2502.05425v1
2502.05425
10.1109/TITS.2025.3535932
8
0
false
null
IEEE Transactions on Intelligent Transportation Systems, 07 February 2025
0.2386
c773caadfd3413931be7eb72d4b28ba5a9f34c3be5b226de1ef3155738974606
[ "arxiv", "semantic_scholar" ]
XAttnMark: Learning Robust Audio Watermarking with Cross-Attention
The rapid proliferation of generative audio synthesis and editing technologies has raised serious concerns about copyright infringement, data provenance, and the spread of misinformation via deepfake audio. Watermarking offers a proactive solution by embedding imperceptible yet identifiable and traceable signals into a...
[ "Yixin Liu", "Lie Lu", "Jihui Jin", "Lichao Sun", "Andrea Fanelli" ]
[ "cs.SD", "cs.AI", "cs.CR", "cs.LG", "eess.AS" ]
[ "Computer Science", "Engineering" ]
2025-02-06T00:00:00
https://arxiv.org/abs/2502.04230
https://arxiv.org/pdf/2502.04230v3
2502.04230
10.48550/arXiv.2502.04230
10
0
false
null
International Conference on Machine Learning
0.2603
241a3d5c10d5a90a8697cc5531181eea04148e2fd1f9a4e6758a051f926129cc
[ "arxiv", "semantic_scholar" ]
DERMARK: A Dynamic, Efficient and Robust Multi-bit Watermark for Large Language Models
As large language models (LLMs) grow more powerful, concerns over copyright infringement of LLM-generated texts have intensified. LLM watermarking has been proposed to trace unauthorized redistribution or resale of generated content by embedding identifiers within the text. Existing approaches primarily rely on one-bit...
[ "Qihao Lin", "Chen Tang", "Lan zhang", "Junyang zhang", "Xiangyang Li" ]
[ "cs.CR", "cs.AI" ]
[ "Computer Science" ]
2025-02-04T00:00:00
https://arxiv.org/abs/2502.05213
https://arxiv.org/pdf/2502.05213v2
2502.05213
10.48550/arXiv.2502.05213
0
0
false
null
arXiv.org
0.0458
81552158272fe9ae25c2f7a0c5550286d33224f88cc8b3307a0ea34de64fd63f
[ "arxiv", "semantic_scholar" ]
Watermarking across Modalities for Content Tracing and Generative AI
Watermarking embeds information into digital content like images, audio, or text, imperceptible to humans but robustly detectable by specific algorithms. This technology has important applications in many challenges of the industry such as content moderation, tracing AI-generated content, and monitoring the usage of AI...
[ "Pierre Fernandez" ]
[ "cs.CR", "cs.AI", "cs.LG" ]
[ "Computer Science" ]
2025-02-04T00:00:00
https://arxiv.org/abs/2502.05215
https://arxiv.org/pdf/2502.05215v1
2502.05215
10.48550/arXiv.2502.05215
2
0
false
null
arXiv.org
0.1193
ea3d1d1db971575665d74d9c090a66547116d25fb6c9519fc5972e97c1134041
[ "arxiv", "semantic_scholar" ]
BiMarker: Enhancing Text Watermark Detection for Large Language Models with Bipolar Watermarks
The rapid growth of Large Language Models (LLMs) raises concerns about distinguishing AI-generated text from human content. Existing watermarking techniques, like \kgw, struggle with low watermark strength and stringent false-positive requirements. Our analysis reveals that current methods rely on coarse estimates of n...
[ "Zhuang Li", "Qiuping Yi", "Zongcheng Ji", "Yijian Lu", "Yanqi Li", "Keyang Xiao", "Hongliang Liang" ]
[ "cs.LG" ]
[ "Computer Science" ]
2025-01-21T00:00:00
https://arxiv.org/abs/2501.12174
https://arxiv.org/pdf/2501.12174v6
2501.12174
10.48550/arXiv.2501.12174
2
0
false
null
Neurocomputing
0.1193
b077655bc0d6a824070feaa58ee24489d22fe1e739f2af171fa7e4d260a69e73
[ "arxiv", "semantic_scholar" ]
GaussMark: A Practical Approach for Structural Watermarking of Language Models
Recent advances in Large Language Models (LLMs) have led to significant improvements in natural language processing tasks, but their ability to generate human-quality text raises significant ethical and operational concerns in settings where it is important to recognize whether or not a given text was generated by a hu...
[ "Adam Block", "Ayush Sekhari", "Alexander Rakhlin" ]
[ "cs.CR", "cs.AI", "cs.CL", "cs.LG" ]
[ "Computer Science" ]
2025-01-17T00:00:00
https://arxiv.org/abs/2501.13941
https://arxiv.org/pdf/2501.13941v1
2501.13941
10.48550/arXiv.2501.13941
11
4
false
null
International Conference on Machine Learning
0.3495
c41ca6eec673179131c842cb9dec4522e1c6e9ddd2fe69e02681fe702cb92aaa
[ "arxiv", "semantic_scholar" ]
SoccerSynth-Detection: A Synthetic Dataset for Soccer Player Detection
In soccer video analysis, player detection is essential for identifying key events and reconstructing tactical positions. The presence of numerous players and frequent occlusions, combined with copyright restrictions, severely restricts the availability of datasets, leaving limited options such as SoccerNet-Tracking an...
[ "Haobin Qin", "Calvin Yeung", "Rikuhei Umemoto", "Keisuke Fujii" ]
[ "cs.CV" ]
[ "Computer Science" ]
2025-01-16T00:00:00
https://arxiv.org/abs/2501.09281
https://arxiv.org/pdf/2501.09281v2
2501.09281
10.48550/arXiv.2501.09281
2
0
true
https://github.com/open-starlab/SoccerSynth-Detection
arXiv.org
0.1193
4abe1b5bddc920b0b2e831d7b30b6c002b91523e3fe97079e760404ec991b74e
[ "arxiv", "semantic_scholar" ]
Leveraging Explainable AI for LLM Text Attribution: Differentiating Human-Written and Multiple LLMs-Generated Text
The development of Generative AI Large Language Models (LLMs) raised the alarm regarding identifying content produced through generative AI or humans. In one case, issues arise when students heavily rely on such tools in a manner that can affect the development of their writing or coding skills. Other issues of plagiar...
[ "Ayat Najjar", "Huthaifa I. Ashqar", "Omar Darwish", "Eman Hammad" ]
[ "cs.CL", "cs.CY" ]
[ "Computer Science" ]
2025-01-06T00:00:00
https://arxiv.org/abs/2501.03212
https://arxiv.org/pdf/2501.03212v1
2501.03212
10.48550/arXiv.2501.03212
16
1
false
null
null
0.3076
15a5a6c1cf4f44e6bc79a5e73a0ff1d8089f0fc8ce9e4ab1f458168dab088d6c
[ "arxiv", "semantic_scholar" ]
RTLMarker: Protecting LLM-Generated RTL Copyright via a Hardware Watermarking Framework
Recent advances of large language models in the field of Verilog generation have raised several ethical and security concerns, such as code copyright protection and dissemination of malicious code. Researchers have employed watermarking techniques to identify codes generated by large language models. However, the exist...
[ "Kun Wang", "Kaiyan Chang", "Mengdi Wang", "Xinqi Zou", "Haobo Xu", "Yinhe Han", "Ying Wang" ]
[ "cs.CR", "cs.AI" ]
[ "Computer Science" ]
2025-01-05T00:00:00
https://arxiv.org/abs/2501.02446
https://arxiv.org/pdf/2501.02446v1
2501.02446
10.1145/3658617.3697774
7
0
false
null
Asia and South Pacific Design Automation Conference
0.2258
aead6aab7e4c405369e3ef22c27834e95cdeb86fd259b351add10691e00f5357
[ "arxiv", "semantic_scholar" ]
A Training-free Method for LLM Text Attribution
Verifying the provenance of content is crucial to the functioning of many organizations, e.g., educational institutions, social media platforms, and firms. This problem is becoming increasingly challenging as text generated by Large Language Models (LLMs) becomes almost indistinguishable from human-generated content. I...
[ "Tara Radvand", "Mojtaba Abdolmaleki", "Mohamed Mostagir", "Ambuj Tewari" ]
[ "stat.ML", "cs.AI", "cs.CL", "cs.IT", "cs.LG" ]
[ "Mathematics", "Computer Science" ]
2025-01-04T00:00:00
https://arxiv.org/abs/2501.02406
https://arxiv.org/pdf/2501.02406v5
2501.02406
null
3
0
true
https://github.com/TaraRadvand74/llm-text-detection
null
0.1505
c28eab89802e1a44dcd98907d5f83788f763afd3dfec603071c4b7256d80ebdf
[ "arxiv", "semantic_scholar" ]
TextSleuth: Towards Explainable Tampered Text Detection
Recently, tampered text detection has attracted increasing attention due to its essential role in information security. Although existing methods can detect the tampered text region, the interpretation of such detection remains unclear, making the prediction unreliable. To address this problem, we propose to explain th...
[ "Chenfan Qu", "Jian Liu", "Haoxing Chen", "Baihan Yu", "Jingjing Liu", "Weiqiang Wang", "Lianwen Jin" ]
[ "cs.CV" ]
[ "Computer Science" ]
2024-12-19T00:00:00
https://arxiv.org/abs/2412.14816
https://arxiv.org/pdf/2412.14816v3
2412.14816
null
6
1
true
null
null
0.2113
2d8fef12b3919ed6092b257d8df7135b12503d4dd35951a3d22e681708684b36
[ "arxiv", "semantic_scholar" ]
SuperMark: Robust and Training-free Image Watermarking via Diffusion-based Super-Resolution
In today's digital landscape, the blending of AI-generated and authentic content has underscored the need for copyright protection and content authentication. Watermarking has become a vital tool to address these challenges, safeguarding both generated and real content. Effective watermarking methods must withstand var...
[ "Runyi Hu", "Jie Zhang", "Yiming Li", "Jiwei Li", "Qing Guo", "Han Qiu", "Tianwei Zhang" ]
[ "cs.CV" ]
[ "Computer Science" ]
2024-12-13T00:00:00
https://arxiv.org/abs/2412.10049
https://arxiv.org/pdf/2412.10049v1
2412.10049
10.48550/arXiv.2412.10049
5
0
false
null
arXiv.org
0.1945
ccf8cf4dbb70b13a4f2c8017a68be6c44e4993aad5d989f13bed5d24d505a1d5
[ "arxiv", "semantic_scholar" ]
Watermarking Training Data of Music Generation Models
Generative Artificial Intelligence (Gen-AI) models are increasingly used to produce content across domains, including text, images, and audio. While these models represent a major technical breakthrough, they gain their generative capabilities from being trained on enormous amounts of human-generated content, which oft...
[ "Pascal Epple", "Igor Shilov", "Bozhidar Stevanoski", "Yves-Alexandre de Montjoye" ]
[ "cs.LG", "cs.SD", "eess.AS" ]
[ "Computer Science", "Engineering" ]
2024-12-11T00:00:00
https://arxiv.org/abs/2412.08549
https://arxiv.org/pdf/2412.08549v2
2412.08549
10.48550/arXiv.2412.08549
6
1
false
null
arXiv.org
0.2113
1840ea2a47d093a0e938bdec90d4ecf8bc5e7bd756ab440d7f040cb0b10f1d89
[ "arxiv", "semantic_scholar" ]
Exploring Memorization and Copyright Violation in Frontier LLMs: A Study of the New York Times v. OpenAI 2023 Lawsuit
Copyright infringement in frontier LLMs has received much attention recently due to the New York Times v. OpenAI lawsuit, filed in December 2023. The New York Times claims that GPT-4 has infringed its copyrights by reproducing articles for use in LLM training and by memorizing the inputs, thereby publicly displaying th...
[ "Joshua Freeman", "Chloe Rippe", "Edoardo Debenedetti", "Maksym Andriushchenko" ]
[ "cs.LG", "cs.AI" ]
[ "Computer Science" ]
2024-12-09T00:00:00
https://arxiv.org/abs/2412.06370
https://arxiv.org/pdf/2412.06370v1
2412.06370
10.48550/arXiv.2412.06370
26
0
false
null
arXiv.org
0.3578
7ea19eb71bd75879f361c946886a4512f1a3d8e52e544f493f31df31a6fa4a40
[ "arxiv", "semantic_scholar" ]
WATER-GS: Toward Copyright Protection for 3D Gaussian Splatting via Universal Watermarking
3D Gaussian Splatting (3DGS) has emerged as a pivotal technique for 3D scene representation, providing rapid rendering speeds and high fidelity. As 3DGS gains prominence, safeguarding its intellectual property becomes increasingly crucial since 3DGS could be used to imitate unauthorized scene creations and raise copyri...
[ "Yuqi Tan", "Xiang Liu", "Shuzhao Xie", "Bin Chen", "Shu-Tao Xia", "Zhi Wang" ]
[ "cs.CR" ]
[ "Computer Science" ]
2024-12-07T00:00:00
https://arxiv.org/abs/2412.05695
https://arxiv.org/pdf/2412.05695v1
2412.05695
10.48550/arXiv.2412.05695
6
1
false
null
arXiv.org
0.2113
d01a29cbef6068f760d536dc024f881080198c15930f76599ab59090935aac62
[ "arxiv", "semantic_scholar" ]
Robust Multi-bit Text Watermark with LLM-based Paraphrasers
We propose an imperceptible multi-bit text watermark embedded by paraphrasing with LLMs. We fine-tune a pair of LLM paraphrasers that are designed to behave differently so that their paraphrasing difference reflected in the text semantics can be identified by a trained decoder. To embed our multi-bit watermark, we use ...
[ "Xiaojun Xu", "Jinghan Jia", "Yuanshun Yao", "Yang Liu", "Hang Li" ]
[ "cs.AI" ]
[ "Computer Science" ]
2024-12-04T00:00:00
https://arxiv.org/abs/2412.03123
https://arxiv.org/pdf/2412.03123v2
2412.03123
10.48550/arXiv.2412.03123
9
0
true
https://github.com/xiaojunxu/multi-bit-text-watermark
International Conference on Machine Learning
0.25
f116e4e278366d4da1cb655023c7ef8d6a2f79772877aa706d143f46a8d5c295
[ "arxiv", "semantic_scholar" ]
CredID: Credible Multi-Bit Watermark for Large Language Models Identification
Large Language Models (LLMs) are widely used in complex natural language processing tasks but raise privacy and security concerns due to the lack of identity recognition. This paper proposes a multi-party credible watermarking framework (CredID) involving a trusted third party (TTP) and multiple LLM vendors to address ...
[ "Haoyu Jiang", "Xuhong Wang", "Ping Yi", "Shanzhe Lei", "Yilun Lin" ]
[ "cs.AI" ]
[ "Computer Science" ]
2024-12-04T00:00:00
https://arxiv.org/abs/2412.03107
https://arxiv.org/pdf/2412.03107v2
2412.03107
10.48550/arXiv.2412.03107
5
0
true
null
IEEE International Conference on Acoustics, Speech, and Signal Processing
0.1945
24af201045f41c5d93bd0a596d58a1cc4d45652ef613570b3929b158160ff451
[ "arxiv", "semantic_scholar" ]
The Efficacy of Transfer-based No-box Attacks on Image Watermarking: A Pragmatic Analysis
Watermarking approaches are widely used to identify if images being circulated are authentic or AI-generated. Determining the robustness of image watermarking methods in the ``no-box'' setting, where the attacker is assumed to have no knowledge about the watermarking model, is an interesting problem. Our main finding i...
[ "Qilong Wu", "Varun Chandrasekaran" ]
[ "cs.CR" ]
[ "Computer Science" ]
2024-12-03T00:00:00
https://arxiv.org/abs/2412.02576
https://arxiv.org/pdf/2412.02576v1
2412.02576
10.48550/arXiv.2412.02576
1
0
true
https://github.com/Ardor-Wu/transfer}
arXiv.org
0.0753
53828638016e5dbef9cbe6efde9397bd0c818af4970507104dc16e13be75de14
[ "arxiv", "semantic_scholar" ]
Ensemble Watermarks for Large Language Models
As large language models (LLMs) reach human-like fluency, reliably distinguishing AI-generated text from human authorship becomes increasingly difficult. While watermarks already exist for LLMs, they often lack flexibility and struggle with attacks such as paraphrasing. To address these issues, we propose a multi-featu...
[ "Georg Niess", "Roman Kern" ]
[ "cs.CL" ]
[ "Computer Science" ]
2024-11-29T00:00:00
https://arxiv.org/abs/2411.19563
https://arxiv.org/pdf/2411.19563v2
2411.19563
10.48550/arXiv.2411.19563
5
0
true
http://github.com/CommodoreEU/ensemble-watermark
Annual Meeting of the Association for Computational Linguistics
0.1945
1a7cebd34771c979bf09cdce75b77b1140510ff52e25b8aa31810b03992c41b9
[ "arxiv", "semantic_scholar" ]
Robust Detection of Watermarks for Large Language Models Under Human Edits
Watermarking has offered an effective approach to distinguishing text generated by large language models (LLMs) from human-written text. However, the pervasive presence of human edits on LLM-generated text dilutes watermark signals, thereby significantly degrading detection performance of existing methods. In this pape...
[ "Xiang Li", "Feng Ruan", "Huiyuan Wang", "Qi Long", "Weijie J. Su" ]
[ "stat.ME", "cs.CL", "cs.LG", "math.ST", "stat.ML" ]
[ "Computer Science", "Medicine", "Mathematics" ]
2024-11-21T00:00:00
https://arxiv.org/abs/2411.13868
https://arxiv.org/pdf/2411.13868v3
2411.13868
10.48550/arXiv.2411.13868
24
1
true
null
Journal of The Royal Statistical Society Series B-statistical Methodology
0.3495
23e043fea6b37d3c816d270160a0f449501196b18f580e9c36b92e9876d7757d
[ "arxiv", "semantic_scholar" ]
Watermark under Fire: A Robustness Evaluation of LLM Watermarking
Various watermarking methods (``watermarkers'') have been proposed to identify LLM-generated texts; yet, due to the lack of unified evaluation platforms, many critical questions remain under-explored: i) What are the strengths/limitations of various watermarkers, especially their attack robustness? ii) How do various d...
[ "Jiacheng Liang", "Zian Wang", "Lauren Hong", "Shouling Ji", "Ting Wang" ]
[ "cs.CR", "cs.CL", "cs.LG" ]
[ "Computer Science" ]
2024-11-20T00:00:00
https://arxiv.org/abs/2411.13425
https://arxiv.org/pdf/2411.13425v4
2411.13425
10.18653/v1/2025.findings-emnlp.1148
4
1
false
null
Conference on Empirical Methods in Natural Language Processing
0.1747
451b36e92b78b141aef773a64a37b9e2e9ff70c2baa36f6d4964ab6ac1d353d9
[ "arxiv", "semantic_scholar" ]
CopyrightMeter: Revisiting Copyright Protection in Text-to-image Models
Text-to-image diffusion models have emerged as powerful tools for generating high-quality images from textual descriptions. However, their increasing popularity has raised significant copyright concerns, as these models can be misused to reproduce copyrighted content without authorization. In response, recent studies h...
[ "Naen Xu", "Changjiang Li", "Tianyu Du", "Minxi Li", "Wenjie Luo", "Jiacheng Liang", "Yuyuan Li", "Xuhong Zhang", "Meng Han", "Jianwei Yin", "Ting Wang" ]
[ "cs.CR", "cs.AI", "cs.CV" ]
[ "Computer Science" ]
2024-11-20T00:00:00
https://arxiv.org/abs/2411.13144
https://arxiv.org/pdf/2411.13144v1
2411.13144
10.48550/arXiv.2411.13144
8
1
false
null
arXiv.org
0.2386
1a4ecbd64ecada11614ce84bcb068f77b10a5966deb9835daebd82dd14850218
[ "arxiv", "semantic_scholar" ]
Data Watermarking for Sequential Recommender Systems
In the era of large foundation models, data has become a crucial component in building high-performance AI systems. As the demand for high-quality and large-scale data continues to rise, data copyright protection is attracting increasing attention. In this work, we explore the problem of data watermarking for sequentia...
[ "Sixiao Zhang", "Cheng Long", "Wei Yuan", "Hongxu Chen", "Hongzhi Yin" ]
[ "cs.IR" ]
[ "Computer Science" ]
2024-11-20T00:00:00
https://arxiv.org/abs/2411.12989
https://arxiv.org/pdf/2411.12989v2
2411.12989
10.1145/3711896.3736903
3
0
false
null
Knowledge Discovery and Data Mining
0.1505
3ba660dcd57553fd69825ea3bffd6c10e0b3a33275372d0c19144ffe25e7c4f4
[ "arxiv", "semantic_scholar" ]
Watermarking Visual Concepts for Diffusion Models
The personalization techniques of diffusion models succeed in generating images with specific concepts. This ability also poses great threats to copyright protection and network security since malicious users can generate unauthorized content and disinformation relevant to a target concept. Model watermarking is an eff...
[ "Liangqi Lei", "Keke Gai", "Jing Yu", "Liehuang Zhu", "Qi Wu" ]
[ "cs.CR", "cs.AI", "cs.MM" ]
[ "Computer Science" ]
2024-11-18T00:00:00
https://arxiv.org/abs/2411.11688
https://arxiv.org/pdf/2411.11688v3
2411.11688
null
2
0
false
null
null
0.1193
8209655e9d9387b1afbb726d5acb4a9b9a92b022cb376558d6e5f85720db6f4f
[ "arxiv", "semantic_scholar" ]
SEFD: Semantic-Enhanced Framework for Detecting LLM-Generated Text
The widespread adoption of large language models (LLMs) has created an urgent need for robust tools to detect LLM-generated text, especially in light of \textit{paraphrasing} techniques that often evade existing detection methods. To address this challenge, we present a novel semantic-enhanced framework for detecting L...
[ "Weiqing He", "Bojian Hou", "Tianqi Shang", "Davoud Ataee Tarzanagh", "Qi Long", "Li Shen" ]
[ "cs.CL", "cs.AI", "cs.IR" ]
[ "Medicine", "Computer Science" ]
2024-11-17T00:00:00
https://arxiv.org/abs/2411.12764
https://arxiv.org/pdf/2411.12764v1
2411.12764
10.1109/BigData62323.2024.10825110
5
0
false
null
BigData Congress [Services Society]
0.1945
378a1e6ff04e3c0e0653ee369ba4e19657b4d2d06d995b86a2fd65ff7acfb042
[ "arxiv", "semantic_scholar" ]
Your Semantic-Independent Watermark is Fragile: A Semantic Perturbation Attack against EaaS Watermark
Embedding-as-a-Service (EaaS) has emerged as a successful business pattern but faces significant challenges related to various forms of copyright infringement, particularly, the API misuse and model extraction attacks. Various studies have proposed backdoor-based watermarking schemes to protect the copyright of EaaS se...
[ "Zekun Fei", "Biao Yi", "Jianing Geng", "Ruiqi He", "Lihai Nie", "Zheli Liu" ]
[ "cs.CR", "cs.AI" ]
[ "Computer Science" ]
2024-11-14T00:00:00
https://arxiv.org/abs/2411.09359
https://arxiv.org/pdf/2411.09359v2
2411.09359
10.18653/v1/2025.findings-emnlp.192
0
0
true
https://github.com/Zk4-ps/EaaS-Embedding-Watermark
Conference on Empirical Methods in Natural Language Processing
0
d98c49f7fe5a336a6cff05b230671d5dfd46e0979353df4e93db83c671d27502
[ "arxiv", "semantic_scholar" ]
Robust Detection of LLM-Generated Text: A Comparative Analysis
The ability of large language models to generate complex texts allows them to be widely integrated into many aspects of life, and their output can quickly fill all network resources. As the impact of LLMs grows, it becomes increasingly important to develop powerful detectors for the generated text. This detector is ess...
[ "Yongye Su", "Yuqing Wu" ]
[ "cs.CL" ]
[ "Computer Science" ]
2024-11-09T00:00:00
https://arxiv.org/abs/2411.06248
https://arxiv.org/pdf/2411.06248v1
2411.06248
10.48550/arXiv.2411.06248
4
0
true
null
arXiv.org
0.1747
33d4c26b7ead665975e752eb35a307065ab425637b87f1747f0b24c4ed316b83
[ "arxiv", "semantic_scholar" ]
Revisiting the Robustness of Watermarking to Paraphrasing Attacks
Amidst rising concerns about the internet being proliferated with content generated from language models (LMs), watermarking is seen as a principled way to certify whether text was generated from a model. Many recent watermarking techniques slightly modify the output probabilities of LMs to embed a signal in the genera...
[ "Saksham Rastogi", "Danish Pruthi" ]
[ "cs.CR", "cs.CL", "cs.LG" ]
[ "Computer Science" ]
2024-11-08T00:00:00
https://arxiv.org/abs/2411.05277
https://arxiv.org/pdf/2411.05277v1
2411.05277
10.18653/v1/2024.emnlp-main.1005
22
2
false
null
Conference on Empirical Methods in Natural Language Processing
0.3404
7a9b02f9329799bfcfa7869feff8dc4e92f1edf286d2e7bface99dd7d3fdb0da
[ "arxiv", "semantic_scholar" ]
Understanding the Effects of Human-written Paraphrases in LLM-generated Text Detection
Natural Language Generation has been rapidly developing with the advent of large language models (LLMs). While their usage has sparked significant attention from the general public, it is important for readers to be aware when a piece of text is LLM-generated. This has brought about the need for building models that en...
[ "Hiu Ting Lau", "Arkaitz Zubiaga" ]
[ "cs.CL" ]
[ "Computer Science" ]
2024-11-06T00:00:00
https://arxiv.org/abs/2411.03806
https://arxiv.org/pdf/2411.03806v1
2411.03806
10.48550/arXiv.2411.03806
16
2
false
null
Natural Language Processing Journal
0.3076
4af686c2bad9f5ae2ab6c665534d06ceb802d975f92858def6b0ca1f4aa6f940
[ "arxiv", "semantic_scholar" ]
Do LLMs Know to Respect Copyright Notice?
Prior study shows that LLMs sometimes generate content that violates copyright. In this paper, we study another important yet underexplored problem, i.e., will LLMs respect copyright information in user input, and behave accordingly? The research problem is critical, as a negative answer would imply that LLMs will beco...
[ "Jialiang Xu", "Shenglan Li", "Zhaozhuo Xu", "Denghui Zhang" ]
[ "cs.CL" ]
[ "Computer Science" ]
2024-11-02T00:00:00
https://arxiv.org/abs/2411.01136
https://arxiv.org/pdf/2411.01136v1
2411.01136
10.18653/v1/2024.emnlp-main.1147
19
2
false
null
Conference on Empirical Methods in Natural Language Processing
0.3253
96bc13fe37d8ccdcb3b9a460115814ddae0b4c5d15d041ff8f9b7a7fbc6208d9
[ "arxiv", "semantic_scholar" ]
$B^4$: A Black-Box Scrubbing Attack on LLM Watermarks
Watermarking has emerged as a prominent technique for LLM-generated content detection by embedding imperceptible patterns. Despite supreme performance, its robustness against adversarial attacks remains underexplored. Previous work typically considers a grey-box attack setting, where the specific type of watermark is a...
[ "Baizhou Huang", "Xiao Pu", "Xiaojun Wan" ]
[ "cs.CL" ]
[ "Computer Science" ]
2024-11-02T00:00:00
https://arxiv.org/abs/2411.01222
https://arxiv.org/pdf/2411.01222v3
2411.01222
10.48550/arXiv.2411.01222
8
0
false
null
North American Chapter of the Association for Computational Linguistics
0.2386
caf92666bf3bf978659e96a333f550339927037dc7a1ccb1af9519094ee7a0b3
[ "arxiv", "semantic_scholar" ]
Segmenting Watermarked Texts From Language Models
Watermarking is a technique that involves embedding nearly unnoticeable statistical signals within generated content to help trace its source. This work focuses on a scenario where an untrusted third-party user sends prompts to a trusted language model (LLM) provider, who then generates a text from their LLM with a wat...
[ "Xingchi Li", "Guanxun Li", "Xianyang Zhang" ]
[ "cs.LG", "cs.MM", "cs.NE", "stat.ML" ]
[ "Computer Science", "Mathematics" ]
2024-10-28T00:00:00
https://arxiv.org/abs/2410.20670
https://arxiv.org/pdf/2410.20670v1
2410.20670
10.48550/arXiv.2410.20670
5
1
true
https://github.com/doccstat/llm-watermark-cpd
Neural Information Processing Systems
0.1945
3196c77fc468f3fe66137565653b381d5293c72613dc9edc779f6fb92beae698
[ "arxiv", "semantic_scholar" ]
Robust and Minimally Invasive Watermarking for EaaS
Embeddings as a Service (EaaS) is emerging as a crucial role in AI applications. Unfortunately, EaaS is vulnerable to model extraction attacks, highlighting the urgent need for copyright protection. Although some preliminary works propose applying embedding watermarks to protect EaaS, recent research reveals that these...
[ "Zongqi Wang", "Baoyuan Wu", "Jingyuan Deng", "Yujiu Yang" ]
[ "cs.CL" ]
[ "Computer Science" ]
2024-10-23T00:00:00
https://arxiv.org/abs/2410.17552
https://arxiv.org/pdf/2410.17552v3
2410.17552
10.18653/v1/2025.findings-acl.112
4
2
false
null
Annual Meeting of the Association for Computational Linguistics
0.2386
e56895c52b1a8a1e5630f8cf8178b7f169507e541b17c238dba2d9d01f3ba52e
[ "arxiv", "semantic_scholar" ]
Beyond Binary: Towards Fine-Grained LLM-Generated Text Detection via Role Recognition and Involvement Measurement
The rapid development of large language models (LLMs), like ChatGPT, has resulted in the widespread presence of LLM-generated content on social media platforms, raising concerns about misinformation, data biases, and privacy violations, which can undermine trust in online discourse. While detecting LLM-generated conten...
[ "Zihao Cheng", "Li Zhou", "Feng Jiang", "Benyou Wang", "Haizhou Li" ]
[ "cs.CL" ]
[ "Computer Science" ]
2024-10-18T00:00:00
https://arxiv.org/abs/2410.14259
https://arxiv.org/pdf/2410.14259v2
2410.14259
10.1145/3696410.3714770
20
3
false
null
The Web Conference
0.3306
de6ad8fa8071a939d68193c5837348647615ca5df28f8baa31bc3f36b214bf33
[ "arxiv", "semantic_scholar" ]
Beyond Dataset Watermarking: Model-Level Copyright Protection for Code Summarization Models
Code Summarization Model (CSM) has been widely used in code production, such as online and web programming for PHP and Javascript. CSMs are essential tools in code production, enhancing software development efficiency and driving innovation in automated code analysis. However, CSMs face risks of exploitation by unautho...
[ "Jiale Zhang", "Haoxuan Li", "Di Wu", "Xiaobing Sun", "Qinghua Lu", "Guodong Long" ]
[ "cs.CR", "cs.CY" ]
[ "Computer Science" ]
2024-10-18T00:00:00
https://arxiv.org/abs/2410.14102
https://arxiv.org/pdf/2410.14102v2
2410.14102
10.1145/3696410.3714641
2
0
false
null
The Web Conference
0.1193
fcb58e9adcc533059d6987fd802b7b30a7a90b6296184a6a6f5d63d3f9632925
[ "arxiv", "semantic_scholar" ]
Which LLMs are Difficult to Detect? A Detailed Analysis of Potential Factors Contributing to Difficulties in LLM Text Detection
As LLMs increase in accessibility, LLM-generated texts have proliferated across several fields, such as scientific, academic, and creative writing. However, LLMs are not created equally; they may have different architectures and training datasets. Thus, some LLMs may be more challenging to detect than others. Using two...
[ "Shantanu Thorat", "Tianbao Yang" ]
[ "cs.CL", "cs.LG" ]
[ "Computer Science" ]
2024-10-18T00:00:00
https://arxiv.org/abs/2410.14875
https://arxiv.org/pdf/2410.14875v2
2410.14875
10.48550/arXiv.2410.14875
2
0
false
null
arXiv.org
0.1193
0b8ff528e2bd027add6fcb9416cae093cf6509fc33975df413ba7485154c0063
[ "arxiv", "semantic_scholar" ]
FreqMark: Frequency-Based Watermark for Sentence-Level Detection of LLM-Generated Text
The increasing use of Large Language Models (LLMs) for generating highly coherent and contextually relevant text introduces new risks, including misuse for unethical purposes such as disinformation or academic dishonesty. To address these challenges, we propose FreqMark, a novel watermarking technique that embeds detec...
[ "Zhenyu Xu", "Kun Zhang", "Victor S. Sheng" ]
[ "cs.CL", "cs.CR", "cs.LG" ]
[ "Computer Science" ]
2024-10-09T00:00:00
https://arxiv.org/abs/2410.10876
https://arxiv.org/pdf/2410.10876v1
2410.10876
10.48550/arXiv.2410.10876
7
0
false
null
arXiv.org
0.2258
7fd098ad791ff0cb34e5532eced836a72df927e5e3701557d4b80d8b1d90e1e4
[ "arxiv", "semantic_scholar" ]
WAPITI: A Watermark for Finetuned Open-Source LLMs
Watermarking of large language models (LLMs) generation embeds an imperceptible statistical pattern within texts, making it algorithmically detectable. Watermarking is a promising method for addressing potential harm and biases from LLMs, as it enables traceability, accountability, and detection of manipulated content,...
[ "Lingjie Chen", "Ruizhong Qiu", "Siyu Yuan", "Zhining Liu", "Tianxin Wei", "Hyunsik Yoo", "Zhichen Zeng", "Deqing Yang", "Hanghang Tong" ]
[ "cs.CR" ]
[ "Computer Science" ]
2024-10-09T00:00:00
https://arxiv.org/abs/2410.06467
https://arxiv.org/pdf/2410.06467v2
2410.06467
10.48550/arXiv.2410.06467
16
0
true
null
arXiv.org
0.3076
c38ca3e5137164af2a3a54d55c286de2210939eacd7516bc8d414ca06bd87ae0
[ "arxiv", "semantic_scholar" ]
Signal Watermark on Large Language Models
As Large Language Models (LLMs) become increasingly sophisticated, they raise significant security concerns, including the creation of fake news and academic misuse. Most detectors for identifying model-generated text are limited by their reliance on variance in perplexity and burstiness, and they require substantial c...
[ "Zhenyu Xu", "Victor S. Sheng" ]
[ "cs.CR", "cs.LG" ]
[ "Computer Science" ]
2024-10-09T00:00:00
https://arxiv.org/abs/2410.06545
https://arxiv.org/pdf/2410.06545v1
2410.06545
10.48550/arXiv.2410.06545
2
0
false
null
arXiv.org
0.1193
2c7398787a71ec499741dc20b0be7bc7fa990efc0741f7bf46d1d14939143c48
[ "arxiv", "semantic_scholar" ]
Inner-Probe: Discovering Copyright-related Data Generation in LLM Architecture
Large Language Models (LLMs) utilize extensive knowledge databases and show powerful text generation ability. However, their reliance on high-quality copyrighted datasets raises concerns about copyright infringements in generated texts. Current research often employs prompt engineering or semantic classifiers to identi...
[ "Qichao Ma", "Rui-Jie Zhu", "Peiye Liu", "Renye Yan", "Fahong Zhang", "Ling Liang", "Meng Li", "Zhaofei Yu", "Zongwei Wang", "Yimao Cai", "Tiejun Huang" ]
[ "cs.CL" ]
[ "Computer Science" ]
2024-10-06T00:00:00
https://arxiv.org/abs/2410.04454
https://arxiv.org/pdf/2410.04454v3
2410.04454
10.1109/TAI.2025.3645710
2
0
false
null
IEEE Transactions on Artificial Intelligence
0.1193
ac3c4f7a6103ca6a68181eb9c94b7323862a8208f49f5d4e9f2f2e7a5b933957
[ "arxiv", "semantic_scholar" ]
Can Watermarked LLMs be Identified by Users via Crafted Prompts?
Text watermarking for Large Language Models (LLMs) has made significant progress in detecting LLM outputs and preventing misuse. Current watermarking techniques offer high detectability, minimal impact on text quality, and robustness to text editing. However, current researches lack investigation into the imperceptibil...
[ "Aiwei Liu", "Sheng Guan", "Yiming Liu", "Leyi Pan", "Yifei Zhang", "Liancheng Fang", "Lijie Wen", "Philip S. Yu", "Xuming Hu" ]
[ "cs.CR", "cs.CL" ]
[ "Computer Science" ]
2024-10-04T00:00:00
https://arxiv.org/abs/2410.03168
https://arxiv.org/pdf/2410.03168v3
2410.03168
10.48550/arXiv.2410.03168
17
2
true
https://github.com/THU-BPM/Watermarked_LLM_Identification
International Conference on Learning Representations
0.3138
0ad4a69022ec2812b4e64ed7f8836d2ba560124dee3d83dd7b811676229162ba
[ "arxiv", "semantic_scholar" ]
Efficiently Identifying Watermarked Segments in Mixed-Source Texts
Text watermarks in large language models (LLMs) are increasingly used to detect synthetic text, mitigating misuse cases like fake news and academic dishonesty. While existing watermarking detection techniques primarily focus on classifying entire documents as watermarked or not, they often neglect the common scenario o...
[ "Xuandong Zhao", "Chenwen Liao", "Yu-Xiang Wang", "Lei Li" ]
[ "cs.CL" ]
[ "Computer Science" ]
2024-10-04T00:00:00
https://arxiv.org/abs/2410.03600
https://arxiv.org/pdf/2410.03600v2
2410.03600
10.48550/arXiv.2410.03600
3
1
true
https://github.com/XuandongZhao/llm-watermark-location
Annual Meeting of the Association for Computational Linguistics
0.1505
e8583aad248ef8cd6df28ddfa2456f4caec77f7cc2f539a7a400f81efd439259
[ "arxiv", "semantic_scholar" ]
Theoretically Grounded Framework for LLM Watermarking: A Distribution-Adaptive Approach
Watermarking has emerged as a crucial method to distinguish AI-generated text from human-created text. Current watermarking approaches often lack formal optimality guarantees or address the scheme and detector design separately. In this paper, we introduce a novel, unified theoretical framework for watermarking Large L...
[ "Haiyun He", "Yepeng Liu", "Ziqiao Wang", "Yongyi Mao", "Yuheng Bu" ]
[ "cs.CR", "cs.IT", "cs.LG" ]
[ "Computer Science", "Mathematics" ]
2024-10-03T00:00:00
https://arxiv.org/abs/2410.02890
https://arxiv.org/pdf/2410.02890v5
2410.02890
null
18
2
true
https://github.com/yepengliu/DAWA
null
0.3197
f994e192099d399ee49aa389834d43f32b9f6ec1a5d13c318879f864c67c21d5
[ "arxiv", "semantic_scholar" ]
Optimizing Adaptive Attacks against Watermarks for Language Models
Large Language Models (LLMs) can be misused to spread unwanted content at scale. Content watermarking deters misuse by hiding messages in content, enabling its detection using a secret watermarking key. Robustness is a core security property, stating that evading detection requires (significant) degradation of the cont...
[ "Abdulrahman Diaa", "Toluwani Aremu", "Nils Lukas" ]
[ "cs.CR", "cs.AI" ]
[ "Computer Science" ]
2024-10-03T00:00:00
https://arxiv.org/abs/2410.02440
https://arxiv.org/pdf/2410.02440v2
2410.02440
null
10
1
true
https://github.com/nilslukas/ada-wm-evasion
International Conference on Machine Learning
0.2603
bce7d1e6974ad83bcf51843195e8373817a1627a0dc3d8e0bccce46f68b000cc
[ "arxiv", "semantic_scholar" ]
Bridging Speech and Text: Enhancing ASR with Pinyin-to-Character Pre-training in LLMs
The integration of large language models (LLMs) with pre-trained speech models has opened up new avenues in automatic speech recognition (ASR). While LLMs excel in multimodal understanding tasks, effectively leveraging their capabilities for ASR remains a significant challenge. This paper presents a novel training appr...
[ "Yang Yuhang", "Peng Yizhou", "Eng Siong Chng", "Xionghu Zhong" ]
[ "cs.CL", "cs.SD", "eess.AS" ]
[ "Computer Science", "Engineering" ]
2024-09-24T00:00:00
https://arxiv.org/abs/2409.16005
https://arxiv.org/pdf/2409.16005v1
2409.16005
10.1109/ISCSLP63861.2024.10800477
3
1
false
null
International Symposium on Chinese Spoken Language Processing
0.1505
aad8807283bcec7100428c5dace1d30936c6feb906cc4a500c05dfd54357a13b
[ "arxiv", "semantic_scholar" ]
Towards Effective User Attribution for Latent Diffusion Models via Watermark-Informed Blending
Rapid advancements in multimodal large language models have enabled the creation of hyper-realistic images from textual descriptions. However, these advancements also raise significant concerns about unauthorized use, which hinders their broader distribution. Traditional watermarking methods often require complex integ...
[ "Yongyang Pan", "Xiaohong Liu", "Siqi Luo", "Yi Xin", "Xiao Guo", "Xiaoming Liu", "Xiongkuo Min", "Guangtao Zhai" ]
[ "cs.MM", "cs.CR", "cs.CV", "eess.IV" ]
[ "Computer Science", "Engineering" ]
2024-09-17T00:00:00
https://arxiv.org/abs/2409.10958
https://arxiv.org/pdf/2409.10958v2
2409.10958
10.48550/arXiv.2409.10958
5
0
false
null
arXiv.org
0.1945
8ad98cf9f14c8dd25c75f155e576467d2c433b849b5eec799fb57a2303b461c4
[ "arxiv", "semantic_scholar" ]
PersonaMark: Personalized LLM watermarking for model protection and user attribution
The rapid advancement of customized Large Language Models (LLMs) offers considerable convenience. However, it also intensifies concerns regarding the protection of copyright/confidential information. With the extensive adoption of private LLMs, safeguarding model copyright and ensuring data privacy have become critical...
[ "Yuehan Zhang", "Peizhuo Lv", "Yinpeng Liu", "Yongqiang Ma", "Wei Lu", "Xiaofeng Wang", "Xiaozhong Liu", "Jiawei Liu" ]
[ "cs.CR", "cs.CL" ]
[ "Computer Science" ]
2024-09-15T00:00:00
https://arxiv.org/abs/2409.09739
https://arxiv.org/pdf/2409.09739v2
2409.09739
10.48550/arXiv.2409.09739
5
0
false
null
arXiv.org
0.1945
434ceb9a5ee181309ded69c1e3dbd6418c3e7a1ad305d222a15b2110c32a7a9d
[ "arxiv", "semantic_scholar" ]
Protecting Copyright of Medical Pre-trained Language Models: Training-Free Backdoor Model Watermarking
With the advancement of intelligent healthcare, medical pre-trained language models (Med-PLMs) have emerged and demonstrated significant effectiveness in downstream medical tasks. While these models are valuable assets, they are vulnerable to misuse and theft, requiring copyright protection. However, existing watermark...
[ "Cong Kong", "Rui Xu", "Weixi Chen", "Jiawei Chen", "Zhaoxia Yin" ]
[ "cs.LG", "cs.AI", "cs.CR" ]
[ "Computer Science" ]
2024-09-14T00:00:00
https://arxiv.org/abs/2409.10570
https://arxiv.org/pdf/2409.10570v2
2409.10570
10.1145/3746027.3755548
1
0
false
null
ACM Multimedia
0.0753
bb2adb3f9e944e68d7424b1fa2240d75272ff7c167e89d172a20692084ea24b5
[ "arxiv", "semantic_scholar" ]
Generating API Parameter Security Rules with LLM for API Misuse Detection
In this paper, we present a new framework, named GPTAid, for automatic APSRs generation by analyzing API source code with LLM and detecting API misuse caused by incorrect parameter use. To validate the correctness of the LLM-generated APSRs, we propose an execution feedback-checking approach based on the observation th...
[ "Jinghua Liu", "Yi Yang", "Kai Chen", "Miaoqian Lin" ]
[ "cs.CR", "cs.SE" ]
[ "Computer Science" ]
2024-09-14T00:00:00
https://arxiv.org/abs/2409.09288
https://arxiv.org/pdf/2409.09288v2
2409.09288
10.14722/ndss.2025.23465
16
0
false
null
Network and Distributed System Security Symposium
0.3076
fad7fbdda2f1004eef64369639ea42086a4f75a540ffd80776436e8a7d6d559c
[ "arxiv", "semantic_scholar" ]
WaterSeeker: Pioneering Efficient Detection of Watermarked Segments in Large Documents
Watermarking algorithms for large language models (LLMs) have attained high accuracy in detecting LLM-generated text. However, existing methods primarily focus on distinguishing fully watermarked text from non-watermarked text, overlooking real-world scenarios where LLMs generate only small sections within large docume...
[ "Leyi Pan", "Aiwei Liu", "Yijian Lu", "Zitian Gao", "Yichen Di", "Shiyu Huang", "Lijie Wen", "Irwin King", "Philip S. Yu" ]
[ "cs.CL" ]
[ "Computer Science" ]
2024-09-08T00:00:00
https://arxiv.org/abs/2409.05112
https://arxiv.org/pdf/2409.05112v6
2409.05112
10.18653/v1/2025.findings-naacl.156
5
1
true
https://github.com/THU-BPM/WaterSeeker
North American Chapter of the Association for Computational Linguistics
0.1945
f4944228cd0a01b15781cd1c5bcd5d6b8c127bd4202a4a0cb179152b0f7e1a8e
[ "arxiv", "semantic_scholar" ]
Agentic Copyright Watermarking against Adversarial Evidence Forgery with Purification-Agnostic Curriculum Proxy Learning
With the proliferation of AI agents in various domains, protecting the ownership of AI models has become crucial due to the significant investment in their development. Unauthorized use and illegal distribution of these models pose serious threats to intellectual property, necessitating effective copyright protection m...
[ "Erjin Bao", "Ching-Chun Chang", "Hanrui Wang", "Isao Echizen" ]
[ "cs.CV", "cs.CR" ]
[ "Computer Science" ]
2024-09-03T00:00:00
https://arxiv.org/abs/2409.01541
https://arxiv.org/pdf/2409.01541v2
2409.01541
10.1109/ICASSP49660.2025.10889676
0
0
false
null
IEEE International Conference on Acoustics, Speech, and Signal Processing
0
bbcbbf05ba799c87890f21b4cdf17dacd007b7a522878cd69d048dcf20ebc252
[ "arxiv", "semantic_scholar" ]
RLCP: A Reinforcement Learning-based Copyright Protection Method for Text-to-Image Diffusion Model
The increasing sophistication of text-to-image generative models has led to complex challenges in defining and enforcing copyright infringement criteria and protection. Existing methods, such as watermarking and dataset deduplication, fail to provide comprehensive solutions due to the lack of standardized metrics and t...
[ "Zhuan Shi", "Jing Yan", "Xiaoli Tang", "Lingjuan Lyu", "Boi Faltings" ]
[ "cs.CY", "cs.AI", "cs.CR" ]
[ "Computer Science" ]
2024-08-29T00:00:00
https://arxiv.org/abs/2408.16634
https://arxiv.org/pdf/2408.16634v3
2408.16634
10.1109/ICME59968.2025.11210136
4
0
false
null
IEEE International Conference on Multimedia and Expo
0.1747
dc9b27bd0e2330f90512057ff4a7be6394b4121beed53148b3e829c9a06fd850
[ "arxiv", "semantic_scholar" ]
Watermarking Techniques for Large Language Models: A Survey
With the rapid advancement and extensive application of artificial intelligence technology, large language models (LLMs) are extensively used to enhance production, creativity, learning, and work efficiency across various domains. However, the abuse of LLMs also poses potential harm to human society, such as intellectu...
[ "Yuqing Liang", "Jiancheng Xiao", "Wensheng Gan", "Philip S. Yu" ]
[ "cs.CR", "cs.AI" ]
[ "Computer Science" ]
2024-08-26T00:00:00
https://arxiv.org/abs/2409.00089
https://arxiv.org/pdf/2409.00089v1
2409.00089
10.1007/s10462-025-11474-6
31
1
false
null
Artificial Intelligence Review
0.3763
fe26000d46f6e01a99726870640909867c4242645cb720f7d388c008f34369b8
[ "arxiv", "semantic_scholar" ]
Systematic Evaluation of LLM-as-a-Judge in LLM Alignment Tasks: Explainable Metrics and Diverse Prompt Templates
LLM-as-a-Judge has been widely applied to evaluate and compare different LLM alignmnet approaches (e.g., RLHF and DPO). However, concerns regarding its reliability have emerged, due to LLM judges' biases and inconsistent decision-making. Previous research has developed evaluation frameworks to assess reliability of LLM...
[ "Hui Wei", "Shenghua He", "Tian Xia", "Fei Liu", "Andy Wong", "Jingyang Lin", "Mei Han" ]
[ "cs.CL" ]
[ "Computer Science" ]
2024-08-23T00:00:00
https://arxiv.org/abs/2408.13006
https://arxiv.org/pdf/2408.13006v2
2408.13006
10.48550/arXiv.2408.13006
81
6
true
null
arXiv.org
0.4785
515ff9cc49af6f74cfe3f5d797eca6900a44896d19774d861072260aa61c6754
[ "arxiv", "semantic_scholar" ]
Authorship Attribution in the Era of LLMs: Problems, Methodologies, and Challenges
Accurate attribution of authorship is crucial for maintaining the integrity of digital content, improving forensic investigations, and mitigating the risks of misinformation and plagiarism. Addressing the imperative need for proper authorship attribution is essential to uphold the credibility and accountability of auth...
[ "Baixiang Huang", "Canyu Chen", "Kai Shu" ]
[ "cs.CY" ]
[ "Computer Science", "Medicine" ]
2024-08-16T00:00:00
https://arxiv.org/abs/2408.08946
https://arxiv.org/pdf/2408.08946v3
2408.08946
10.1145/3715073.3715076
72
1
false
null
SIGKDD Explorations
0.4658
0591003e9abc546c2fefc7f344c6e996f6cd4caf6a1316636cb5368c863975ca
[ "arxiv", "semantic_scholar" ]
LLM-DetectAIve: a Tool for Fine-Grained Machine-Generated Text Detection
The ease of access to large language models (LLMs) has enabled a widespread of machine-generated texts, and now it is often hard to tell whether a piece of text was human-written or machine-generated. This raises concerns about potential misuse, particularly within educational and academic domains. Thus, it is importan...
[ "Mervat Abassy", "Kareem Elozeiri", "Alexander Aziz", "Minh Ngoc Ta", "Raj Vardhan Tomar", "Bimarsha Adhikari", "Saad El Dine Ahmed", "Yuxia Wang", "Osama Mohammed Afzal", "Zhuohan Xie", "Jonibek Mansurov", "Ekaterina Artemova", "Vladislav Mikhailov", "Rui Xing", "Jiahui Geng", "Hasan ...
[ "cs.CL" ]
[ "Computer Science" ]
2024-08-08T00:00:00
https://arxiv.org/abs/2408.04284
https://arxiv.org/pdf/2408.04284v3
2408.04284
10.48550/arXiv.2408.04284
44
1
true
https://github.com/mbzuai-nlp/LLM-DetectAIve
Conference on Empirical Methods in Natural Language Processing
0.4133
d691f14829429afe0a29b1b0508b380b77c8f2f9e9db807862370dcea28a479a
[ "arxiv", "semantic_scholar" ]
Learning to Rewrite: Generalized LLM-Generated Text Detection
Large language models (LLMs) present significant risks when used to generate non-factual content and spread disinformation at scale. Detecting such LLM-generated content is crucial, yet current detectors often struggle to generalize in open-world contexts. We introduce Learning2Rewrite, a novel framework for detecting ...
[ "Ran Li", "Wei Hao", "Weiliang Zhao", "Junfeng Yang", "Chengzhi Mao" ]
[ "cs.CL" ]
[ "Computer Science" ]
2024-08-08T00:00:00
https://arxiv.org/abs/2408.04237
https://arxiv.org/pdf/2408.04237v2
2408.04237
10.48550/arXiv.2408.04237
15
1
false
null
arXiv.org
0.301
a258520a0fd0c5c8916989b72cfe55019f961f7a301771436a3717e4f41f21ff
[ "arxiv", "semantic_scholar" ]
Robustness of Watermarking on Text-to-Image Diffusion Models
Watermarking has become one of promising techniques to not only aid in identifying AI-generated images but also serve as a deterrent against the unethical use of these models. However, the robustness of watermarking techniques has not been extensively studied recently. In this paper, we investigate the robustness of ge...
[ "Xiaodong Wu", "Xiangman Li", "Jianbing Ni" ]
[ "cs.CR" ]
[ "Computer Science" ]
2024-08-04T00:00:00
https://arxiv.org/abs/2408.02035
https://arxiv.org/pdf/2408.02035v2
2408.02035
10.48550/arXiv.2408.02035
1
0
false
null
arXiv.org
0.0753
3642ad86a75528b313dfc68eb886caa45cbb2a03a08418452028513f8fd1d04d
[ "arxiv", "semantic_scholar" ]
Can Watermarking Large Language Models Prevent Copyrighted Text Generation and Hide Training Data?
Large Language Models (LLMs) have demonstrated impressive capabilities in generating diverse and contextually rich text. However, concerns regarding copyright infringement arise as LLMs may inadvertently produce copyrighted material. In this paper, we first investigate the effectiveness of watermarking LLMs as a deterr...
[ "Michael-Andrei Panaitescu-Liess", "Zora Che", "Bang An", "Yuancheng Xu", "Pankayaraj Pathmanathan", "Souradip Chakraborty", "Sicheng Zhu", "Tom Goldstein", "Furong Huang" ]
[ "cs.LG" ]
[ "Computer Science" ]
2024-07-24T00:00:00
https://arxiv.org/abs/2407.17417
https://arxiv.org/pdf/2407.17417v3
2407.17417
10.48550/arXiv.2407.17417
23
0
true
https://github.com/michael-panaitescu/watermark_copyright_aaai25
AAAI Conference on Artificial Intelligence
0.3451