Papers with detectors
Synthetic Text Detection in the Age of Large Language Models: Watermark vs. Automatic Detection (2026.acl-industry)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) are ubiquitous and capable of generating long coherent texts that look almost indistinguishable from human-written texts. |
| Approach: | They propose to use watermark and automatic detection to detect synthetic texts generated from Large Language Models (LLMs) they evaluate six different models, six different watermark techniques and two different automatic detectors for different levels of syntactic changes. |
| Outcome: | The proposed methods outperform on unperturbed and perturbed datasets on six different sizes of Qwen2.5 models, six watermark techniques and detectors, and two automatic detectors. |
SearchLLM: Detecting LLM Paraphrased Text by Measuring the Similarity with Regeneration of the Candidate Source via Search Engine (2026.eacl-long)
Copied to clipboard
| Challenge: | Large language models (LLMs) can be used to enhance text quality but can sometimes result in loss or distortion of original meaning. |
| Approach: | They propose a method to identify LLM-paraphrased text by leveraging search engine capabilities to locate potential original text sources. |
| Outcome: | The proposed approach distinguishes LLM-paraphrased text from genuine human writing . it uses search engine capabilities to integrate with existing detectors to improve performance . |
M4: Multi-generator, Multi-domain, and Multi-lingual Black-Box Machine-Generated Text Detection (2024.eacl-long)
Copied to clipboard
Yuxia Wang, Jonibek Mansurov, Petar Ivanov, Jinyan Su, Artem Shelmanov, Akim Tsvigun, Chenxi Whitehouse, Osama Mohammed Afzal, Tarek Mahmoud, Toru Sasaki, Thomas Arnold, Alham Aji, Nizar Habash, Iryna Gurevych, Preslav Nakov
| Challenge: | Large language models generate fluent responses to user queries, but they are also susceptible to misuse in journalism, education, and academia. |
| Approach: | They propose a large-scale benchmark for machine-generated text detection that is a multi-generator, multi-domain, and multi-lingual corpus. |
| Outcome: | The proposed system can detect machine-generated text and pinpoint misuse . the proposed system is based on a large-scale benchmark dataset . |
Adapting Fake News Detection to the Era of Large Language Models (2024.findings-naacl)
Copied to clipboard
| Challenge: | a gap exists in understanding the interplay between machine-paraphrased real news, machine-generated fake news, and human-written real news . false information is easier to generate but harder to detect due to the bias of detectors against machine-generated texts . |
| Approach: | They propose a strategy to adapt fake news detectors to the era of large language models and AI-driven content creation . |
| Outcome: | The proposed detectors perform well on human-written articles but not vice versa . the proposed detector should be trained on datasets with lower machine-generated news ratio than the test set . |
Ghostbuster: Detecting Text Ghostwritten by Large Language Models (2024.naacl-long)
Copied to clipboard
| Challenge: | Ghostbuster is a system that passes documents through weaker language models, runs a structured search over possible combinations of their features, and trains a classifier on the selected features. |
| Approach: | They propose a method that passes documents through weaker language models, runs a structured search over possible combinations of their features, and trains a classifier on the selected features. |
| Outcome: | The proposed method outperforms existing detectors and a new baseline on student essays, creative writing, and news articles. |
Double Entendre: Robust Audio-Based AI-Generated Lyrics Detection via Multi-View Fusion (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing methods for detecting AI-generated music are weak and vulnerable to audio perturbations. |
| Approach: | They propose a multimodal late-fusion pipeline that combines automatically transcribed sung lyrics and speech features capturing lyrics related information within the audio. |
| Outcome: | The proposed method outperforms existing detectors while being more robust to audio perturbations. |
Threat Scenarios and Best Practices to Detect Neural Fake News (2022.coling-1)
Copied to clipboard
| Challenge: | During the COVID-19 pandemic, inaccurate information made it hard for people to find reliable guidance when they needed it. |
| Approach: | They propose to use pretrained language models to generate fluent, original text . they argue that strong detectors should be released along with new generators . |
| Outcome: | The proposed system is prone to shortcut learning and should be released along with new generators. |
Do LLM hallucination detectors suffer from low-resource effect? (2026.eacl-long)
Copied to clipboard
| Challenge: | a long line of work suggests that LLMs face issues along both dimensions . |
| Approach: | They investigate hallucination detectors' failure modes and their effects on the task accuracy of four LLMs and three halluciner detectors. |
| Outcome: | The models show impressive performance in high-resource languages like English but the performance degrades significantly in low-resourced languages like Bengali. |
Stress-testing Machine Generated Text Detection: Shifting Language Models Writing Style to Fool Detectors (2025.findings-acl)
Copied to clipboard
Andrea Pedrotti, Michele Papucci, Cristiano Ciaccio, Alessio Miaschi, Giovanni Puccetti, Felice Dell’Orletta, Andrea Esuli
| Challenge: | Recent advances in Generative AI and Large Language Models (LLMs) have enabled the creation of highly realistic synthetic content, raising concerns about the potential for malicious use, such as misinformation and manipulation. |
| Approach: | They evaluate the resilience of state-of-the-art MGT detectors to linguistically informed adversarial attacks by using Direct Preference Optimization to shift the MGT style toward human-written text. |
| Outcome: | The proposed pipeline fine-tunes language models to shift the MGT style toward human-written text (HWT) it obtains generations more challenging to detect by current models, and shows that detectors can be easily fooled with relatively few examples, resulting in a significant drop in detecting performances. |
Stumbling Blocks: Stress Testing the Robustness of Machine-Generated Text Detectors Under Attacks (2024.acl-long)
Copied to clipboard
| Challenge: | Existing studies on this topic focus on the robustness of specific detectors or particular attack methods. |
| Approach: | They stress test the detectors’ robustness to malicious attacks under realistic scenarios using LLMs and metric-based detectors. |
| Outcome: | The proposed methods are based on a set of LLM-based models and their performance is compared under different budget levels. |
Detecting Stealthy Backdoor Samples based on Intra-class Distance for Large Language Models (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Existing detectors use classifier-style probability signals or rely on rewriting, which can degrade quality and introduce new triggers. |
| Approach: | They propose to efficiently remove poisoned examples before or during fine-tuning . |
| Outcome: | The proposed method outperforms prior detectors on two machine translation datasets and one QA dataset. |
The Two Paradigms of LLM Detection: Authorship Attribution vs Authorship Verification (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing methods for detecting texts generated by large language models are disputed . authors argue that there are limitations in the current technology . |
| Approach: | They propose to make LLM detectors robust against domain shifts and build benchmarks . they argue that the limitations lie elsewhere, and open the realm of authorship analysis technology . |
| Outcome: | The proposed method systematically analyzes the benchmarks and validates it using state-of-the-art detectors. |
On the Zero-Shot Generalization of Machine-Generated Text Detectors (2023.findings-emnlp)
Copied to clipboard
| Challenge: | rampant proliferation of large language models generates text indistinguishable from human-written language. |
| Approach: | They train neural detectors on outputs of a new generator and test their performance on held-out generators. |
| Outcome: | The proposed detectors can be built on training data from medium-sized models. |
DAMASHA: Detecting AI in Mixed Adversarial Texts via Segmentation with Human-interpretable Attribution (2026.findings-eacl)
Copied to clipboard
| Challenge: | a new framework for mixed authorship detection addresses the challenge of segmenting mixed-authorship text . mixed-authored text detection is a growing concern in the age of advanced large language models . a recent survey highlighted the greater challenges of detecting AI content in realworld settings . |
| Approach: | They propose a framework for mixed authorship detection that integrates stylometric cues, perplexity-driven signals, and structured boundary modeling to accurately segment collaborative human-AI content. |
| Outcome: | The proposed framework improves robustness against adversarial perturbations while revealing limitations. |
WaveDetect: Robust Framework for Machine-Generated Text Detection via Wavelet Transform (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing methods for detecting LLM-generated texts falter when faced with adversarial perturbations, cross-domain shifts, and the rapid temporal evolution of the foundation model. |
| Approach: | They propose a framework that reformulates text detection as a signal processing task within the time-frequency domain. |
| Outcome: | The proposed framework achieves superior accuracy and robustness against sophisticated attacks and generalization across out-of-distribution topics. |
Your Language Model Can Secretly Write Like Humans: Contrastive Paraphrase Attacks on LLM-Generated Text Detectors (2025.emnlp-main)
Copied to clipboard
Hao Fang, Jiawei Kong, Tianqu Zhuang, Yixiang Qiu, Kuofeng Gao, Bin Chen, Shu-Tao Xia, Yaowei Wang, Min Zhang
| Challenge: | Existing methods to detect large language models (LLMs) generated for plagiarism use paraphrases to rewrite them to evade detection. |
| Approach: | They propose a training-free method that effectively fools text detectors using off-the-shelf LLMs by rewriting them to evade detection. |
| Outcome: | The proposed method deceives text detectors using off-the-shelf LLMs by rewriting them to produce human-like sentences that are less discernible by detectors. |
Mitigating Biases in Hate Speech Detection from A Causal Perspective (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Existing methods to detect hate speech are prone to spurious correlations between training data and labels, which could lead to biased treatment of vulnerable and minority groups. |
| Approach: | They propose to use grammar induction to find grammar patterns for hate speech and analyze this phenomenon from a causal perspective. |
| Outcome: | The proposed methods can detect hate speech from a causal perspective and are robust to different datasets. |
BUST: Benchmark for the evaluation of detectors of LLM-Generated Text (2024.naacl-long)
Copied to clipboard
| Challenge: | Using the benchmark, we evaluated 5 detectors and found substantial performance variance across tasks. |
| Approach: | They propose to evaluate detectors of texts generated by instruction-tuned large language models (LLMs) using a benchmark dataset, they evaluated 5 detectors and found substantial performance variance across tasks. |
| Outcome: | The proposed benchmarks evaluated 5 detectors and found substantial performance variance across tasks. |
Machine Translated Text Detection Through Text Similarity with Round-Trip Translation (2021.naacl-main)
Copied to clipboard
| Challenge: | Existing detectors for translating texts fail to detect a text from a strange translator . Existing methods for detection of translated texts use text structure and complex words to detect translations . |
| Approach: | They propose a detector using text similarity with round-trip translation (TSRT) TSRT achieves 86.9% accuracy in detecting a translated text from a strange translator . Existing detectors have been built around a specific translator but fail to detect a translation from skeptics . |
| Outcome: | Existing detectors fail to detect translated texts from a strange translator . a detector achieves 86.9% accuracy in detecting a translated text from skeptic translators . |
On the Risk of Evidence Pollution for Malicious Social Text Detection in the Era of LLMs (2025.acl-long)
Copied to clipboard
| Challenge: | Evidence-enhanced detectors are able to detect malicious social text, but they are prone to evidence pollution. |
| Approach: | They propose three defense strategies to mitigate evidence pollution by large language models by machine-generated text detection and a mixture of experts. |
| Outcome: | The proposed defense strategies could mitigate evidence pollution, but they faced limitations for practical employment. |
From Form to Logic: Masked Reconstruction and Reasoning Distillation for Short Video Fake News Detection (2026.acl-long)
Copied to clipboard
| Challenge: | Existing detectors that detect short video fake news suffer from global-alignment bias and lack generative reasoning are too late. |
| Approach: | They propose a Perception-Cognition Dual-driven Detector that jointly observes the form and probes the logic for short video fake news detection. |
| Outcome: | The proposed detector outperforms baseline detectors on real-world datasets while improving interpretability and robustness in data scarcity scenarios. |
Hidding the Ghostwriters: An Adversarial Evaluation of AI-Generated Student Essay Detection (2023.emnlp-main)
Copied to clipboard
| Challenge: | Large language models (LLMs) have exhibited remarkable capabilities in text generation tasks, but their utilization carries inherent risks, including plagiarism and the dissemination of fake news. |
| Approach: | They propose to use a dataset to construct an AI-generated student essay that employs a range of text perturbation methods to evade detection. |
| Outcome: | The proposed methods evade detection and maintain quality of the generated essays while avoiding plagiarism and fake news. |
Optimal Transport for Unsupervised Hallucination Detection in Neural Machine Translation (2023.acl-long)
Copied to clipboard
| Challenge: | Neural machine translation models can unpredictably produce severely pathological translations, known as hallucinations, that seriously undermine user trust. |
| Approach: | They propose a fully unsupervised, plug-in detector that can be used with any attention-based NMT model. |
| Outcome: | The proposed detector outperforms existing models and is competitive with detectors that employ external models trained on millions of samples. |
Types of Out-of-Distribution Texts and How to Detect Them (2021.emnlp-main)
Copied to clipboard
| Challenge: | Current NLP models produce unreliable or catastrophic predictions when training and test distributions differ . current models tend to produce unreliability or even catastrophic predictions that hurt user trust. |
| Approach: | They categorize examples as exhibiting a background shift or semantic shift and use calibration and density estimation methods to detect OOD examples. |
| Outcome: | The proposed methods beat calibration methods in background shift settings and perform worse in semantic shift settings. |
How You Prompt Matters! Even Task-Oriented Constraints in Instructions Affect LLM-Generated Text Detection (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Recent studies have presented LLM-generated-text detectors with promising performance, but they do not cover such diverse instruction patterns when creating datasets for LLM detection. |
| Approach: | They propose to use task-oriented constraints that would naturally be included in an instruction and are not related to detection-evasion to create detectors with large variances in detection performance. |
| Outcome: | The proposed detectors have a large variance in detection performance on student essay writing with task-oriented constraints, and the standard deviation is significantly larger than that on texts generated by the constraint with such a constraint. |
CoCoNUTS: Concentrating on Content while Neglecting Uninformative Textual Styles for AI-Generated Peer Review Detection (2026.acl-long)
Copied to clipboard
| Challenge: | Existing detectors rely on stylistic cues to distinguish between surface-level language refinement and genuine content generation. |
| Approach: | They propose a content-based detection paradigm to detect substantive AI-generation . they propose 'CoCoDet' detector that can detect surface-level language refinement . |
| Outcome: | The proposed detector achieves a macro F1 score of 98.24% on permissible machine-polished reviews and maintains 3.89% false positive rate on real-world reviews. |
Summary Factual Inconsistency Detection Based on LLMs Enhanced by Universal Information Extraction (2025.findings-acl)
Copied to clipboard
| Challenge: | Recent studies have shown that Large language models can detect factual inconsistencies in summaries but they lack the efficiency and explainability needed to be effective. |
| Approach: | They propose to decouple LLMs’ information extraction and reasoning capabilities to address key challenges and propose a framework for UIEFID to guide fine-tuned LLM methods in extracting unified structured information from documents and summaries. |
| Outcome: | The proposed framework improves the detection accuracy and reduces redundant reasoning on the AGGREFACT benchmark. |
OpenTuringBench: An Open-Model-based Benchmark and Framework for Machine-Generated Text Detection and Attribution (2025.emnlp-main)
Copied to clipboard
| Challenge: | Open Large Language Models (OLLMs) are increasingly leveraged in generative AI applications, posing new challenges for detecting their outputs. |
| Approach: | They propose a benchmark to train and evaluate machine-generated text detectors on Turing Test and Authorship Attribution problems. |
| Outcome: | The proposed detector outperforms existing detectors in varying degrees of difficulty and relevance across tasks. |
Droid: A Resource Suite for AI-Generated Code Detection (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing detectors fail to generalise to diverse coding domains and programming languages outside of their narrow training data. |
| Approach: | They propose to use DroidCollection to train machine-generated code detectors that can be trained on a multi-task objective. |
| Outcome: | The proposed detectors fail to generalise to diverse coding domains and programming languages outside of their narrow training data. |
TempParaphraser: “Heating Up” Text to Evade AI-Text Detection through Paraphrasing (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing detectors that perform well on benchmark datasets have weaknesses that can be exploited to manipulate AI-text. |
| Approach: | They propose a framework that simulates high-temperature sampling effects through multiple normal-temperaturing generations, effectively evading detection. |
| Outcome: | The proposed framework reduces detector accuracy by an average of 82.5% while preserving high text quality. |
DetectRL-X: Towards Reliable Multilingual and Real-World LLM-Generated Text Detection (2026.acl-long)
Copied to clipboard
Junchao Wu, Yefeng Liu, Chenyu Zhu, Hao Zhang, Zeyu Wu, Tianqi Shi, Yichao Du, Longyue Wang, Weihua Luo, Jinsong Su, Derek F. Wong
| Challenge: | Existing detectors are limited in their ability to detect large language models generated content in multilingual environments. |
| Approach: | They propose a multilingual benchmark to evaluate advanced detectors across 8 dimensions to better align with real-world applications. |
| Outcome: | The proposed benchmark encompasses 8 languages commonly used in commercial contexts and collects human-written texts from 6 domains highly susceptible to LLM misuse. |