MoSEs: Uncertainty-Aware AI-Generated Text Detection via Mixture of Stylistics Experts with Conditional Thresholds (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods neglect stylistic modeling and rely on static thresholds, which greatly limits the detection performance. |
| Approach: | They propose a framework that enables stylistics-aware uncertainty quantification through conditional threshold estimation. |
| Outcome: | The proposed framework achieves an average improvement 11.34% in detection performance compared to baselines. |
Similar Papers
AID: Adaptive Integration of Detectors for Safe AI with Language Models (2025.naacl-long)
Copied to clipboard
| Challenge: | Large language models (LLMs) are increasingly used to generate human-like text, but safety concerns have emerged with the deployment of LLMs. |
| Approach: | They propose an approach that orchestrates the strengths of multiple pretrained detectors to ensure comprehensive effectiveness in diverse scenarios. |
| Outcome: | The proposed approach can improve the area under the curve (AUC) by 0.07 to 0.21, with a median of 0.12, compared to the best individual detectors developed for specific safety aspects. |
Towards Robust and Semantically Organised Latent Representations for Unsupervised Text Style Transfer (2022.naacl-main)
Copied to clipboard
| Challenge: | Recent studies show that auto-encoders perform language generation, smooth sentence interpolation, and style transfer over unseen attributes using unlabelled datasets in a zero-shot manner. |
| Approach: | They propose a discrete token-based perturbation approach to map "similar" sentences close by in latent space. |
| Outcome: | The proposed model can generate and perform language generation, style transfer and sentence interpolation tasks on unlabelled datasets in a zero-shot manner. |
Automatic Detection of Machine Generated Text: A Critical Survey (2020.coling-main)
Copied to clipboard
| Challenge: | Current text generative models excel in producing text that matches the style of human language reasonably well. |
| Approach: | They conduct an in-depth error analysis of the state-of-the-art detector and discuss research directions to guide future work in this exciting area. |
| Outcome: | The proposed detectors can distinguish between human and text generated by the model and can be used to generate fake news and fake product reviews. |
Counter Turing Test (CT2): Investigating AI-Generated Text Detection for Hindi - Ranking LLMs based on Hindi AI Detectability Index (ADI_hi) (2024.findings-emnlp)
Copied to clipboard
| Challenge: | a growing number of large language models are being used to detect AI-generated text . a recent study has found that some techniques to bypass detection are fragile . |
| Approach: | They propose to use 26 LLMs to evaluate their proficiency in generating Hindi text . they propose to introduce a Hindi AI Detectability Index to assess and rank LLM models based on their detectability levels. |
| Outcome: | The proposed methods are effective in English, but struggle in Hindi . the proposed methods show that they are susceptible to fragility . |
Leveraging Human and Machine Preferences for Zero-shot Detection of AI-Generated Text (2026.findings-acl)
Copied to clipboard
| Challenge: | Recent advances in large language models have enabled generated texts to closely mimic human writing, posing significant challenges to the detection of AI-generated content. |
| Approach: | They propose a human-machine prediction discrepancy adapter for AI-generated text detection . they use a joint fine-tuning strategy and a discrepany-aware reweighting mechanism . |
| Outcome: | The proposed framework improves the detection performance of five representative models under various evaluation scenarios. |
LLM-as-a-Coauthor: Can Mixed Human-Written and Machine-Generated Text Be Detected? (2024.findings-naacl)
Copied to clipboard
Qihui Zhang, Chujie Gao, Dongping Chen, Yue Huang, Yixin Huang, Zhenyang Sun, Shilin Zhang, Weiye Li, Zhengyan Fu, Yao Wan, Lichao Sun
| Challenge: | Current research focuses on purely MGT detection without adequately addressing mixed scenarios including AI-revised Human-Written Text (HWT) and human-revealed MGT. |
| Approach: | They define mixtext, a form of mixed text involving both AI and human-generated content, and then use a MixSet dataset to assess their effectiveness. |
| Outcome: | The proposed detectors struggle to identify mixtext, particularly in dealing with subtle modifications and style adaptability. |
FAID: Fine-grained AI-generated Text Detection using Multi-task Auxiliary and Multi-level Contrastive Learning (2026.eacl-long)
Copied to clipboard
Minh Ngoc Ta, Dong Cao Van, Duc-Anh Hoang, Minh Le-Anh, Truong Nguyen, My Anh Tran Nguyen, Yuxia Wang, Preslav Nakov, Dinh Viet Sang
| Challenge: | Existing binary detection frameworks for human-written, LLM-generated and human-LLM collaborative texts are challenging . a recent study focused on binary detection, i.e., human vs. LLM, or on fine-grained detection limited to English. |
| Approach: | They propose a fine-grained detection framework to classify text into three categories . they use multilingual datasets and a multi-domain, multi-generator dataset . |
| Outcome: | The proposed framework outperforms baselines on unseen domains and new LLMs. |
Uncertainty Quantification with Pre-trained Language Models: A Large-Scale Empirical Analysis (2022.findings-emnlp)
Copied to clipboard
Yuxin Xiao, Paul Pu Liang, Umang Bhatt, Willie Neiswanger, Ruslan Salakhutdinov, Louis-Philippe Morency
| Challenge: | Pre-trained language models (PLMs) have gained increasing popularity due to compelling prediction performance in diverse natural language processing tasks. |
| Approach: | They compare three popular options for encoding and Temp Scaling for PLMs . they recommend using Temp Loss as uncertainty quantifier and Focal Loss for fine-tuning . |
| Outcome: | Using pre-trained language models, we compare three options on NLP classification tasks and domain shift. |
AGSC: Adaptive Granularity and Semantic Clustering for Uncertainty Quantification in Long-text Generation (2026.acl-long)
Copied to clipboard
| Challenge: | Existing methods for aggregating large-form outputs overlook the nuance of neutral information and suffer from the high computational cost of fine-grained decomposition. |
| Approach: | They propose a UQ framework that uses NLI neutral probabilities as triggers to distinguish irrelevance from uncertainty, reducing computation costs. |
| Outcome: | Experiments on BIO and LongFact show that the proposed framework reduces inference time by 60% compared to full atomic decomposition. |
MASH: Evading Black-Box AI-Generated Text Detectors via Style Humanization (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing detection methods rely on white-box assumptions or require prohibitively high computational and interaction costs, rendering them ineffective under practical black-box scenarios. |
| Approach: | They propose a framework that evades black-box detection methods based on style transfer by using style-injection supervised fine-tuning and direct preference optimization to shape distributions of AI-generated texts to resemble those of human-written texts. |
| Outcome: | The proposed framework achieves an average Attack Success Rate (ASR) of 92%, surpassing the strongest baselines by an average of 24% while maintaining superior linguistic quality. |