English
Related papers

Related papers: CFL: Causally Fair Language Models Through Token-l…

200 papers

Large language models (LLMs) exhibit exceptional performance but pose inherent risks of generating toxic content, restricting their safe deployment. While traditional methods (e.g., alignment) adjust output preferences, they fail to…

Machine Learning · Computer Science 2026-01-13 Zenghao Duan , Zhiyi Yin , Zhichao Shi , Liang Pang , Shaoling Jing , Zihe Huang , Jiayi Wu , Yu Yan , Jingcheng Deng , Huawei Shen , Xueqi Cheng

Multilingual toxicity detection remains a significant challenge due to the scarcity of training data and resources for many languages. While prior work has leveraged the translate-test paradigm to support cross-lingual transfer across a…

Computation and Language · Computer Science 2025-09-19 Samuel J. Bell , Eduardo Sánchez , David Dale , Pontus Stenetorp , Mikel Artetxe , Marta R. Costa-jussà

Controllable text generation (CTG) aims to generate text with desired attributes, and decoding-time-based methods have shown promising performance on this task. However, in this paper, we identify the phenomenon of Attribute Collapse for…

Computation and Language · Computer Science 2023-11-03 Tianqi Zhong , Quan Wang , Jingxuan Han , Yongdong Zhang , Zhendong Mao

Detecting machine-generated text (MGT) from contemporary Large Language Models (LLMs) is increasingly crucial amid risks like disinformation and threats to academic integrity. Existing zero-shot detection paradigms, despite their…

Computation and Language · Computer Science 2025-08-19 Yue Wang , Liesheng Wei , Yuxiang Wang

Detoxification in large language models (LLMs) remains a significant research challenge. Existing decoding detoxification methods are all based on external constraints, which require additional resource overhead and lose generation fluency.…

Computation and Language · Computer Science 2025-10-16 Ming Dong , Jinkui Zhang , Bolong Zheng , Xinhui Tu , Po Hu , Tingting He

The dissemination of Large Language Models (LLMs), trained at scale, and endowed with powerful text-generating abilities, has made it easier for all to produce harmful, toxic, faked or forged content. In response, various proposals have…

Computation and Language · Computer Science 2025-06-12 Matthieu Dubois , François Yvon , Pablo Piantanida

Large language models (LLMs) have shown great potential in decision-making due to the vast amount of knowledge stored within the models. However, these pre-trained models are prone to lack reasoning abilities and are difficult to adapt to…

Machine Learning · Computer Science 2025-06-02 Wei Chen , Jiahao Zhang , Haipeng Zhu , Boyan Xu , Zhifeng Hao , Keli Zhang , Junjian Ye , Ruichu Cai

Large Language Models (LLMs) present massive inherent knowledge and superior semantic comprehension capability, which have revolutionized various tasks in natural language processing. Despite their success, a critical gap remains in…

Computation and Language · Computer Science 2025-02-11 Ben Liu , Jihai Zhang , Fangquan Lin , Cheng Yang , Min Peng

Biased associations have been a challenge in the development of classifiers for detecting toxic language, hindering both fairness and accuracy. As potential solutions, we investigate recently introduced debiasing methods for text…

Computation and Language · Computer Science 2021-02-02 Xuhui Zhou , Maarten Sap , Swabha Swayamdipta , Noah A. Smith , Yejin Choi

As large language models (LLMs) become increasingly prevalent in global applications, ensuring that they are toxicity-free across diverse linguistic contexts remains a critical challenge. We explore "Cross-lingual Detoxification", a…

Computation and Language · Computer Science 2025-10-24 Himanshu Beniwal , Youngwoo Kim , Maarten Sap , Soham Dan , Thomas Hartvigsen

Large-scale language models often learn behaviors that are misaligned with user expectations. Generated text may contain offensive or toxic language, contain significant repetition, or be of a different sentiment than desired by the user.…

Computation and Language · Computer Science 2022-11-18 Ximing Lu , Sean Welleck , Jack Hessel , Liwei Jiang , Lianhui Qin , Peter West , Prithviraj Ammanabrolu , Yejin Choi

Large language models can produce toxic or inappropriate text even for benign inputs, creating risks when deployed at scale. Detoxification is therefore important for safety and user trust, particularly when we want to reduce harmful…

Computation and Language · Computer Science 2026-02-04 Baturay Saglam , Dionysis Kalogerias

The rise of Large Language Models (LLMs) necessitates accurate AI-generated text detection. However, current approaches largely overlook the influence of author characteristics. We investigate how sociolinguistic attributes-gender, CEFR…

Computation and Language · Computer Science 2025-12-30 Jiatao Li , Xiaojun Wan

Causal explanations of the predictions of NLP systems are essential to ensure safety and establish trust. Yet, existing methods often fall short of explaining model predictions effectively or efficiently and are often model-specific. In…

Computation and Language · Computer Science 2023-11-23 Yair Gat , Nitay Calderon , Amir Feder , Alexander Chapanin , Amit Sharma , Roi Reichart

In this paper we aim to investigate the mechanisms that guide text generation with pre-trained Transformer-based Language Models (TLMs). Grounded on the Product of Experts formulation by Hinton (1999), we describe a generative mechanism…

Computation and Language · Computer Science 2023-06-16 Xavier Suau , Luca Zappella , Nicholas Apostoloff

Causal reasoning and compositional reasoning are two core aspirations in AI. Measuring the extent of these behaviors requires principled evaluation methods. We explore a unified perspective that considers both behaviors simultaneously,…

Computation and Language · Computer Science 2025-06-11 Jacqueline R. M. A. Maasch , Alihan Hüyük , Xinnuo Xu , Aditya V. Nori , Javier Gonzalez

Constrained decoding approaches aim to control the meaning or style of text generated by the pre-trained large language models (LLMs or also PLMs) for various tasks at inference time. However, these methods often guide plausible…

Computation and Language · Computer Science 2025-05-06 Chen Xu , Tian Lan , Yu Ji , Changlong Yu , Wei Wang , Jun Gao , Qunxi Dong , Kun Qian , Piji Li , Wei Bi , Bin Hu

Sparse autoencoders can localize where concepts live in language models, but not how they interact during multi-step reasoning. We propose Causal Concept Graphs (CCG): a directed acyclic graph over sparse, interpretable latent features,…

Machine Learning · Computer Science 2026-04-27 Md Muntaqim Meherab , Noor Islam S. Mohammad , Faiza Feroz

While language models (LMs) can sometimes generate factually correct text and estimate truth values of individual claims, these generally do not reflect a globally coherent, manipulable model of the world. As a consequence, current LMs also…

Computation and Language · Computer Science 2024-06-28 Afra Feyza Akyürek , Ekin Akyürek , Leshem Choshen , Derry Wijaya , Jacob Andreas

Ensuring that Large Language Models (LLMs) generate text representative of diverse sub-populations is essential, particularly when key concepts related to under-represented groups are scarce in the training data. We address this challenge…

Computation and Language · Computer Science 2024-12-17 Sabit Hassan , Anthony Sicilia , Malihe Alikhani