English
Related papers

Related papers: Domain Restriction via Multi SAE Layer Transitions

200 papers

There is growing interest in leveraging mechanistic interpretability and controllability to better understand and influence the internal dynamics of large language models (LLMs). However, current methods face fundamental challenges in…

Computation and Language · Computer Science 2025-12-02 Ruben Härle , Felix Friedrich , Manuel Brack , Stephan Wäldchen , Björn Deiseroth , Patrick Schramowski , Kristian Kersting

Sparse autoencoders (SAEs) have emerged as powerful techniques for interpretability of large language models (LLMs), aiming to decompose hidden states into meaningful semantic features. While several SAE variants have been proposed, there…

Machine Learning · Computer Science 2025-10-03 Xudong Zhu , Mohammad Mahdi Khalili , Zhihui Zhu

Watermarking LLM-generated text is critical for content attribution and misinformation prevention. However, existing methods compromise text quality, require white-box model access and logit manipulation. These limitations exclude API-based…

Computation and Language · Computer Science 2026-01-13 Zhuohao Yu , Xingru Jiang , Weizheng Gu , Yidong Wang , Qingsong Wen , Shikun Zhang , Wei Ye

This paper explores enabling large language models (LLMs) to understand spatial information from multichannel audio, a skill currently lacking in auditory LLMs. By leveraging LLMs' advanced cognitive and inferential abilities, the aim is to…

Sound · Computer Science 2024-06-17 Changli Tang , Wenyi Yu , Guangzhi Sun , Xianzhao Chen , Tian Tan , Wei Li , Jun Zhang , Lu Lu , Zejun Ma , Yuxuan Wang , Chao Zhang

Sparse autoencoders (SAEs) are used to analyze embeddings, but their role and practical value are debated. We propose a new perspective on SAEs by demonstrating that they can be naturally understood as topic models. We propose a continuous…

Computer Vision and Pattern Recognition · Computer Science 2026-05-19 Leander Girrbach , Zeynep Akata

Interpretability can improve the safety, transparency and trust of AI models, which is especially important in healthcare applications where decisions often carry significant consequences. Mechanistic interpretability, particularly through…

Modern LLMs face inference efficiency challenges due to their scale. To address this, many compression methods have been proposed, such as pruning and quantization. However, the effect of compression on a model's interpretability remains…

Machine Learning · Computer Science 2025-07-23 Suchit Gupte , Vishnu Kabir Chhabra , Mohammad Mahdi Khalili

Understanding and mitigating the potential risks associated with foundation models (FMs) hinges on developing effective interpretability methods. Sparse Autoencoders (SAEs) have emerged as a promising tool for disentangling FM…

Machine Learning · Computer Science 2024-11-04 Aashiq Muhamed , Mona Diab , Virginia Smith

Large Language Models (LLMs), being generic task solvers, are versatile. However, despite the vast amount of data they are trained on, there are speculations about their adaptation capabilities to a new domain. Additionally, the simple…

Computation and Language · Computer Science 2025-09-03 Anum Afzal , Mehul Kumawat , Florian Matthes

Large Language Models (LLMs) are traditionally viewed as black-box algorithms, therefore reducing trustworthiness and obscuring potential approaches to increasing performance on downstream tasks. In this work, we apply an effective LLM…

Computation and Language · Computer Science 2025-07-10 Shun Wang , Tyler Loakman , Youbo Lei , Yi Liu , Bohao Yang , Yuting Zhao , Dong Yang , Chenghua Lin

Variational autoencoders (VAEs) are widely used deep generative models capable of learning unsupervised latent representations of data. Such representations are often difficult to interpret or control. We consider the problem of…

Machine Learning · Computer Science 2018-12-18 Jack Klys , Jake Snell , Richard Zemel

Out-of-domain (OOD) intent detection aims to examine whether the user's query falls outside the predefined domain of the system, which is crucial for the proper functioning of task-oriented dialogue (TOD) systems. Previous methods address…

Computation and Language · Computer Science 2024-03-05 Pei Wang , Keqing He , Yejie Wang , Xiaoshuai Song , Yutao Mou , Jingang Wang , Yunsen Xian , Xunliang Cai , Weiran Xu

Decomposing model activations into interpretable components is a key open problem in mechanistic interpretability. Sparse autoencoders (SAEs) are a popular method for decomposing the internal activations of trained transformers into sparse,…

Machine Learning · Computer Science 2024-06-26 Connor Kissane , Robert Krzyzanowski , Joseph Isaac Bloom , Arthur Conmy , Neel Nanda

Reducing the likelihood of generating harmful and toxic output is an essential task when aligning large language models (LLMs). Existing methods mainly rely on training an external reward model (i.e., another language model) or fine-tuning…

Sparse Autoencoders (SAEs) aim to decompose the activation space of large language models (LLMs) into human-interpretable latent directions or features. As we increase the number of features in the SAE, hierarchical features tend to split…

Computation and Language · Computer Science 2025-11-18 David Chanin , James Wilken-Smith , Tomáš Dulka , Hardik Bhatnagar , Satvik Golechha , Joseph Bloom

We are witnessing a bloom of AI-powered software driven by Large Language Models (LLMs). Although the applications of these LLMs are impressive and seemingly countless, their unreliability hinders adoption. In fact, the tendency of LLMs to…

Software Engineering · Computer Science 2025-10-30 Vincenzo Scotti , Jan Keim , Tobias Hey , Andreas Metzger , Anne Koziolek , Raffaela Mirandola

Large language models (LLMs) have showcased their capability with few-shot inference known as in-context learning. However, in-domain demonstrations are not always readily available in real scenarios, leading to cross-domain in-context…

Computation and Language · Computer Science 2023-11-21 Quanyu Long , Wenya Wang , Sinno Jialin Pan

Debiasing techniques such as SentDebias aim to reduce bias in large language models (LLMs). Previous studies have evaluated their cross-lingual transferability by directly applying these methods to LLM representations, revealing their…

Computation and Language · Computer Science 2025-08-26 Qiwei Peng , Guimin Hu , Yekun Chai , Anders Søgaard

Sparse autoencoders (SAEs) have emerged as a powerful tool for uncovering interpretable features in large language models (LLMs) through the sparse directions they learn. However, the sheer number of extracted directions makes comprehensive…

Computation and Language · Computer Science 2025-11-11 Xinyuan Yan , Shusen Liu , Kowshik Thopalli , Bei Wang

Page Stream Segmentation (PSS) is an essential prerequisite for automated document processing at scale. However, research progress has been limited by the absence of realistic public benchmarks. This paper works towards addressing this gap…

Computation and Language · Computer Science 2024-08-23 Hunter Heidenreich , Ratish Dalvi , Rohith Mukku , Nikhil Verma , Neven Pičuljan
‹ Prev 1 3 4 5 6 7 10 Next ›