English
Related papers

Related papers: Steered Generation via Gradient Descent on Sparse …

200 papers

Large Language Models (LLMs) offer extensive knowledge across various domains, but they may inadvertently memorize sensitive, unauthorized, or malicious data, such as personal information in the medical and financial sectors. Machine…

Computation and Language · Computer Science 2024-10-16 YuXuan Wu , Bonaventure F. P. Dossou , Dianbo Liu

Understanding the latent space geometry of large language models (LLMs) is key to interpreting their behavior and improving alignment. Yet it remains unclear to what extent LLMs linearly organize representations related to semantic…

Computation and Language · Computer Science 2026-01-22 Baturay Saglam , Paul Kassianik , Blaine Nelson , Sajana Weerawardhena , Yaron Singer , Amin Karbasi

Fine-tuning large language models (LLMs) for downstream tasks has become increasingly crucial due to their widespread use and the growing availability of open-source models. However, the high memory costs associated with fine-tuning remain…

Machine Learning · Computer Science 2025-02-04 David H. Yang , Mohammad Mohammadi Amiri , Tejaswini Pedapati , Subhajit Chaudhury , Pin-Yu Chen

Sparse autoencoders (SAEs) are a useful tool for uncovering human-interpretable features in the activations of large language models (LLMs). While some expect SAEs to find the true underlying features used by a model, our research shows…

Machine Learning · Computer Science 2025-01-31 Gonçalo Paulo , Nora Belrose

In this note we present a generative model of natural images consisting of a deep hierarchy of layers of latent random variables, each of which follows a new type of distribution that we call rectified Gaussian. These rectified Gaussian…

Machine Learning · Statistics 2016-03-01 Tim Salimans

Changing the behavior of large language models (LLMs) can be as straightforward as editing the Transformer's residual streams using appropriately constructed "steering vectors." These modifications to internal neural activations, a form of…

Computation and Language · Computer Science 2025-05-20 Jian-Qiao Zhu , Haijiang Yan , Thomas L. Griffiths

While sparse autoencoders (SAEs) successfully extract interpretable features from language models, applying them to audio generation faces unique challenges: audio's dense nature requires compression that obscures semantic meaning, and…

Machine Learning · Computer Science 2025-10-31 Nathan Paek , Yongyi Zang , Qihui Yang , Randal Leistikow

Behavioral patterns captured in embeddings learned from interaction data are pivotal across various stages of production recommender systems. However, in the initial retrieval stage, practitioners face an inherent tradeoff between embedding…

Information Retrieval · Computer Science 2026-02-11 Vojtěch Vančura , Martin Spišák , Rodrigo Alves , Ladislav Peška

This paper proposes a novel latent semantic learning method for extracting high-level features (i.e. latent semantics) from a large vocabulary of abundant mid-level features (i.e. visual keywords) with structured sparse representation,…

Multimedia · Computer Science 2015-03-19 Zhiwu Lu , Yuxin Peng

Large language models (LLMs) have shown remarkable capabilities across various tasks, that are learned from massive amounts of text-based data. Although LLMs can control output sequence length, particularly in instruction-based settings,…

Computation and Language · Computer Science 2025-08-22 Sangjun Moon , Dasom Choi , Jingun Kwon , Hidetaka Kamigaito , Manabu Okumura

Sparse Autoencoders (SAEs) have emerged as a popular tool for interpreting the hidden states of large language models (LLMs). By learning to reconstruct activations from a sparse bottleneck layer, SAEs discover interpretable features from…

Computer Vision and Pattern Recognition · Computer Science 2025-09-19 Matthew Lyle Olson , Musashi Hinck , Neale Ratzlaff , Changbai Li , Phillip Howard , Vasudev Lal , Shao-Yen Tseng

Understanding how Transformer-based Language Models (LMs) learn and recall information is a key goal of the deep learning community. Recent interpretability methods project weights and hidden states obtained from the forward pass to the…

Computation and Language · Computer Science 2024-02-21 Shahar Katz , Yonatan Belinkov , Mor Geva , Lior Wolf

As the development and application of Large Language Models (LLMs) continue to advance rapidly, enhancing their trustworthiness and aligning them with human preferences has become a critical area of research. Traditional methods rely…

Computation and Language · Computer Science 2024-11-06 Yuxin Xiao , Chaoqun Wan , Yonggang Zhang , Wenxiao Wang , Binbin Lin , Xiaofei He , Xu Shen , Jieping Ye

Sparse Autoencoders (SAEs) have recently emerged as powerful tools for interpreting and steering the internal representations of large language models (LLMs). However, conventional approaches to analyzing SAEs typically rely solely on…

Machine Learning · Computer Science 2025-09-24 Dong Shu , Xuansheng Wu , Haiyan Zhao , Mengnan Du , Ninghao Liu

Sparse autoencoders (SAEs) have lately been used to uncover interpretable latent features in large language models. By projecting dense embeddings into a much higher-dimensional and sparse space, learned features become disentangled and…

Machine Learning · Computer Science 2025-07-30 Viktoria Schuster

Large Language Models (LLMs) have transformed natural language processing, yet their internal mechanisms remain largely opaque. Recently, mechanistic interpretability has attracted significant attention from the research community as a…

Machine Learning · Computer Science 2025-09-24 Dong Shu , Xuansheng Wu , Haiyan Zhao , Daking Rai , Ziyu Yao , Ninghao Liu , Mengnan Du

Various Large Language Models~(LLMs) from the Generative Pretrained Transformer(GPT) family have achieved outstanding performances in a wide range of text generation tasks. However, the enormous model sizes have hindered their practical use…

Computation and Language · Computer Science 2024-04-24 Hang Shao , Bei Liu , Bo Xiao , Ke Zeng , Guanglu Wan , Yanmin Qian

Reward models (RMs) are a core component in the post-training of large language models (LLMs), serving as proxies for human preference evaluation and guiding model alignment. However, training reliable RMs under limited resources remains…

Artificial Intelligence · Computer Science 2025-11-12 Dengcan Liu , Jiahao Li , Zheren Fu , Yi Tu , Jiajun Li , Zhendong Mao , Yongdong Zhang

Despite their strong performance, Dense Passage Retrieval (DPR) models suffer from a lack of interpretability. In this work, we propose a novel interpretability framework that leverages Sparse Autoencoders (SAEs) to decompose previously…

Information Retrieval · Computer Science 2025-08-28 Seongwan Park , Taeklim Kim , Youngjoong Ko

Scaling autoregressive large language models (LLMs) has driven unprecedented progress but comes with vast computational costs. In this work, we tackle these costs by leveraging unstructured sparsity within an LLM's feedforward layers, the…

Machine Learning · Computer Science 2026-05-11 Edoardo Cetin , Stefano Peluchetti , Emilio Castillo , Akira Naruse , Mana Murakami , Llion Jones