English
Related papers

Related papers: Patterns and Mechanisms of Contrastive Activation …

200 papers

Sparse Autoencoders (SAEs) can extract interpretable features from large language models (LLMs) without supervision. However, their effectiveness in downstream steering tasks is limited by the requirement for contrastive datasets or large…

Computation and Language · Computer Science 2026-05-05 Seonglae Cho , Zekun Wu , Adriano Koshiyama

Large Language Models (LLMs) are increasingly deployed in high-stakes decision-making contexts. While prior work has shown that LLMs exhibit cognitive biases behaviorally, whether these biases correspond to identifiable internal…

Artificial Intelligence · Computer Science 2026-04-03 Fan Huang , Songheng Zhang , Haewoon Kwak , Jisun An

Large Language Models (LLMs) exhibit impressive capabilities but often hallucinate, confidently providing incorrect answers instead of admitting ignorance. Prior work has shown that models encode linear representations of their own…

Computation and Language · Computer Science 2025-12-09 Wannan , Yang , Xinchi Qiu , Lei Yu , Yuchen Zhang , Aobo Yang , Narine Kokhlikyan , Nicola Cancedda , Diego Garcia-Olano

Large language models have simplified the production of personalized translations reflecting predefined stylistic constraints. However, these systems still struggle when stylistic requirements are implicitly represented by a set of…

Computation and Language · Computer Science 2025-10-15 Daniel Scalena , Gabriele Sarti , Arianna Bisazza , Elisabetta Fersini , Malvina Nissim

Mixture-of-Experts (MoE) architectures in Large Language Models (LLMs) have significantly reduced inference costs through sparse activation. However, this sparse activation paradigm also introduces new safety challenges. Since only a subset…

Cryptography and Security · Computer Science 2026-05-01 Jona te Lintelo , Lichao Wu , Marina Krček , Sengim Karayalçin , Stjepan Picek

Large language models (LLMs) have achieved remarkable performance across many generation tasks. Nevertheless, effectively aligning them with desired behaviors remains a significant challenge. Activation steering is an effective and…

Computation and Language · Computer Science 2025-10-02 Zifeng Cheng , Jinwei Gan , Zhiwei Jiang , Cong Wang , Yafeng Yin , Xiang Luo , Yuchen Fu , Qing Gu

Sparse autoencoders (SAEs) have recently emerged as a powerful tool for language model steering. Prior work has explored top-k SAE latents for steering, but we observe that many dimensions among the top-k latents capture non-semantic…

Computation and Language · Computer Science 2025-10-03 Jiaqing Xie

Activation-based steering enables Large Language Models (LLMs) to exhibit targeted behaviors by intervening on intermediate activations without retraining. Despite its widespread use, the mechanistic factors that govern when steering…

Computation and Language · Computer Science 2026-03-13 Mehdi Jafari , Hao Xue , Flora Salim

Representation Engineering (RepE) is a novel paradigm for controlling the behavior of LLMs. Unlike traditional approaches that modify inputs or fine-tune the model, RepE directly manipulates the model's internal representations. As a…

Machine Learning · Computer Science 2025-10-09 Jan Wehner , Sahar Abdelnabi , Daniel Tan , David Krueger , Mario Fritz

Activation steering methods control large language model (LLM) behavior by modifying internal activations at inference time. However, most existing activation steering methods rely on a fixed steering strength, leading to either…

Computation and Language · Computer Science 2025-10-16 Arthur Vogels , Benjamin Wong , Yann Choho , Annabelle Blangero , Milan Bhan

This paper investigates how large language models (LLMs) behave when faced with discrepancies between their parametric knowledge and conflicting information contained in a prompt. Building on prior question-answering (QA) research, we…

Computation and Language · Computer Science 2025-10-23 Jaesung Bae , Cameron Churchwell , Mitchell Hermon , Tsun-An Hsieh , Jocelyn Xu , Yekaterina Yegorova , Mark Hasegawa-Johnson , Heng Ji

Controlling undesirable Large Language Model (LLM) behaviors, such as the generation of unsafe content or failing to adhere to safety guidelines, often relies on costly fine-tuning. Activation steering provides an alternative for…

Computation and Language · Computer Science 2026-03-17 Amr Hegazy , Mostafa Elhoushi , Amr Alanwar

Large language models (LLMs) struggle in knowledge-intensive tasks, as retrievers often overfit to surface similarity and fail on queries involving complex logical relations. The capacity for logical analysis is inherent in model…

Computation and Language · Computer Science 2026-02-03 Wenxuan Zhang , Yuan-Hao Jiang , Changyong Qi , Rui Jia , Yonghe Wu

State-of-the-art pre-trained image models predominantly adopt a two-stage approach: initial unsupervised pre-training on large-scale datasets followed by task-specific fine-tuning using Cross-Entropy loss~(CE). However, it has been…

Computer Vision and Pattern Recognition · Computer Science 2024-11-18 Zijun Long , George Killick , Lipeng Zhuang , Gerardo Aragon-Camarasa , Zaiqiao Meng , Richard Mccreadie

Understanding how generative models represent and transform data is a foundational problem in deep learning interpretability. While mechanistic interpretability of discriminative architectures has yielded substantial insights, relatively…

Machine Learning · Computer Science 2026-04-07 Dip Roy , Rajiv Misra , Sanjay Kumar Singh , Anisha Roy

Large Language Models (LLMs) exhibit remarkable capabilities in the hierarchical decomposition of complex tasks through semantic reasoning. However, their application in embodied systems faces challenges in ensuring reliable execution of…

Robotics · Computer Science 2025-03-04 Mingcong Lei , Ge Wang , Yiming Zhao , Zhixin Mai , Qing Zhao , Yao Guo , Zhen Li , Shuguang Cui , Yatong Han , Jinke Ren

Large language models (LLMs) excel at handling human queries, but they can occasionally generate flawed or unexpected responses. Understanding their internal states is crucial for understanding their successes, diagnosing their failures,…

Computation and Language · Computer Science 2025-02-24 Xuansheng Wu , Jiayi Yuan , Wenlin Yao , Xiaoming Zhai , Ninghao Liu

To facilitate robust and trustworthy deployment of large language models (LLMs), it is essential to quantify the reliability of their generations through uncertainty estimation. While recent efforts have made significant advancements by…

Computation and Language · Computer Science 2025-07-22 Rui Li , Jing Long , Muge Qi , Heming Xia , Lei Sha , Peiyi Wang , Zhifang Sui

The rapid advancement of large vision-language models (LVLMs) has driven significant progress in multimodal tasks, enabling models to interpret, reason, and generate outputs across both visual and textual domains. While excelling in…

Computer Vision and Pattern Recognition · Computer Science 2025-03-26 Hao Yu , Zhuokai Zhao , Shen Yan , Lukasz Korycki , Jianyu Wang , Baosheng He , Jiayi Liu , Lizhu Zhang , Xiangjun Fan , Hanchao Yu

Large language models (LLMs) risk retaining sensitive, copyrighted, or harmful information from their training data. Entity-level unlearning addresses this issue by removing all knowledge of a specific entity while preserving the model's…

Computation and Language · Computer Science 2026-01-15 Xiaoqi Han , Víctor Gutiérrez-Basulto , Ru Li , Xiaoli Li , Jiye Liang , Jeff Z. Pan