English
Related papers

Related papers: Conceptors for Semantic Steering

200 papers

Steering vectors are a promising approach to aligning language model behavior at inference time. In this paper, we propose a framework to assess the limitations of steering vectors as alignment mechanisms. Using a framework of transformer…

Computation and Language · Computer Science 2025-05-05 Chebrolu Niranjan , Kokil Jaidka , Gerard Christopher Yeo

Concept Bottleneck Models (CBMs) provide a basis for semantic abstractions within a neural network architecture. Such models have primarily been seen through the lens of interpretability so far, wherein they offer transparency by inferring…

Computer Vision and Pattern Recognition · Computer Science 2025-12-09 Deepika SN Vemuri , Gautham Bellamkonda , Aditya Pola , Vineeth N Balasubramanian

Having an LLM that aligns with human preferences is essential for accommodating individual needs, such as maintaining writing style or generating specific topics of interest. The majority of current alignment methods rely on fine-tuning or…

Computation and Language · Computer Science 2025-03-06 Bingqing Song , Boran Han , Shuai Zhang , Hao Wang , Haoyang Fang , Bonan Min , Yuyang Wang , Mingyi Hong

As large language models (LLMs) become widely deployed, concerns about their safety and alignment grow. An approach to steer LLM behavior, such as mitigating biases or defending against jailbreaks, is to identify which parts of a prompt…

Computation and Language · Computer Science 2025-05-20 Kenza Amara , Rita Sevastjanova , Mennatallah El-Assady

Large Language Models (LLMs), despite advances in instruction tuning, often fail to follow complex user instructions. Activation steering techniques aim to mitigate this by manipulating model internals, but have a potential risk of…

Machine Learning · Computer Science 2026-03-10 Minjae Kang , Jaehyung Kim

Despite the remarkable success of Multimodal Large Language Models (MLLMs) across diverse tasks, the internal mechanisms governing how they encode and ground distinct visual concepts remain poorly understood. To bridge this gap, we propose…

Artificial Intelligence · Computer Science 2026-05-08 Zehao Deng , Tianjie Ju , Zheng Wu , Liangbo He , Jun Lan , Huijia Zhu , Weiqiang Wang , Zhuosheng Zhang

Steering methods influence Large Language Model behavior by identifying semantic directions in hidden representations, but are typically realized through inference-time activation interventions that apply a fixed, global modification to the…

Computation and Language · Computer Science 2026-03-04 Chung-En Sun , Ge Yan , Zimo Wang , Tsui-Wei Weng

Recent advances in Large Reasoning Models (LRMs) have demonstrated remarkable capabilities in solving complex tasks such as mathematics and coding. However, these models frequently exhibit a phenomenon known as overthinking during…

Machine Learning · Computer Science 2025-11-18 Yao Huang , Huanran Chen , Shouwei Ruan , Yichi Zhang , Xingxing Wei , Yinpeng Dong

Recent work has shown that LLMs can sometimes detect when steering vectors are injected into their residual stream and identify the injected concept -- a phenomenon termed "introspective awareness." We investigate the mechanisms underlying…

Machine Learning · Computer Science 2026-05-18 Uzay Macar , Li Yang , Atticus Wang , Peter Wallich , Emmanuel Ameisen , Jack Lindsey

Large Reasoning Models (LRMs) excel at complex reasoning tasks, but their efficiency is often hampered by overly verbose outputs. Prior steering methods attempt to address this issue by applying a single, global vector to hidden…

Machine Learning · Computer Science 2026-02-06 Yawei Li , Benjamin Bergner , Yinghan Zhao , Vihang Prakash Patil , Bei Chen , Cheng Wang

This work characterizes large language models' chain-of-thought generation as a structured trajectory through representation space. We show that mathematical reasoning traverses functionally ordered, step-specific subspaces that become…

Computation and Language · Computer Science 2026-04-08 Lihao Sun , Hang Dong , Bo Qiao , Qingwei Lin , Dongmei Zhang , Saravan Rajmohan

Understanding how large audio models represent music, and using that understanding to steer generation, is both challenging and underexplored. Inspired by mechanistic interpretability in language models, where direction vectors in…

Sparse autoencoders have become a standard tool for uncovering interpretable latent representations in neural networks. Yet salient concepts often span manifolds that current linear methods cannot capture without post hoc analysis. This…

Machine Learning · Computer Science 2026-05-12 Thomas Dooms , Ward Gauderis , Geraint Wiggins , Jose Oramas

Large language models can be steered at inference time through prompting or activation interventions, but activation steering methods often underperform compared to prompt-based approaches. We propose a framework that formulates prompt…

Computation and Language · Computer Science 2026-05-06 Geert Heyman , Frederik Vandeputte

Large language models handle single-turn generation well, but multi-turn interactions still require the model to reconstruct user intent and task state from an expanding token history because internal representations do not persist across…

Computation and Language · Computer Science 2025-12-11 Vishwas Hegde , Vindhya Shigehalli

Alignment in LLMs is more brittle than commonly assumed: misalignment can be triggered by adversarial prompts, benign fine-tuning, emergent misalignment, and goal misgeneralization. Recent evidence suggests that some misalignment behaviors…

Artificial Intelligence · Computer Science 2026-04-10 Niklas Herbster , Martin Zborowski , Alberto Tosato , Gauthier Gidel , Tommaso Tosato

Activation steering, or representation engineering, offers a lightweight approach to align large language models (LLMs) by manipulating their internal activations at inference time. However, current methods suffer from two key limitations:…

Artificial Intelligence · Computer Science 2026-02-24 Hongjue Zhao , Haosen Sun , Jiangtao Kong , Xiaochang Li , Qineng Wang , Liwei Jiang , Qi Zhu , Tarek Abdelzaher , Yejin Choi , Manling Li , Huajie Shao

Concept-based models perform prediction using a set of concepts that are interpretable to stakeholders. However, such models often involve a fixed, large number of concepts, which may place a substantial cognitive load on stakeholders. We…

Human-Computer Interaction · Computer Science 2023-06-16 Matthew Barker , Katherine M. Collins , Krishnamurthy Dvijotham , Adrian Weller , Umang Bhatt

There has been considerable recent interest in interpretable concept-based models such as Concept Bottleneck Models (CBMs), which first predict human-interpretable concepts and then map them to output classes. To reduce reliance on…

Machine Learning · Computer Science 2024-07-08 Simon Schrodi , Julian Schur , Max Argus , Thomas Brox

While large language models (LLMs) have seen unprecedented advancements in capabilities and applications across a variety of use-cases, safety alignment of these models is still an area of active research. The fragile nature of LLMs, even…

Computation and Language · Computer Science 2024-10-03 Amrita Bhattacharjee , Shaona Ghosh , Traian Rebedea , Christopher Parisien