English
Related papers

Related papers: Steerable but Not Decodable: Function Vectors Oper…

200 papers

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

Reasoning in language models is difficult to evaluate: natural-language traces are unverifiable, symbolic datasets are too small, and most benchmarks conflate heuristics with inference. We present FOL-Traces, the first large-scale dataset…

Artificial Intelligence · Computer Science 2026-01-27 Isabelle Lee , Sarah Liaw , Dani Yogatama

Steering vectors offer a training-free mechanism for controlling reasoning behaviors in large language models, but constructing effective vectors requires identifying genuine behavioral signals in the model's hidden states. For behaviors…

Computation and Language · Computer Science 2026-04-03 Haomin Zhuang , Hojun Yoo , Xiaonan Luo , Kehan Guo , Xiangliang Zhang

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

Multimodal LLMs (MLLMs) have reached remarkable levels of proficiency in understanding multimodal inputs. However, understanding and interpreting the behavior of such complex models is a challenging task, not to mention the dynamic shifts…

Artificial Intelligence · Computer Science 2025-08-14 Pegah Khayatan , Mustafa Shukor , Jayneel Parekh , Arnaud Dapogny , Matthieu Cord

Activation steering has emerged as a promising alternative for controlling language-model behavior at inference time by modifying intermediate representations while keeping model parameters frozen. However, large-scale evaluations such as…

Computation and Language · Computer Science 2026-05-08 Zehao Jin , Ruixuan Deng , Junran Wang , Xinjie Shen , Chao Zhang

Steering vectors have emerged as a lightweight and effective approach for aligning large language models (LLMs) at inference time, enabling modulation over model behaviors by shifting LLM representations towards a target behavior. However,…

Machine Learning · Computer Science 2026-04-07 Soham Gadgil , Chris Lin , Su-In Lee

Large Language Models (LLMs) can perform new tasks from in-context demonstrations, a phenomenon known as in-context learning (ICL). Recent work suggests that these demonstrations are compressed into task vectors (TVs), compact task…

Computation and Language · Computer Science 2026-05-04 Haolin Yang , Hakaze Cho , Kaize Ding , Naoya Inoue

Activation steering is a widely used approach for controlling large language model (LLM) behavior by intervening on internal representations. Existing methods largely rely on the Linear Representation Hypothesis, assuming behavioral…

Artificial Intelligence · Computer Science 2026-03-24 Shivam Raval , Hae Jin Song , Linlin Wu , Abir Harrasse , Jeff M. Phillips , Fazl Barez , Amirali Abdullah

Inference-time LLM alignment methods, particularly activation steering, offer an alternative to fine-tuning by directly modifying activations during generation. Existing methods, however, often rely on non-anticipative interventions that…

Machine Learning · Computer Science 2026-04-22 Julian Skifstad , Xinyue Annie Yang , Glen Chou

Vision-Language Models (VLMs) have recently demonstrated incredible strides on diverse vision language tasks. We dig into vision-based deductive reasoning, a more sophisticated but less explored realm, and find previously unexposed…

Artificial Intelligence · Computer Science 2024-10-02 Yizhe Zhang , He Bai , Ruixiang Zhang , Jiatao Gu , Shuangfei Zhai , Josh Susskind , Navdeep Jaitly

How interpretable are the features of leading vision models? The question is increasingly pressing as these models move from research benchmarks into high-stakes deployments, yet existing methods cannot answer it reliably. We close this gap…

Computer Vision and Pattern Recognition · Computer Science 2026-05-21 Julien Colin , Lore Goetschalckx , Nuria Oliver , Thomas Serre

Recent advances in mechanistic interpretability have revealed that large language models (LLMs) develop internal representations corresponding not only to concrete entities but also distinct, human-understandable abstract concepts and…

Machine Learning · Computer Science 2025-12-01 Rio Alexa Fear , Payel Mukhopadhyay , Michael McCabe , Alberto Bietti , Miles Cranmer

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

Recent work in activation steering has demonstrated the potential to better control the outputs of Large Language Models (LLMs), but it involves finding steering vectors. This is difficult because engineers do not typically know how…

Computation and Language · Computer Science 2023-12-08 Ole Jorgensen , Dylan Cope , Nandi Schoots , Murray Shanahan

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 is a promising technique for controlling LLM behavior by adding semantically meaningful vectors directly into a model's hidden states during inference. It is often framed as a precise, interpretable, and potentially…

Machine Learning · Computer Science 2026-02-17 Anton Korznikov , Andrey Galichin , Alexey Dontsov , Oleg Y. Rogov , Ivan Oseledets , Elena Tutubalina

Vision-Language Model (VLM) have gained widespread adoption in Open-Vocabulary (OV) object detection and segmentation tasks. Despite they have shown promise on OV-related tasks, their effectiveness in conventional vision tasks has thus far…

Computer Vision and Pattern Recognition · Computer Science 2025-04-15 Yongchao Feng , Yajie Liu , Shuai Yang , Wenrui Cai , Jinqing Zhang , Qiqi Zhan , Ziyue Huang , Hongxi Yan , Qiao Wan , Chenguang Liu , Junzhe Wang , Jiahui Lv , Ziqi Liu , Tengyuan Shi , Qingjie Liu , Yunhong Wang

Pretrained vision-language models (VLMs) can make semantic and visual inferences across diverse settings, providing valuable common-sense priors for robotic control. However, effectively grounding this knowledge in robot behaviors remains…

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