paper-with-me

Papers

Steering Code LLMs with Activation Directions for Language and Library Control

2026-03-24 · Md Mahbubur Rahman, Arjun Guha, Harshitha Menon arxiv

Code LLMs often default to particular programming languages and libraries under neutral prompts. We investigate whether these preferences are encoded as approximately linear directions in activation space that can be manipulated at inference time. Using a difference-in-means method, we estimate layer-wise steering vectors for five language/library pairs and add them to model hidden states during generation. Across three open-weight code LLMs, these interventions substantially increase generation toward the target ecosystem under neutral prompts and often remain effective even when prompts explicitly request the opposite choice. Steering strength varies by model and target, with common ecosystems easier to induce than rarer alternatives, and overly strong interventions can reduce output quality. Overall, our results suggest that code-style preferences in LLMs are partly represented by compact, steerable structure in activation space.

📄 PDF Abstract BibTeX arXiv:2603.23629

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

FairSteer: Inference Time Debiasing for LLMs with Dynamic Activation Steering

2025-04-20 · Yichen Li, Zhiting Fan, Ruizhe Chen, Xiaotang Gai 외

Large language models (LLMs) are prone to capturing biases from training corpus, leading to potential negative social impacts. Existing prompt-based debiasing methods exhibit instability due to their sensitivity to promp…

counterfactualFairnessQuestion AnsweringText Generation

Do LLMs Experience an Internal Polylogue? Investigating Reasoning through the Lens of Personas

2026-05-09 · Nils A. Herrmann, Leander Girrbach, Kirill Bykov, Zeynep Akata arxiv

Recent work shows that large language models (LLMs) encode behavioural traits ("personas") as linear directions in activation space, often called "persona vectors". Prior work has used such directions as static handles f…

Neural FOXP2 -- Language Specific Neuron Steering for Targeted Language Improvement in LLMs

2026-02-01 · Anusa Saha, Tanmay Joshi, Vinija Jain, Aman Chadha 외 arxiv

LLMs are multilingual by training, yet their lingua franca is often English, reflecting English language dominance in pretraining. Other languages remain in parametric memory but are systematically suppressed. We argue t…

Activation Steering with a Feedback Controller

2025-10-05 · Dung V. Nguyen, Hieu M. Vu, Nhi Y. Pham, Lei Zhang 외 arxiv

Controlling the behaviors of large language models (LLM) is fundamental to their safety alignment and reliable deployment. However, existing steering methods are primarily driven by empirical insights and lack theoretica…

Linear Personality Probing and Steering in LLMs: A Big Five Study

2025-12-19 · Michel Frising, Daniel Balcells arxiv

Large language models (LLMs) exhibit distinct and consistent personalities that greatly impact trust and engagement. While this means that personality frameworks would be highly valuable tools to characterize and control…

Prompt Engineering