paper-with-me

홈 › Papers

Prompt-Activation Duality: Improving Activation Steering via Attention-Level Interventions

2026-05-11 · Diancheng Kang, Zheyuan Liu, Ningshan Ma, Yue Huang, Zhaoxuan Tan, Meng Jiang arxiv

Activation steering controls language model behavior by adding directions to internal representations at inference time, but standard residual-stream steering can fail in stateful dialogue. We identify KV-cache contamination as a key failure mode: steered token states are stored and repeatedly reused, turning a local perturbation into cumulative coherence degradation. To address this challenge, we propose Gated Cropped Attention-Delta steering (GCAD), which extracts steering signals from system-prompt contributions to self-attention and applies them with token-level gating. Across persona-steering experiments, GCAD preserves trait control while substantially improving long-horizon coherence. On the main multi-turn benchmark, GCAD improves average coherence drift from -18.6 to -1.9 and raises turn-10 trait expression from 78.0 to 93.1. These results suggest that activation steering becomes more reliable when interventions follow the prompt-mediated pathways that models already use for behavioral control.

📄 PDF Abstract BibTeX arXiv:2605.10664

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Steer Like the LLM: Activation Steering that Mimics Prompting

2026-05-05 · Geert Heyman, Frederik Vandeputte arxiv

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 fo…

Steering Where to Listen: Instruction-Based Activation Steering Redirects Temporal Attention in Large Audio-Language Models

2026-06-09 · Tsung-En Lin, Hung-Yi Lee arxiv

Large Audio-Language Models (LALMs) excel at audio understanding but expose little about where in an audio signal they attend. We introduce instruction-based vector steering, which constructs a steering vector by contras…

TADA! Tuning Audio Diffusion Models through Activation Steering

2026-02-12 · Łukasz Staniszewski, Katarzyna Zaleska, Mateusz Modrzejewski, Kamil Deja arxiv

Audio diffusion models can synthesize high-fidelity music from text, yet achieving fine-grained control over specific musical attributes remains challenging, as their internal mechanisms for representing high-level conce…

Steered LLM Activations are Non-Surjective

2026-04-10 · Aayush Mishra, Daniel Khashabi, Anqi Liu arxiv

Activation steering is a popular white-box control technique that modifies model activations to elicit an abstract change in its behavior. It has also become a standard tool in interpretability (e.g., probing truthfulnes…

Contextual Linear Activation Steering of Language Models

2026-04-27 · Brandon Hsu, Daniel Beaglehole, Adityanarayanan Radhakrishnan, Mikhail Belkin arxiv

Linear activation steering is a powerful approach for eliciting the capabilities of large language models and specializing their behavior using limited labeled data. While effective, existing methods often apply a fixed …