{"ID":2868511,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.16820","arxiv_id":"2509.16820","title":"DISCO: Disentangled Communication Steering for Large Language Models","abstract":"A variety of recent methods guide large language model outputs via the inference-time addition of steering vectors to residual-stream or attention-head representations. In contrast, we propose to inject steering vectors directly into the query and value representation spaces within attention heads. We provide evidence that a greater portion of these spaces exhibit high linear discriminability of concepts --a key property motivating the use of steering vectors-- than attention head outputs. We analytically characterize the effect of our method, which we term DISentangled COmmunication (DISCO) Steering, on attention head outputs. Our analysis reveals that DISCO disentangles a strong but underutilized baseline, steering attention inputs, which implicitly modifies queries and values in a rigid manner. In contrast, DISCO's direct modulation of these components enables more granular control. We find that DISCO achieves superior performance over a number of steering vector baselines across multiple datasets on LLaMA 3.1 8B and Gemma 2 9B, with steering efficacy scoring up to 19.1% higher than the runner-up. Our results support the conclusion that the query and value spaces are powerful building blocks for steering vector methods.","short_abstract":"A variety of recent methods guide large language model outputs via the inference-time addition of steering vectors to residual-stream or attention-head representations. In contrast, we propose to inject steering vectors directly into the query and value representation spaces within attention heads. We provide evidence...","url_abs":"https://arxiv.org/abs/2509.16820","url_pdf":"https://arxiv.org/pdf/2509.16820v1","authors":"[\"Max Torop\",\"Aria Masoomi\",\"Masih Eskandar\",\"Jennifer Dy\"]","published":"2025-09-20T21:56:03Z","proceeding":"cs.LG","tasks":"[\"cs.LG\"]","methods":"[\"Language Model\"]","has_code":false}