@inproceedings{ye-etal-2026-riser,
    title = "{RISER}: Orchestrating Latent Reasoning Skills for Adaptive Activation Steering",
    author = "Ye, Wencheng  and
      Yuan, Xiaoyang  and
      Bin, Yi  and
      Jin, Hengyu  and
      Peng, Liang  and
      Zeng, Pengpeng  and
      Shen, Heng Tao",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.findings-acl.226/",
    pages = "4627--4644",
    ISBN = "979-8-89176-395-1",
    abstract = "Recent work on domain-specific reasoning with large language models (LLMs) has largely relied on training-intensive approaches that require updating model parameters. Although activation steering has emerged as a parameter-efficient alternative, existing methods typically rely on static and manually designed interventions, limiting their ability to adapt to the dynamic nature of complex reasoning. To address this limitation, we propose RISER (Router-based Intervention for Steerable Enhancement of Reasoning), a plug-and-play intervention framework that adaptively steers LLM reasoning in activation space. RISER builds a library of reusable reasoning vectors and employs a lightweight Router to dynamically compose these vectors for each input. The Router is optimized via reinforcement learning under task-level rewards, enabling the emergent and compositional activation of latent cognitive primitives. Across seven diverse benchmarks, RISER achieves average zero-shot accuracy improvements of 3.4{--}6.5{\%} over the base model, while outperforming chain-of-thought-style reasoning with 2{--}3{\texttimes} higher token efficiency and robust accuracy gains. Further analysis demonstrates that RISER autonomously combines multiple vectors into interpretable and precise control strategies, pointing toward more controllable and efficient LLM reasoning."
}