@inproceedings{shahriar-etal-2026-meav,
    title = "{MEAV}: Model Editing with Alignment Vectors for inference time {LLM} alignment in single and multidomain preference spectrum",
    author = "Shahriar, Sadat  and
      Qi, Zheng  and
      Pappas, Nikolaos  and
      Doss, Srikanth  and
      Halder, Kishaloy  and
      Sunkara, Monica  and
      Mager, Manuel  and
      Benajiba, Yassine",
    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.2035/",
    pages = "40972--40985",
    ISBN = "979-8-89176-395-1",
    abstract = "Aligning Large Language Models (LLM) to address subjectivity and nuanced preference levels requires adequate flexibility and control, which can be a resource-intensive and time-consuming procedure. Existing training-time alignment methods require full re-training when a change is needed and inference-time ones typically require access to the reward model at each inference step. We introduce **MEAV**, an inference-time model-editing-based LLM alignment method that learns encoded representations of preference dimensions, called *Alignment Vectors* (AV). These representations enable dynamic adjusting of the model behavior during inference through simple linear operations. Here, we focus on three gradual response levels across three specialized domains: medical, legal, and financial, exemplifying its practical potential. This new alignment paradigm introduces adjustable preference knobs during inference, allowing users to tailor their LLM outputs while reducing the inference cost by half compared to the prompt engineering approach. Additionally, we find that AVs are transferable across different fine-tuning stages of the same model, demonstrating their flexibility. AVs also facilitate multidomain, diverse preference alignment, making the process 12x faster than the retraining approach."
}