Researchers from Japan Advanced Institute of Science and Technology (JAIST), including Professor Le-Minh Nguyen, doctoral student Dinh-Truong Do, and Dr. Nguyen-Khang Le, developed UniSpec, a plug-and-play, training-free speculative decoding framework that accelerates large language model (LLM) inference without changing model outputs or requiring additional model training. The framework automatically calibrates the optimal draft size for each hardware platform, estimates confidence scores for retrieved n-grams, and builds a more effective draft tree through confidence-guided expansion. The team also introduced Multi-SpecBench, a multilingual benchmark spanning seven languages and seven generation tasks. These innovations delivered up to 2.6× faster inference than existing training-free speculative decoding methods across multiple LLM architectures, hardware platforms, and languages while producing outputs identical to standard autoregressive decoding. The researchers publicly released both the UniSpec implementation and the Multi-SpecBench benchmark. The paper was made available online on July 01, 2026, and was published in Volume 1: Long Papers, pages 6288–6310 in the Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics, which took place between July 2–7, 2026. The paper was selected as an ACL 2026 Best Paper nominee and received the SAC Highlight Award.