PARD-2: Target-Aligned Parallel Draft Model for Dual-Mode Speculative Decoding
Speculative decoding accelerates Large Language Models (LLMs) inference by using a lightweight draft model to propose candidate tokens that are verified in parallel by the target model. However, existing draft model training objectives are not directly aligned with the inference-time goal of maximizing consecutive token acceptance. To address this issue, we reformulate the draft model optimization objective, shifting the focus from token prediction accuracy to the overall acceptance length. In this paper, we build upon PARD to propose PARD-2, a dual-mode speculative decoding framework with Confidence-Adaptive Token (CAT) optimization. This approach adaptively reweights each token to bette...
| Authors | Zihao An, Taichi Liu, Ziqiong Liu et al. |
|---|---|
| Primary category | cs.CL |
| Published | 9 May 2026 |
| Last updated at source | 9 May 2026 |
Connected in the knowledge graph
- amd/PARD2-Qwen3-8BModel · cited by
- amd/PARD2-Llama-3.1-8BModel · cited by
Models and datasets referencing this paper
- amd/PARD2-Qwen3-8BModel
- amd/PARD2-Llama-3.1-8BModel
Silytics shows public metadata from arXiv and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.