ProVe：让大模型自己出草案再自验，解码提速2.5倍
ProVe: Proposer-Verifier Decoding
一项把投机解码反着来的解码新方法，32B 模型自己出草稿再验证，吞吐提2.5倍还不用训草稿头，推理部署的可以看看。
ProVe 提出反转投机解码的角色：将 32B 大模型微调为提议者，一次前向预测 16 个 token 的块，再用一次前向验证，验证器可用 3B 小模型或提议者本身。提议者自验的 ProVe-lossless 在六个基准上平均达到 AR 吞吐的 2.5 倍，且不需要单独训练草稿头。搭配 3B 验证器时吞吐达 2.7 倍，六任务平均精度比 AR 低 2.6 分，可在速度与精度间权衡。在 GSM8K 上块大小为 16 时，平均有 6.6 个 token 的前缀与 AR 解码匹配。
ProVe: Proposer-Verifier Decoding
Speculative decoding accelerates autoregressive (AR) language models by letting a small model draft tokens serially and a large model verify them in parallel. We reverse these roles. Starting from a standard AR checkpoint, we fine-tune the large model into a proposer that predicts a block of future tokens in a single forward pass, then use one additional forward pass to verify the proposed block. This gives ProVe, Proposer-Verifier decoding. The verifier can be a smaller AR model or the proposer itself; the latter gives ProVe-lossless, which reproduces the proposer's greedy AR decoding. Block proposals already provide substantial room for acceleration: with a block size of 16, the proposed block has an average AR-matching prefix of 6.6 tokens on GSM8K. Across six benchmarks with a 32B proposer, ProVe-lossless reaches 2.5x average AR throughput without training a separate draft head. With a small 3B verifier, ProVe reaches 2.7x average throughput at 2.6 accuracy points below AR on the six-task average, providing a controllable tradeoff between speed and accuracy.