停止翻转：面向快速可撤销扩散解码的上下文保持验证
Stop the Flip-Flop: Context-Preserving Verification for Fast Revocable Diffusion Decoding
AI总结：
针对并行扩散解码中因激进并行导致的翻转振荡问题，提出COVER方法，通过KV缓存覆盖和稳定性感知评分实现单次前向传递中的留一验证与稳定草稿，减少不必要修订并加速解码。
摘要：
并行扩散解码可以通过每步解掩多个令牌来加速扩散语言模型推理，但激进的并行常常损害质量。可撤销解码通过重新检查早期令牌来缓解这一问题，然而我们观察到现有的验证方案频繁触发翻转振荡，即令牌被重新掩码后又原样恢复。这种行为以两种方式减慢推理：重新掩码已验证位置削弱了并行草稿的条件上下文，且重复的重新掩码循环消耗修订预算而进展甚微。我们提出COVER（用于高效修订的缓存覆盖验证），它在单次前向传递中执行留一验证和稳定草稿。COVER通过KV缓存覆盖构建两种注意力视图：选定的种子被掩码用于验证，而其缓存的键值状态被注入到所有其他查询中以保留上下文信息，同时通过闭式对角校正防止种子位置的自泄漏。COVER进一步使用稳定性感知评分对种子进行优先级排序，该评分平衡不确定性、下游影响和缓存漂移，并自适应调整每步验证的种子数量。在多个基准测试中，COVER显著减少不必要的修订，实现更快的解码，同时保持输出质量。
英文摘要：
Parallel diffusion decoding can accelerate diffusion language model inference by unmasking multiple tokens per step, but aggressive parallelism often harms quality. Revocable decoding mitigates this by rechecking earlier tokens, yet we observe that existing verification schemes frequently trigger flip-flop oscillations, where tokens are remasked and later restored unchanged. This behaviour slows inference in two ways: remasking verified positions weakens the conditioning context for parallel drafting, and repeated remask cycles consume the revision budget with little net progress. We propose COVER (Cache Override Verification for Efficient Revision), which performs leave-one-out verification and stable drafting within a single forward pass. COVER constructs two attention views via KV cache override: selected seeds are masked for verification, while their cached key value states are injected for all other queries to preserve contextual information, with a closed form diagonal correction preventing self leakage at the seed positions. COVER further prioritises seeds using a stability aware score that balances uncertainty, downstream influence, and cache drift, and it adapts the number of verified seeds per step. Across benchmarks, COVER markedly reduces unnecessary revisions and yields faster decoding while preserving output quality.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
