arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.09555cs.CL
收藏

推进块扩散语言模型用于测试时间扩展

Adaptive Test-Time Compute Allocation for Block Diffusion Language Models in Complex Reasoning

Fudan University(复旦大学);Peking University(北京大学)

Yi Lu, Deyang Kong, Jianing Wang, Linsen Guo, Xue Wang, Qi Guo, Tao Gui, Xuanjing Huang, Wei Ye, Shikun Zhang, Wei Wang

AI总结:

本文提出BACD和TCCF方法,提升块扩散语言模型在测试时间扩展中的推理效率和效果。

摘要:

近期块扩散语言模型(BDLM)的进步在推理任务中展示了竞争性的性能和强大的可扩展性。然而,现有的BDLM在测试时间扩展设置下探索有限,并在长链推理中面临更严重的解码挑战,特别是在平衡解码速度和有效性方面。在本文中,我们提出了一种统一的BDLM测试时间扩展框架,该框架在解码和块级生成中引入了适应性。在解码层面,我们提出了有界自适应置信解码(BACD),一种基于模型置信度的采样策略,能够动态调整去噪过程,从而加速推理并控制误差积累。除了逐步适应性之外,我们引入了Think Coarse, Critic Fine(TCCF)测试时间扩展范式,该范式为探索性推理分配较大的块大小,为细化阶段分配较小的块大小,从而实现有效的效率-有效性平衡。为了使大块大小解码高效且有效,我们采用了渐进块大小扩展,这可以缓解在扩展块大小时的性能下降。广泛的实验表明,将BACD和TCCF应用于TDAR-8B在强基线模型如TraDo-8B上取得了显著改进(2.26倍的速度提升,+11.2分在AIME24上)。这些结果标志着在复杂推理任务中解锁BDLM测试时间扩展潜力的重要一步。

英文摘要:

Recent advances in block diffusion language models have demonstrated competitive performance and strong scalability on reasoning tasks. However, their test-time compute allocation remains largely unexplored, leaving a critical speed-effectiveness trade-off unresolved in long Chain-of-Thought reasoning. To address this, we propose a unified test-time compute allocation framework that introduces adaptivity in both step-wise decoding and blockwise generation. At the decoding level, we propose Bounded Adaptive Confidence Decoding (BACD), a difficulty-aware sampling strategy that dynamically adjusts denoising based on model confidence, accelerating inference while controlling error accumulation. Beyond step-wise adaptivity, we introduce the Think Coarse, Critic Fine (TCCF) paradigm that allocates large block sizes for exploratory thinking and smaller block sizes for precise refinement. To stabilize training under varying block configurations, we adopt Progressive Block Size Extension, which mitigates quality degradation when scaling up block sizes. Extensive evaluations on six benchmarks show that our TDAR-8B model with BACD and TCCF achieves a 2.38$\times$ speedup and +3.4% average accuracy over the strong TraDo-8B baseline, unlocking the potential of block diffusion in complex reasoning.

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。

↑