Thinking Diffusion: 通过惩罚和引导在扩散多模态语言模型中抑制和引导视觉推理
Thinking Diffusion: Penalize and Guide Visual-Grounded Reasoning in Diffusion Multimodal Language Models
AI总结：
本文针对扩散多模态语言模型在视觉推理中生成过早答案的问题，提出位置与步数惩罚和视觉推理引导方法，提升推理准确率并加快推理速度。
摘要：
扩散大语言模型（dLLMs）正逐渐成为自回归（AR）LLMs的有前途的替代品。最近，这一范式已扩展到多模态任务，导致开发了扩散多模态大语言模型（dMLLMs）。这些模型预计能够保留LLMs的推理能力，同时通过并行生成实现更快的推理。然而，当与链式思维（CoT）推理结合时，dMLLMs表现出两个关键问题。首先，我们观察到dMLLMs经常在非常早的时间步生成最终答案标记。这种趋势表明，模型在充分推理之前就决定了答案，导致推理性能下降。其次，在初始时间步，dMLLMs对视觉提示的依赖极低，表现出与AR视觉-语言模型根本不同的视觉信息利用模式。总之，这些发现表明，dMLLMs倾向于在充分依赖视觉输入之前生成过早的答案。为了解决这些限制，我们提出了位置与步数惩罚（PSP）和视觉推理引导（VRG）。PSP在早期时间步惩罚后期位置的标记，延迟提前答案生成，鼓励跨时间步的逐步推理。VRG受分类器自由引导启发，放大视觉 grounding 信号以增强模型与视觉证据的一致性。在各种dMLLMs上的广泛实验表明，我们的方法在准确率上达到最高7.5%的提升，同时相比使用四倍更多扩散步骤的推理，速度提升超过3倍。
英文摘要：
Diffusion large language models (dLLMs) are emerging as promising alternatives to autoregressive (AR) LLMs. Recently, this paradigm has been extended to multimodal tasks, leading to the development of diffusion multimodal large language models (dMLLMs). These models are expected to retain the reasoning capabilities of LLMs while enabling faster inference through parallel generation. However, when combined with Chain-of-Thought (CoT) reasoning, dMLLMs exhibit two critical issues. First, we observe that dMLLMs often generate the final answer token at a very early timestep. This trend indicates that the model determines the answer before sufficient reasoning, leading to degraded reasoning performance. Second, during the initial timesteps, dMLLMs show minimal dependency on visual prompts, exhibiting a fundamentally different pattern of visual information utilization compared to AR vision-language models. In summary, these findings indicate that dMLLMs tend to generate premature final answers without sufficiently grounding on visual inputs. To address these limitations, we propose Position and Step Penalty (PSP) and Visual Reasoning Guidance (VRG). PSP penalizes tokens in later positions during early timesteps, delaying premature answer generation and encouraging progressive reasoning across timesteps. VRG, inspired by classifier-free guidance, amplifies visual grounding signals to enhance the model's alignment with visual evidence. Extensive experiments across various dMLLMs demonstrate that our method achieves up to 7.5% higher accuracy while delivering more than 3x speedup compared to reasoning with four times more diffusion steps.
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。
