让大模型先选关键信息，再处理标点，提升推理准确性。
Information-Guided Frontier Decoding: Contextual Utility-Driven Commitment in dMLLMs
- 用信息量、邻近不确定性等三维度排序候选词，优先选重要语义词
- 在多个任务和模型上均优于现有方法，错误率显著降低
- 无需训练或额外计算，可直接部署于现有扩散多模态模型
扩散多模态语言模型（dMLLMs）的解码质量高度依赖于掩码标记的提交顺序。现有基于置信度的策略倾向于优先提交局部容易的标记，但置信度并不等于上下文有用性。因此，结构上简单的标记如标点可能先于有信息量的语义锚点被提交，削弱了上下文传播并加剧误差累积。本文提出信息引导的前沿解码（IGFD），一种无需训练的解码策略，通过标记置信度、邻域不确定性及结构提交风险对候选项进行排序。IGFD鼓励早期提交可靠的语义锚点，推迟脆弱的结构标记，增强解码过程中的上下文支持。动态候选前沿进一步在相同解码预算下限制选择范围，仅允许在局部可扩展区域进行拓展。该方法无需额外训练、辅助模型或前向传播。在多模态理解、推理、定位及幻觉检测等多个基准测试中，IGFD在多数情况下超越现有解码策略，并在多种扩散型多模态大模型上保持一致优势。
原文摘要 · Abstract (English)
Decoding quality in diffusion multimodal language models (dMLLMs) depends heavily on the order in which masked tokens are committed. Existing confidence-based strategies prioritize locally easy tokens, but confidence does not necessarily reflect contextual usefulness. As a result, structurally easy tokens such as punctuation may be committed before informative semantic anchors, weakening context propagation and increasing error accumulation. We propose Information-Guided Frontier Decoding (IGFD), a training-free decoding strategy that ranks candidates using token confidence, neighborhood uncertainty, and structural commitment risk. IGFD encourages early commitment of reliable semantic anchors while delaying fragile structural tokens, improving contextual support during decoding. A dynamic candidate frontier further constrains token selection to locally expandable regions under the same decoding budget. The method requires no additional training, auxiliary models, or extra forward passes. Experiments across multimodal understanding, reasoning, grounding, and hallucination benchmarks show that IGFD consistently outperforms existing decoding strategies across the majority of benchmarks and diffusion MLLM backbones under identical decoding budgets.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。