LogicDiff: Logic-Guided Denoising Improves Zero-Shot Reasoning in Masked Diffusion Language Models
Masked diffusion language models (MDLMs) generate text by iteratively unmasking tokens from a fully masked sequence. Their standard confidence-based unmasking strategy systematically defers high-entropy logical connective tokens, degrading reasoning performance. We introduce LogicDiff, an inference-time method that replaces confidence-based unmasking with logic-role-guided unmasking. A lightweight classification head (4.2M parameters, 0.05% of the base model) predicts the logical role of each masked position (premise, connective, derived step, conclusion, or filler) from the base model's hidden states with 98.4% accuracy, and a dependency-ordered scheduler unmasks tokens in logical order. In zero-shot settings, LogicDiff improves LLaDA-8B-Instruct accuracy from 22.0% to 60.7% on GSM8K (+38.7 percentage points) and from 23.6% to 29.2% on MATH-500 (+5.6 pp), with less than 6% speed overhead. However, with 8-shot chain-of-thought prompting, the baseline reaches approximately 70% and LogicDiff provides no additional improvement. Analysis reveals that few-shot prompting implicitly resolves the same ordering problem that LogicDiff explicitly addresses, and that fixed role-based ordering can cause premature commitment to numerical values before sufficient context is available. Our results characterize the Flexibility Trap as primarily a zero-shot phenomenon and identify context-adaptive ordering as a key direction for future work.
Code (0)
등록된 구현이 없습니다.
Results from the Paper
Similar Papers 제목 키워드 기반
Zero-Shot Denoising for Fluorescence Lifetime Imaging Microscopy with Intensity-Guided Learning
Multimodal and multi-information microscopy techniques such as Fluorescence Lifetime Imaging Microscopy (FLIM) extend the informational channels beyond intensity-based fluorescence microscopy but suffer from reduced imag…
DenoisingIGDNet: Zero-Shot Robust Underexposed Image Enhancement via Illumination-Guided and Denoising
Current methods for restoring underexposed images typically rely on supervised learning with paired underexposed and well-illuminated images. However, collecting such datasets is often impractical in real-world scenarios…
Image EnhancementZERO-IG: Zero-Shot Illumination-Guided Joint Denoising and Adaptive Enhancement for Low-Light Images
This paper presents a novel zero-shot method for jointly denoising and enhancing real-word low-light images. The proposed method is independent of training data and noise distribution. Guided by illumination we integ…
DenoisingZo3T: Zero-Shot 3D-Aware Trajectory-Guided Image-to-Video Generation via Test-Time Training
Trajectory-Guided image-to-video (I2V) generation aims to synthesize videos that adhere to user-specified motion instructions. Existing methods typically rely on computationally expensive fine-tuning on scarce annotated …
Video GenerationA Zero-Shot Deep Image Prior Framework for Denoising and Deconvolution in Fluorescence Microscopy
Fluorescence microscopy images are degraded by noise and diffraction-induced blur, which compromise structural fidelity and limit quantitative analysis. Supervised deep learning methods achieve impressive restoration per…