repoGitHubTrust 82 · PrimaryPublished 1mo agoLive · 1mo ago
VCG-team/DiffSegmenter
(TIP 2025, CCF-A) Diffusion Model is Secretly a Training-free Open Vocabulary Semantic Segmenter
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- PossiblePossibly related (embedding) · 58%Adaptive Block Diffusion: Resolving Training-Inference Mismatch in Diffusion Language Models →
- PossiblePossibly related (embedding) · 54%Learning Unmasking Policies for Diffusion Language Models - Apple Machine Learning Research →
- PossiblePossibly related (embedding) · 50%What if context compression is a diffusion noise function? Proposal + honest results from untrained-model experiments [R] →
- PossiblePossibly related (embedding) · 49%Multi-Block Diffusion Language Models →
- PossiblePossibly related (embedding) · 49%DiffusionGemma: 4x faster text generation →
- PossiblePossibly related (embedding) · 51%Scaling Properties of Continuous Diffusion Spoken Language Models - Apple Machine Learning Research →
- PossiblePossibly related (embedding) · 50%Erasing Without Collateral Damage: Precise Concept Removal in Diffusion Models →
- PossiblePossibly related (embedding) · 53%TILDE: TILt-based Distributional Erasure for Concept Unlearning →
Implements
Covers
Covers (incoming)
Implements (incoming)
paperErasing Without Collateral Damage: Precise Concept Removal in Diffusion ModelspaperTILDE: TILt-based Distributional Erasure for Concept UnlearningpaperAnalysis-by-Proxy: Localization Signals in VLMs Operating as Condition EncoderspaperVocaDet: Sample-Driven Open-Vocabulary Object Detection and Segmentation via Visual Tokenization and Vector Database Retrieval
Related across the graph
newsDiffusionGemma: 4x faster text generationpaperTILDE: TILt-based Distributional Erasure for Concept UnlearningpaperMulti-Block Diffusion Language ModelspaperAnalysis-by-Proxy: Localization Signals in VLMs Operating as Condition EncodersnewsLearning Unmasking Policies for Diffusion Language Models - Apple Machine Learning ResearchpaperVocaDet: Sample-Driven Open-Vocabulary Object Detection and Segmentation via Visual Tokenization and Vector Database RetrievalnewsWhat if context compression is a diffusion noise function? Proposal + honest results from untrained-model experiments [R]newsScaling Properties of Continuous Diffusion Spoken Language Models - Apple Machine Learning ResearchpaperErasing Without Collateral Damage: Precise Concept Removal in Diffusion ModelspaperAdaptive Block Diffusion: Resolving Training-Inference Mismatch in Diffusion Language Models