9 papers
GHOST: Geometry-Constrained Hybrid Diffusion for Open-Vocabulary 3D Scene Graph Generation
Changqun Feng, Wangxiandi Yin, Xin Hu +4
3D scene graph generation requires understanding both object semantics and metric spatial relations. However, existing open-vocabulary methods often struggle with spatial predicate…
Visual Graph Reasoning via Knowledge Compilation
Rongzheng Wang, Zhe Wang, Ke Qin +6
Visual graph reasoning requires answering graph-theoretic questions directly from graph images, where graph topology and state are conveyed visually rather than given in symbolic f…
UniDFKD: A Unified Semantic Prior Framework for Architecture-Agnostic Data-Free Knowledge Distillation
Xuewan He, Tong Chu, Zihan Cheng +5
Data-Free Knowledge Distillation (DFKD) transfers knowledge from a pretrained teacher model to a compact student model by synthesizing semantically informative data, eliminating th…
What to Remove, What to Preserve: Dual-Ambiguity Rectification for All-in-One Image Restoration
Cencen Liu, Wen Yin, Dongyang Zhang +6
All-in-one image restoration aims to handle diverse degradations within a unified framework. Existing methods commonly encode heterogeneous degradation conditions in a shared laten…
DIVER: Diving Deeper into Distilled Data via Expressive Semantic Recovery
Qianxin Xia, Zhiyong Shu, Wenbo Jiang +3
Dataset distillation aims to synthesize a compact proxy dataset that is unreadable or non-raw from the original dataset for privacy protection and highly efficient learning. Howeve…
AnchorDiff: Topology-Aware Masked Diffusion with Confidence-based Rewriting for Radiology Report Generation
Shiying Yu, Jielei Wang, Guoming Lu
Radiology report generation (RRG) aims to automatically produce clinically accurate textual reports from medical images. Existing methods predominantly rely on autoregressive (AR)…
FSCE: A Target-Aware Frequency-Spatial Collaborative Enhancement Framework for Noise-Resilient SAR ATR
Yansong Lin, Zihan Cheng, Ziyue Yang +5
Synthetic aperture radar automatic target recognition (SAR ATR) is severely challenged by coherent speckle noise, whose interference can be progressively amplified by hierarchical…
Efficient Dataset Distillation for Pre-Trained Self-Supervised Models via Statistical Flow Matching
Qianxin Xia, Jiawei Du, Xin Zhang +6
Dataset distillation seeks to synthesize a compact surrogate dataset that enables performance comparable to training on the original dataset for downstream tasks. For the scenario…
EDITS: Enhancing Dataset Distillation with Implicit Textual Semantics
Qianxin Xia, Jiawei Du, Guoming Lu +2
Dataset distillation aims to synthesize a compact dataset from the original large-scale one, enabling highly efficient learning while preserving competitive model performance. Howe…