1

Rano: Restorable Speaker Anonymization via Conditional Invertible Neural Network
TBD
Cite
Rano: Restorable Speaker Anonymization via Conditional Invertible Neural Network
Enhancing Multi-Agent Systems via Reinforcement Learning with LLM-based Planner and Graph-based Policy
Multi-agent systems (MAS) have shown great potential in executing complex tasks, but coordination and safety remain significant …
Cite arXiv
Enhancing Multi-Agent Systems via Reinforcement Learning with LLM-based Planner and Graph-based Policy
CycleFlow: Leveraging Cycle Consistency in Flow Matching for Speaker Style Adaptation
Voice Conversion (VC) aims to convert the style of a source speaker, such as timbre and pitch, to the style of any target speaker while …
Cite arXiv IEEE
CycleFlow: Leveraging Cycle Consistency in Flow Matching for Speaker Style Adaptation
Graph Contrastive Learning with Decoupled Augmentation
Graph contrastive learning based on augmentation strategies has recently demonstrated remarkable performance. Existing methods …
PDF Cite IEEE
Graph Contrastive Learning with Decoupled Augmentation
Homogeneous Graph Extraction: An Approach to Learning Heterogeneous Graph Embedding
Heterogeneous Graph Neural Networks (HGNNs) aim to embed rich structural and semantic information of heterogeneous graphs into …
PDF Cite IEEE
Homogeneous Graph Extraction: An Approach to Learning Heterogeneous Graph Embedding
PointActionCLIP: Preventing Transfer Degradation in Point Cloud Action Recognition with a Triple-Path CLIP
Directly applying CLIP to point cloud action recognition can cause severe accuracy collapse. In this paper, we propose PointActionCLIP, …
PDF Cite IEEE
PointActionCLIP: Preventing Transfer Degradation in Point Cloud Action Recognition with a Triple-Path CLIP
VisTa: Visual-contextual and Text-augmented Zero-shot Object-level OOD Detection
As object detectors are increasingly deployed as black-box cloud services or pre-trained models with restricted access to the original …
PDF Cite arXiv IEEE
VisTa: Visual-contextual and Text-augmented Zero-shot Object-level OOD Detection
ACCon: Angle-Compensated Contrastive Regularizer for Deep Regression
In deep regression, capturing the relationship among continuous labels in feature space is a fundamental challenge that has attracted …
Cite arXiv
ACCon: Angle-Compensated Contrastive Regularizer for Deep Regression
RUNA: Object-level Out-of-Distribution Detection via Regional Uncertainty Alignment of Multimodal Representations
Enabling object detectors to recognize out-of-distribution (OOD) objects is vital for building reliable systems. A primary obstacle …
Cite arXiv
RUNA: Object-level Out-of-Distribution Detection via Regional Uncertainty Alignment of Multimodal Representations
Cocktail:Chunk-AdaptiveMixed-Precision Ouanization for Long-Context LLM inference
Recently, large language models (LLMs) have been able to handle longer and longer contexts. However, a context that is too long may …
Cite arXiv
Cocktail:Chunk-AdaptiveMixed-Precision Ouanization for Long-Context LLM inference