Profiles2023-2026
Wei Xue
Hong Kong University of Science and Technology
15Papers
4Last author
21Unique co-authors
0First author
0Solo
First and last author come from byline position, which some venues assign alphabetically.
Recent papers15
- ActorMind: Emulating Human Actor Reasoning for Speech Role-PlayingAuthor 2 of 3ACL Findings 2026
- Causal Dependency-Aware Unsupervised Routing for Large Reasoning ModelsAuthor 4 of 5ICML 2026
- STAR-VAE: Structured Topology-Aware Regularization for Audio Reconstruction and GenerationAuthor 6 of 6ICML 2026
- VMChill: A Dataset for Fine-Grained Visual-Musical SynergyAuthor 4 of 4AAAI 2026
- Importance Weighting Can Help Large Language Models Self-ImproveAuthor 3 of 5AAAI 2025 · 12 citations
- BayesKD: Bayesian Knowledge Distillation for Compact LLMs in Constrained Fine-tuning ScenariosAuthor 6 of 7ACL Findings 2025 · 3 citations
- Efficient Fine-Tuning of Large Models Via Nested Low-Rank AdaptationAuthor 8 of 10ICCV 2025 · 1 citation
- STBLLM: Breaking the 1-Bit Barrier with Structured Binary LLMsAuthor 9 of 11ICLR 2025 · 1 citation
- AIRA: Activation-Informed Low-Rank Adaptation for Large ModelsAuthor 5 of 7ICCV 2025
- OmniAudio: Generating Spatial Audio from 360-Degree VideoAuthor 14 of 14ICML 2025
- Discovering Sparsity Allocation for Layer-wise Pruning of Large Language ModelsAuthor 7 of 10NeurIPS 2024 · 51 citations
- DetKDS: Knowledge Distillation Search for Object DetectorsAuthor 8 of 9ICML 2024 · 35 citations
- Weakly-Supervised Emotion Transition Learning for Diverse 3D Co-Speech Gesture GenerationAuthor 8 of 11CVPR 2024 · 9 citations
- PyramidCodec: Hierarchical Codec for Long-form Music Generation in Audio DomainAuthor 7 of 7EMNLP Findings 2024
- CoMoSpeech: One-Step Speech and Singing Voice Synthesis via Consistency ModelAuthor 2 of 6ACM MM 2023 · 30 citations
