6 papers
MVCL-DAF++: Enhancing Multimodal Intent Recognition via Prototype-Aware Contrastive Alignment and Coarse-to-Fine Dynamic Attention Fusion
Haofeng Huang, Yifei Han, Long Zhang +3
Multimodal intent recognition (MMIR) suffers from weak semantic grounding and poor robustness under noisy or rare-class conditions. We propose MVCL-DAF++, which extends MVCL-DAF wi…
SiameseNorm: Breaking the Barrier to Reconciling Pre/Post-Norm
Tianyu Li, Dongchen Han, Zixuan Cao +7
The long-standing tension between Pre- and Post-Norm remains an open problem in Transformer architecture, reflecting a fundamental trade-off between training stability and represen…
Q-BERT4Rec: Quantized Semantic-ID Representation Learning for Multimodal Recommendation
Haofeng Huang, Ling Gai
Sequential recommendation plays a critical role in modern online platforms such as e-commerce, advertising, and content streaming, where accurately predicting users' next interacti…
A Unified View of Attention and Residual Sinks: Outlier-Driven Rescaling is Essential for Transformer Training
Zihan Qiu, Zeyu Huang, Kaiyue Wen +16
We investigate the functional role of emergent outliers in large language models, specifically attention sinks (a few tokens that consistently receive large attention logits) and r…
VSA: Faster Video Diffusion with Trainable Sparse Attention
Peiyuan Zhang, Yongqi Chen, Haofeng Huang +5
Scaling video diffusion transformers (DiTs) is limited by their quadratic 3D attention, even though most of the attention mass concentrates on a small subset of positions. We turn…
XAttention: Block Sparse Attention with Antidiagonal Scoring
Ruyi Xu, Guangxuan Xiao, Haofeng Huang +2
Long-Context Transformer Models (LCTMs) are vital for real-world applications but suffer high computational costs due to attention's quadratic complexity. Block-sparse attention mi…