Publications (54)
Dynamic Programming in Rank Space: Scaling Structured Inference with Low-Rank HMMs and PCFGs
Songlin Yang, Wei Liu, Kewei Tu
Hidden Markov Models (HMMs) and Probabilistic Context-Free Grammars (PCFGs) are widely used structured models, both of which can be represented as factor graph grammars (FGGs), a p…
Designing a 3D-Aware StyleNeRF Encoder for Face Editing
Songlin Yang, Wei Wang, Bo Peng +1
GAN inversion has been exploited in many face manipulation tasks, but 2D GANs often fail to generate multi-view 3D consistent images. The encoders designed for 2D GANs are not able…
Generation of S-shaped photonic hooks from microcylinders with engineered surface patches
Chu Xu, Fen Tang, Qingqing Shang +8
Photonic hooks (PHs) are non-evanescent light beams with a highly concentrated curved optical fields. Since their discovery, PHs always have one single inflection point and thus ha…
Distilling to Hybrid Attention Models via KL-Guided Layer Selection
Yanhong Li, Songlin Yang, Shawn Tan +4
Distilling pretrained softmax attention Transformers into more efficient hybrid architectures that interleave softmax and linear attention layers is a promising approach for improv…
Joint Entity and Relation Extraction with Span Pruning and Hypergraph Neural Networks
Zhaohui Yan, Songlin Yang, Wei Liu +1
Entity and Relation Extraction (ERE) is an important task in information extraction. Recent marker-based pipeline models achieve state-of-the-art performance, but still suffer from…
ARFlow: Autoregressive Flow with Hybrid Linear Attention
Mude Hui, Rui-Jie Zhu, Songlin Yang +5
Flow models are effective at progressively generating realistic images, but they generally struggle to capture long-range dependencies during the generation process as they compres…