X
Xiang Ren
Researcher at Facebook
Publications - 5
Citations - 1
Xiang Ren is an academic researcher from Facebook. The author has contributed to research in topics: Modality (human–computer interaction) & Computer science. The author has co-authored 5 publications. Previous affiliations of Xiang Ren include Amazon.com & Indian Institute of Technology Kanpur.
Papers
More filters
Posted Content
MSD: Saliency-aware Knowledge Distillation for Multimodal Understanding.
TL;DR: In this paper, a modality-specific knowledge distillation (MSD) framework is proposed to transfer knowledge from a teacher on multimodal tasks by learning the teacher's behavior within each modality.
Posted Content
Sparse Distillation: Speeding Up Text Classification by Using Bigger Models.
TL;DR: The authors distill a transformer-based text classifier into a billion-parameter, sparsely-activated student model with a embedding-averaging architecture, which achieves up to 600x speedup on both GPUs and CPUs, compared to the teacher models.
Posted Content
On the Robustness of Reading Comprehension Models to Entity Renaming.
TL;DR: The authors proposed a general and scalable method to replace person names with names from a variety of sources, ranging from common English names to names from other languages to arbitrary strings, and found that this can further improve the robustness of MRC models.
Posted Content
Think Before You Speak: Using Self-talk to Generate Implicit Commonsense Knowledge for Response Generation
Pei Zhou,Karthik Gopalakrishnan,Behnam Hedayatnia,Seokhwan Kim,Jay Pujara,Xiang Ren,Yang Liu,Dilek Hakkani-Tur +7 more
TL;DR: This article presented a self-talk approach that first generates the implicit commonsense knowledge and then generates response by referencing the externalized knowledge, all using one generative model, and evaluated three evaluation aspects: knowledge quality, knowledge-response connection, and response quality.
Proceedings Article
MSD: Saliency-aware Knowledge Distillation for Multimodal Understanding
TL;DR: In this article, a modality-specific knowledge distillation (MSD) framework is proposed to transfer knowledge from a teacher on multimodal tasks by learning the teacher's behavior within each modality.