Recent News

  • 2023-10-07: Two papers (Document OOD Detection and Document CLIP) get accepted by EMNLP 2023.
  • 2023-09-21: One paper on entity segmentation gets accepted to NeurIPS 2023 (Spotlight).
  • 2023-07-14: One paper gets accepted to ICCV 2023 (Oral).
  • 2023-05-25: I will be serving as an Area Chair of WACV 2024.
  • 2023-01-04: One paper gets accepted to AAAI 2023.
  • 2022-12-05: One paper (Entity Segmentation) gets accepted to PAMI.
  • 2022-11-19: One paper gets accepted to AAAI 2023.
  • 2022-10-24: One paper (Differential Privacy in Learning Language Models) gets accepted to BigData 2022.
  • 2022-10-10: One paper gets accepted to WACV 2023.
  • 2022-10-06: Our most recent Doc-series paper (SelfDoc->UniDoc->MGDoc) gets accepted to EMNLP 2022.
  • 2022-09-14: One paper (OOD Detection) gets accepted to NeurIPS 2022.
  • 2022-07-03: Three papers get accepted to ECCV 2022.
  • 2022-06-14: One paper gets accepted to Interspeech 2022.
  • 2022-04-07: One paper gets accepted to NAACL 2022.
  • 2022-03-01: Three papers get accepted to CVPR 2022.
  • 2022-02-24: One paper (Adaptive Axis Attentions) gets accepted to ACL 2022 Findings.
  • 2022-01-13: One paper gets accepted to WWW 2022.
  • More >>

Publications

Quickly discover relevant content by filtering publications.

We introduce a new image segmentation task, called Entity Segmentation (ES), which aims to segment all visual entities (objects and …

Professional document editing tools require a certain level of expertise to perform complex edit operations. To make editing tools …

In this paper, we introduce a novel concept of user-entity differential privacy (UeDP) to provide formal privacy protection …

Document images are a ubiquitous source of data where the text is organized in a complex hierarchical structure ranging from fine …

Recognizing out-of-distribution (OOD) samples is critical for machine learning systems deployed in the open world. The vast majority of …

We propose a new task of synthesizing speech directly from semi-structured documents where the extracted text tokens from OCR systems …

We introduce DocTime - a novel temporal dependency graph (TDG) parser that takes as input a text document and produces a temporal …

This work presents one of the first comprehensive studies on different sparse attention patterns in Transformer models. We first …

Multilingual natural language understanding, which aims to comprehend multilingual documents, is an important task. Existing efforts …