SAIS: Supervising and Augmenting Intermediate Steps for Document-Level Relation Extraction
Comments Accepted by NAACL 2022
作者
Jiawei Han
Data Mining
OA-Mine: Open-World Attribute Mining for E-Commerce Products with Weak Supervision
Comments WWW 2022
The Future is not One-dimensional: Complex Event Schema Induction by Graph Modeling for Event Prediction
Pretraining Text Encoders with Adversarial Mixture of Training Signal Generators
Comments ICLR 2022. (Code and Models: https://github.com/microsoft/AMOS)
Shift-Robust Node Classification via Graph Adversarial Clustering
Corpus-based Open-Domain Event Type Induction
Comments 15 pages, EMNLP 2021 main conference, updated for new related work
Eider: Empowering Document-level Relation Extraction with Efficient Evidence Extraction and Inference-stage Fusion
Comments ACL 2022 Findings
Training ELECTRA Augmented with Multi-word Selection
Comments Accepted in Findings of ACL 2021
TaxoEnrich: Self-Supervised Taxonomy Completion via Structure-Semantic Representations
Comments WWW 2022
Topic Discovery via Latent Space Clustering of Pretrained Language Model Representations
Comments WWW 2022. (Code: https://github.com/yumeng5/TopClus)
TaxoCom: Topic Taxonomy Completion with Hierarchical Discovery of Novel Topic Clusters
Comments The Web Conference (WWW) 2022, 11 pages, 7 figures
Constrained Abstractive Summarization: Preserving Factual Consistency with Constrained Generation
Out-of-Category Document Identification Using Target-Category Names as Weak Supervision
Comments ICDM 2021. 10 pages, 4 figures
Shift-Robust GNNs: Overcoming the Limitations of Localized Graph Training Data
Comments NeurIPS 2021
COCO-LM: Correcting and Contrasting Text Sequences for Language Model Pretraining
Comments NeurIPS 2021. (Code and Models: https://github.com/microsoft/COCO-LM)
Transfer Learning of Graph Neural Networks with Ego-graph Information Maximization
Comments NeurIPS 2021
On the Variance of the Adaptive Learning Rate and Beyond
Comments ICLR 2020. Fix several typos in the previous version
Fine-Grained Opinion Summarization with Minimal Supervision
Entity Linking Meets Deep Learning: Techniques and Solutions
Comments To appear in IEEE TKDE
Chemical-Reaction-Aware Molecule Representation Learning
Distantly-Supervised Named Entity Recognition with Noise-Robust Learning and Language Model Augmented Self-Training
Comments EMNLP 2021. (Code: https://github.com/yumeng5/RoSTER)
Extract, Denoise and Enforce: Evaluating and Improving Concept Preservation for Text-to-Text Generation
Comments EMNLP 2021
Rider: Reader-Guided Passage Reranking for Open-Domain Question Answering
Comments Minor format updates. TLDR: Reranking retrieved passages by reader predictions can achieve 10~20 gains in top-1 retrieval accuracy and 1~4 gains in Exact Match (EM) without any training
Generation-Augmented Retrieval for Open-domain Question Answering
Comments Minor format updates
On the Transformer Growth for Progressive BERT Training
Comments NAACL 2021
Multi-head or Single-head? An Empirical Comparison for Transformer Training
Comments Work in progress
Scaling Deep Contrastive Learning Batch Size under Memory Limited Setup
Comments RepL4NLP 2021
UCPhrase: Unsupervised Context-aware Quality Phrase Tagging
Comments KDD 2021
