ICLR 2020 accepted papers on Large language models
23 of the 687 papers accepted at ICLR 2020 (3.3%) are on Large language models, compared with 2.2% at ICLR 2019. They fall into 3 topics.
See ICLR 2020 in the interactive map
Submitting to ICLR? Checklist →
Working on something in this area? Describe your idea in scime atlas to see the closest papers by meaning and how crowded the spot has become.
Topics and papers
Language, safety and interpretability · 15 papers
- The Curious Case of Neural Text Degeneration (1,122 citations)
- Neural Text Generation With Unlikelihood Training (235 citations)
- TabFact: A Large-scale Dataset for Table-based Fact Verification (235 citations)
- Cross-Lingual Ability of Multilingual BERT: An Empirical Study (167 citations)
- Plug and Play Language Models: A Simple Approach to Controlled Text Generation (138 citations)
- ReClor: A Reading Comprehension Dataset Requiring Logical Reasoning (133 citations)
- StructBERT: Incorporating Language Structures into Pre-training for Deep Language Understanding (99 citations)
- Mixout: Effective Regularization to Finetune Large-scale Pretrained Language Models (97 citations)
- Thieves on Sesame Street! Model Extraction of BERT-based APIs (96 citations)
- Permutation Equivariant Models for Compositional Generalization in Language (65 citations)
- Measuring Compositional Generalization: A Comprehensive Method on Realistic Data (51 citations)
- Are Pre-trained Language Models Aware of Phrases? Simple but Strong Baselines for Grammar Induction (35 citations)
and 3 more in the interactive map.
Retrieval and knowledge · 5 papers
- Learning to Retrieve Reasoning Paths over Wikipedia Graph for Question Answering (145 citations)
- Low-Resource Knowledge-Grounded Dialogue Generation (83 citations)
- Query2box: Reasoning over Knowledge Graphs in Vector Space Using Box Embeddings (74 citations)
- Differentiable Reasoning over a Virtual Knowledge Base (41 citations)
- Learning from Explanations with Neural Execution Tree (17 citations)
