AAAI 2020 accepted papers on Large language models
107 of the 1,610 papers accepted at AAAI 2020 (6.6%) are on Large language models, compared with 5.8% at AAAI 2019. They fall into 7 topics.
See AAAI 2020 in the interactive map
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 · 46 papers
- ERNIE 2.0: A Continual Pre-Training Framework for Language Understanding (811 citations)
- Towards Scalable Multi-Domain Conversational Agents: The Schema-Guided Dialogue Dataset (481 citations)
- Semantics-Aware BERT for Language Understanding (392 citations)
- Do Not Have Enough Data? Deep Learning to the Rescue! (371 citations)
- Effective Modeling of Encoder-Decoder Architecture for Joint Entity and Relation Extraction (252 citations)
- Automatically Neutralizing Subjective Bias in Text (173 citations)
- Capturing the Style of Fake News (164 citations)
- A Pre-Training Based Personalized Dialogue Generation Model with Persona-Sparse Data (158 citations)
- Inducing Relational Knowledge from BERT (140 citations)
- Probing Natural Language Inference Models through Semantic Fragments (118 citations)
- A Large-Scale Dataset for Argument Quality Ranking: Construction and Analysis (93 citations)
- Rare Words: A Major Problem for Contextualized Embeddings and How to Fix it by Attentive Mimicking (88 citations)
and 34 more in the interactive map.
Retrieval and knowledge · 40 papers
- WinoGrande: An Adversarial Winograd Schema Challenge at Scale (1,090 citations)
- PIQA: Reasoning about Physical Commonsense in Natural Language (959 citations)
- K-BERT: Enabling Language Representation with Knowledge Graph (899 citations)
- QASC: A Dataset for Question Answering via Sentence Composition (213 citations)
- JEC-QA: A Legal-Domain Question Answering Dataset (168 citations)
- Evaluating Commonsense in Pre-Trained Language Models (156 citations)
- Select, Answer and Explain: Interpretable Multi-Hop Reading Comprehension over Multiple Documents (144 citations)
- DCMN+: Dual Co-Matching Network for Multi-Choice Reading Comprehension (136 citations)
- SPARQA: Skeleton-Based Semantic Parsing for Complex Questions over Knowledge Bases (121 citations)
- Iteratively Questioning and Answering for Interpretable Legal Judgment Prediction (114 citations)
- A Knowledge-Aware Attentional Reasoning Network for Recommendation (79 citations)
- Differentiable Reasoning on Large Knowledge Bases and Natural Language (77 citations)
and 28 more in the interactive map.
AI and society · 10 papers
- Atari-HEAD: Atari Human Eye-Tracking and Demonstration Dataset (68 citations)
- Predictive Engagement: An Efficient Metric for Automatic Evaluation of Open-Domain Dialogue Systems (47 citations)
- Crowd-Assisted Disaster Scene Assessment with Human-AI Interactive Attention (36 citations)
- TrueLearn: A Family of Bayesian Algorithms to Match Lifelong Learners to Open Educational Resources (36 citations)
- Towards Socially Responsible AI: Cognitive Bias-Aware Multi-Objective Learning (22 citations)
- A Unified Framework for Knowledge Intensive Gradient Boosting: Leveraging Human Experts for Noisy Sparse Domains (18 citations)
- Asymptotically Unambitious Artificial General Intelligence (18 citations)
- Dempster-Shafer Theoretic Learning of Indirect Speech Act Comprehension Norms (12 citations)
- Actionable Ethics through Neural Learning (8 citations)
- Tracking Disaster Footprints with Social Streaming Data (4 citations)
Agents, code and math · 4 papers
- MALA: Cross-Domain Dialogue Generation with Action Learning (15 citations)
- Grammar Filtering for Syntax-Guided Synthesis (12 citations)
- Narrative Planning Model Acquisition from Text Summaries and Descriptions (9 citations)
- ElGolog: A High-Level Programming Language with Memory of the Execution History (2 citations)
Alignment and preferences · 3 papers
Instruction and fine-tuning · 3 papers
- Fine-Tuning by Curriculum Learning for Non-Autoregressive Neural Machine Translation (75 citations)
- Learning to Compare for Better Training and Evaluation of Open Domain Natural Language Generation Models (39 citations)
- Learning to Learn Morphological Inflection for Resource-Poor Languages (8 citations)
