NeurIPS 2022 accepted papers on Large language models
76 of the 2,901 papers accepted at NeurIPS 2022 (2.6%) are on Large language models, compared with 2.6% at NeurIPS 2021. They fall into 6 topics.
See NeurIPS 2022 in the interactive map
Submitting to NeurIPS? 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 · 39 papers
- Large Language Models are Zero-Shot Reasoners (1,089 citations)
- Deep Bidirectional Language-Knowledge Graph Pretraining (86 citations)
- A Contrastive Framework for Neural Text Generation (83 citations)
- Generating Training Data with Language Models: Towards Zero-Shot Language Understanding (79 citations)
- The BigScience ROOTS Corpus: A 1.6TB Composite Multilingual Dataset (61 citations)
- Two-Stream Network for Sign Language Recognition and Translation (57 citations)
- Factuality Enhanced Language Models for Open-Ended Text Generation (55 citations)
- The Unreliability of Explanations in Few-shot Prompting for Textual Reasoning (52 citations)
- Memorization Without Overfitting: Analyzing the Training Dynamics of Large Language Models (44 citations)
- COLD Decoding: Energy-based Constrained Text Generation with Langevin Dynamics (43 citations)
- Pile of Law: Learning Responsible Data Filtering from the Law and a 256GB Open-Source Legal Dataset (43 citations)
- Exploring Length Generalization in Large Language Models (41 citations)
and 27 more in the interactive map.
Agents, code and math · 15 papers
- Solving Quantitative Reasoning Problems with Language Models (278 citations)
- Autoformalization with Large Language Models (42 citations)
- pyKT: A Python Library to Benchmark Deep Learning based Knowledge Tracing Models (30 citations)
- Thor: Wielding Hammers to Integrate Language Models and Automated Theorem Provers (20 citations)
- Fault-Aware Neural Code Rankers (13 citations)
- ELASTIC: Numerical Reasoning with Adaptive Symbolic Compiler (9 citations)
- Evaluation beyond Task Performance: Analyzing Concepts in AlphaZero in Hex (7 citations)
- NaturalProver: Grounded Mathematical Proof Generation with Language Models (7 citations)
- Using natural language and program abstractions to instill human inductive biases in machines (7 citations)
- Forecasting Future World Events With Neural Networks (6 citations)
- Improving Certified Robustness via Statistical Learning with Logical Reasoning (6 citations)
- Communicating Natural Programs to Humans and Machines (4 citations)
and 3 more in the interactive map.
AI and society · 9 papers
- Picking on the Same Person: Does Algorithmic Monoculture lead to Outcome Homogenization? (28 citations)
- When to Make Exceptions: Exploring Language Models as Accounts of Human Moral Judgment (26 citations)
- Uncalibrated Models Can Improve Human-AI Collaboration (15 citations)
- METS-CoV: A Dataset of Medical Entity and Targeted Sentiment on COVID-19 Related Tweets (10 citations)
- OpenFilter: A Framework to Democratize Research Access to Social Media AR Filters (4 citations)
- Human-AI Collaborative Bayesian Optimisation
- Towards Trustworthy Automatic Diagnosis Systems by Emulating Doctors' Reasoning with Deep Reinforcement Learning
- What I Cannot Predict, I Do Not Understand: A Human-Centered Evaluation Framework for Explainability Methods
- 🏘️ ProcTHOR: Large-Scale Embodied AI Using Procedural Generation
Retrieval and knowledge · 8 papers
- Autoregressive Search Engines: Generating Substrings as Document Identifiers (65 citations)
- Decoupling Knowledge from Memorization: Retrieval-augmented Prompt Learning (25 citations)
- EHRSQL: A Practical Text-to-SQL Benchmark for Electronic Health Records (13 citations)
- Decoupled Context Processing for Context Augmented Language Modeling (5 citations)
- Neural-Symbolic Entangled Framework for Complex Query Answering (5 citations)
- TweetNERD - End to End Entity Linking Benchmark for Tweets (4 citations)
- Eliciting Thinking Hierarchy without a Prior (3 citations)
- Retrieve, Reason, and Refine: Generating Accurate and Faithful Patient Instructions (2 citations)
Alignment and preferences · 3 papers
Reasoning and chain of thought · 2 papers
- Chain-of-Thought Prompting Elicits Reasoning in Large Language Models (2,251 citations)
- STaR: Bootstrapping Reasoning With Reasoning (114 citations)
