Recitation-Augmented Language Models
Explore this paper's citation graph
Summary
It is shown that by utilizing recitation as the intermediate step, a recite-and-answer scheme can achieve new state-of-the-art performance in various closed-book question answering (CBQA) tasks.
- Type
- preprint
- Published
- 2022-10-04
- Cited by
- 89
- References
- 73
- Access
- Open access
- OpenAlex
- https://openalex.org/W4302305884
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:252692968
Keywords
Computer science, Code (set theory), Scheme (mathematics), Artificial intelligence, Natural language processing
References
- The Read-Recite-Review Study Strategy
- Reading Wikipedia to Answer Open-Domain Questions
- TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehension
- SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text Processing
- HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering
- End-to-End Open-Domain Question Answering with BERTserini
- Natural Questions: A Benchmark for Question Answering Research
- MathQA: Towards Interpretable Math Word Problem Solving with Operation-Based Formalisms
- Program Induction by Rationale Generation: Learning to Solve and Explain Algebraic Word Problems
- BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
- Equation Parsing : Mapping Sentences to Grounded Equations
- The Secret Sharer: Evaluating and Testing Unintended Memorization in Neural Networks
- RoBERTa: A Robustly Optimized BERT Pretraining Approach
- Language Models as Knowledge Bases?
- Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
- Contextualized Sparse Representation with Rectified N-Gram Attention for Open-Domain Question Answering
- How Can We Know What Language Models Know?
- Neural Symbolic Reader: Scalable Integration of Distributed and Symbolic Representations for Reading Comprehension
- oLMpics-On What Language Model Pre-training Captures
- REALM: Retrieval-Augmented Language Model Pre-Training
Cited by
- Interleaving Retrieval with Chain-of-Thought Reasoning for Knowledge-Intensive Multi-Step Questions
- When Not to Trust Language Models: Investigating Effectiveness of Parametric and Non-Parametric Memories
- Demonstrate-Search-Predict: Composing retrieval and language models for knowledge-intensive NLP
- Query2doc: Query Expansion with Large Language Models
- Language Models can Solve Computer Tasks
- Answering Questions by Meta-Reasoning over Multiple Chains of Thought
- Search-in-the-Chain: Towards the Accurate, Credible and Traceable Content Generation for Complex Knowledge-intensive Tasks
- Principle-Driven Self-Alignment of Language Models from Scratch with Minimal Human Supervision
- Active Retrieval Augmented Generation
- Towards Expert-Level Medical Question Answering with Large Language Models
- Can Language Models Solve Graph Problems in Natural Language?
- Adaptive Chameleon or Stubborn Sloth: Unraveling the Behavior of Large Language Models in Knowledge Clashes
- Allies: Prompting Large Language Model with Beam Search
- Improving Language Models via Plug-and-Play Retrieval Feedback
- Leveraging Few-Shot Data Augmentation and Waterfall Prompting for Response Generation
- Enhancing Large Language Models in Coding Through Multi-Perspective Self-Consistency
- Knowledge Crosswords: Geometric Reasoning over Structured Knowledge with Large Language Models
- Large Language Models as Analogical Reasoners
- FreshLLMs: Refreshing Large Language Models with Search Engine Augmentation
- Take a Step Back: Evoking Reasoning via Abstraction in Large Language Models
Related papers
- Evaluating an earlybird scheme: encouraging early assignment writing and revising
- A Novel Scheme Producing and Evaluation System in HWME
- Study on Multi-Feedback Model Based Cultivating Scheme Planning of Industrial Design
- Improved topic-dependent language modeling using information retrieval techniques
- The state and private education : an evaluation of the Assisted Places Scheme
- The scheme of things: portability
- Dealing with adaptation: the Courtauld classification scheme
- BanglaLM: Data Mining based Bangla Corpus for Language Model Research