Results 1 to 10 of about 1,352,218 (128)
Plan-and-Solve Prompting: Improving Zero-Shot Chain-of-Thought Reasoning by Large Language Models [PDF]
Annual Meeting of the Association for Computational Linguistics, 2023 Large language models (LLMs) have recently been shown to deliver impressive performance in various NLP tasks. To tackle multi-step reasoning tasks, Few-shot chain-of-thought (CoT) prompting includes a few manually crafted step-by-step reasoning ...
Lei Wang +6 more
semanticscholar +1 more source
Better Zero-Shot Reasoning with Role-Play Prompting [PDF]
North American Chapter of the Association for Computational Linguistics, 2023 Modern large language models (LLMs) exhibit a remarkable capacity for role-playing, enabling them to embody not only human characters but also non-human entities.
Aobo Kong +6 more
semanticscholar +1 more source
Precise Zero-Shot Dense Retrieval without Relevance Labels [PDF]
Annual Meeting of the Association for Computational Linguistics, 2022 While dense retrieval has been shown to be effective and efficient across tasks and languages, it remains difficult to create effective fully zero-shot dense retrieval systems when no relevance labels are available.
Luyu Gao +3 more
semanticscholar +1 more source
Aligning Instruction Tasks Unlocks Large Language Models as Zero-Shot Relation Extractors [PDF]
Annual Meeting of the Association for Computational Linguistics, 2023 Recent work has shown that fine-tuning large language models (LLMs) on large-scale instruction-following datasets substantially improves their performance on a wide range of NLP tasks, especially in the zero-shot setting.
Kai Zhang +2 more
semanticscholar +1 more source
Better Zero-Shot Reasoning with Self-Adaptive Prompting [PDF]
Annual Meeting of the Association for Computational Linguistics, 2023 Modern large language models (LLMs) have demonstrated impressive capabilities at sophisticated tasks, often through step-by-step reasoning similar to humans.
Xingchen Wan +4 more
semanticscholar +1 more source
On Second Thought, Let’s Not Think Step by Step! Bias and Toxicity in Zero-Shot Reasoning [PDF]
Annual Meeting of the Association for Computational Linguistics, 2022 Generating a Chain of Thought (CoT) has been shown to consistently improve large language model (LLM) performance on a wide range of NLP tasks. However, prior work has mainly focused on logical reasoning tasks (e.g.
Omar Shaikh +4 more
semanticscholar +1 more source
ChatGPT for Zero-shot Dialogue State Tracking: A Solution or an Opportunity? [PDF]
Annual Meeting of the Association for Computational Linguistics, 2023 Recent research on dialog state tracking (DST) focuses on methods that allow few- and zero-shot transfer to new domains or schemas. However, performance gains heavily depend on aggressive data augmentation and fine-tuning of ever larger language model ...
Michael Heck +8 more
semanticscholar +1 more source
Tab-CoT: Zero-shot Tabular Chain of Thought [PDF]
Annual Meeting of the Association for Computational Linguistics, 2023 The chain-of-though (CoT) prompting methods were successful in various natural language processing (NLP) tasks thanks to their ability to unveil the underlying complex reasoning processes.
Ziqi Jin, Wei Lu
semanticscholar +1 more source
ReGen: Zero-Shot Text Classification via Training Data Generation with Progressive Dense Retrieval [PDF]
Annual Meeting of the Association for Computational Linguistics, 2023 With the development of large language models (LLMs), zero-shot learning has attracted much attention for various NLP tasks. Different from prior works that generate training data with billion-scale natural language generation (NLG) models, we propose a ...
Yue Yu +5 more
semanticscholar +1 more source
Zero-shot Faithful Factual Error Correction [PDF]
Annual Meeting of the Association for Computational Linguistics, 2023 Faithfully correcting factual errors is critical for maintaining the integrity of textual knowledge bases and preventing hallucinations in sequence-to-sequence models.
Kung-Hsiang Huang +2 more
semanticscholar +1 more source