Large Language Model

Direct Preference Optimization (DPO) is a sophisticated training technique used for refining large language models (LLMs). It does not depend on a single gold reference like traditional supervised fine-tuning, instead, it trains models to identify quality differences among multiple outputs. Adding reinforcement learning approaches, DPO can learn from feedback, making it a useful technique for…

Researchers from Baidu Inc., China, have unveiled a self-reasoning framework that greatly improves the reliability and traceability of Retrieval-Augmented Language Models (RALMs). RALMs augment language models with external knowledge, decreasing factual inaccuracies. However, they face reliability and traceability issues, as noisy retrieval may lead to incorrect responses, and a lack of citations makes verifying these…

A recent research paper by the University of Washington and Allen Institute for AI researchers has examined the use of abstention in large language models (LLMs), emphasizing its potential to minimize false results and enhance the safety of AI. The study investigates the current methods of abstention incorporated during the different development stages of LLMs…

Large Language Models (LLMs), which focus on understanding and generating human language, are a subset of artificial intelligence. However, their use of the Transformer architecture to process long texts introduces a significant challenge due to its quadratic time complexity. This complexity is a barrier to efficient performance with extended text inputs. To deal with this issue,…

Large language models (LLMs), used in applications such as machine translation, content creation, and summarization, present significant challenges due to their tendency to generate hallucinations - plausible sounding but factually inaccurate statements. This major issue affects the reliability of AI-produced copy, particularly in high-accuracy-required domains like medical and legal texts. Thus, reducing hallucinations in LLMs…

A team of researchers from Meta FAIR have been studying Large Language Models (LLMs) and found that these can produce more nuanced responses by distilling System 2 reasoning methods into System 1 responses. While System 1 operates quickly and directly, generating responses without intermediate steps, System 2 uses intermediate strategies, such as token generation and…