Know Your Limits: A Survey of Abstention in Large Language Models
It addresses safety and reliability issues for LLM users, but is incremental as it synthesizes existing work without new empirical results.
This survey tackles the problem of abstention in large language models (LLMs) to mitigate hallucinations and enhance safety, organizing literature on methods, benchmarks, and evaluation metrics to identify future research areas.
Abstention, the refusal of large language models (LLMs) to provide an answer, is increasingly recognized for its potential to mitigate hallucinations and enhance safety in LLM systems. In this survey, we introduce a framework to examine abstention from three perspectives: the query, the model, and human values. We organize the literature on abstention methods, benchmarks, and evaluation metrics using this framework, and discuss merits and limitations of prior work. We further identify and motivate areas for future research, such as whether abstention can be achieved as a meta-capability that transcends specific tasks or domains, and opportunities to optimize abstention abilities in specific contexts. In doing so, we aim to broaden the scope and impact of abstention methodologies in AI systems.