It is often desirable for Large Language Models (LLMs) to capture multiple objectives when providing a response. In document-grounded response generation, for example, agent responses are expected to be relevant to a user's query while also being grounded in a given document. In this paper, we introduce Proxy Metric-based Self-Refinement (ProMiSe), which enables an LLM to refine its own initial response along key dimensions of quality guided by external metrics feedback, yielding an overall better final response. ProMiSe leverages feedback on response quality through principle-specific proxy metrics, and iteratively refines its response one principle at a time. We apply ProMiSe to open source language models Flan-T5-XXL and Llama-2-13B-Chat, to evaluate its performance on document-grounded question answering datasets, MultiDoc2Dial and QuAC, demonstrating that self-refinement improves response quality. We further show that fine-tuning Llama-2-13B-Chat on the synthetic dialogue data generated by ProMiSe yields significant performance improvements over the zero-shot baseline as well as a supervised fine-tuned model on human annotated data.

Federal officials overseeing billions in benefit payments to millions of Canadians are hoping artificial intelligence can resolve ongoing snags in the system. The government is looking to "push the boundaries" of what artificial intelligence can do to improve a variety of services, including the pace of benefit decisions to Canadians applying for disability pensions, say documents obtained by The Canadian Press under the access to information law. Employment and Social Development Canada is currently facing processes that are "slow, inefficient, inconsistent, and prone to error," reads a presentation about the AI efforts. Instead of being able to proactively change the way federal services are delivered to Canadians, the documents say, officials are bogged down with millions of "low value" issues that need to be taken care of, crowding out more critical work. Among the "low value" work are about 50,000 tasks required to issue tax forms to Canada Pension Plan and old age security recipients.

Federal officials overseeing billions in benefit payments to millions of Canadians are hoping artificial intelligence can resolve ongoing snags in the system. The government is looking to "push the boundaries" of what artificial intelligence can do to improve a variety of services, including the pace of benefit decisions to Canadians applying for disability pensions, say documents obtained by The Canadian Press under the access to information law. Employment and Social Development Canada is currently facing processes that are "slow, inefficient, inconsistent, and prone to error," reads a presentation about the AI efforts. Instead of being able to proactively change the way federal services are delivered to Canadians, the documents say, officials are bogged down with millions of "low value" issues that need to be taken care of, crowding out more critical work. Among the "low value" work are about 50,000 tasks required to issue tax forms to Canada Pension Plan and old age security recipients.

Minister of Employment, Workforce Development and Labour Patty Hajdu speaks with Lori Sterling, deputy minister of Labour and associate deputy minister of Employment and Social Development, as they appear at a Commons human resources committee hearing on Parliament Hill in Ottawa on Monday, Feb. 12, 2018. Federal officials overseeing billions in benefit payments to millions of Canadians are hoping machine learning tools can solve ongoing snags in the system. Federal officials overseeing billions in benefit payments to millions of Canadians are hoping artificial intelligence can resolve ongoing snags in the system. The government is looking to "push the boundaries" of what artificial intelligence can do to improve a variety of services, including the pace of benefit decisions to Canadians applying for disability pensions, say documents obtained by The Canadian Press under the access to information law. Employment and Social Development Canada is currently facing processes that are "slow, inefficient, inconsistent, and prone to error," reads a presentation about the AI efforts.