Learning to Summarize from LLM-generated Feedback
Developing effective text summarizers remains a challenge due to issues like hallucinations, key information omissions, and verbosity in LLM-generated summaries. This work explores using LLM-generated feedback to improve summary quality by aligning the summaries with human preferences for faithfulness, completeness, and conciseness. We introduce FeedSum, a large-scale dataset containing multi-dimensional LLM feedback on summaries of varying quality across diverse domains. Our experiments show how feedback quality, dimensionality, and granularity influence preference learning, revealing that high-quality, multi-dimensional, fine-grained feedback significantly improves summary generation. We also compare two methods for using this feedback: supervised fine-tuning and direct preference optimization. Finally, we introduce SummLlama3-8b, a model that outperforms the nearly 10x larger Llama3-70b-instruct in generating human-preferred summaries, demonstrating that smaller models can achieve superior performance with appropriate training. The full dataset and SummLlama3-8B model are available at https://huggingface.co/datasets/DISLab/FeedSum and https://huggingface.co/DISLab/SummLlama3-8B.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Reasoning-Aware Query-Focused Summarization over Multi-Table Data
Query-focused summarization over multi-table data is a challenging yet critical task for extracting precise and relevant information from structured data. Existing methods often rely on complex preprocessing steps and st…
Logical ReasoningQuery-focused SummarizationUnsupervised Dual-Cascade Learning with Pseudo-Feedback Distillation for Query-based Extractive Summarization
We propose Dual-CES -- a novel unsupervised, query-focused, multi-document extractive summarizer. Dual-CES is designed to better handle the tradeoff between saliency and focus in summarization. To this end, Dual-CES empl…
Extractive SummarizationQuery-Based Extractive SummarizationEffects of spike-triggered negative feedback on receptive-field properties
Sensory neurons are often described in terms of a receptive field, that is, a linear kernel through which stimuli are filtered before they are further processed. If information transmission is assumed to proceed in a fee…
Reexamining zero-shot summarization: Empirical investigation of trustworthiness of LLM-summarizers
Zero-shot summarization using Large Language Models (LLMs) has significantly advanced the abstractive summarization task by producing coherent and fluent summaries. However, underlying stochasticity of the large language…
NetReAct: Interactive Learning for Network Summarization
Generating useful network summaries is a challenging and important problem with several applications like sensemaking, visualization, and compression. However, most current work in this space does not take human feedbac…