{"ID":2846405,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2511.03005","arxiv_id":"2511.03005","title":"Error-Aware Knowledge Distillation via Targeted Revision for Customer-Service Summarization","abstract":"We introduce an Analyze-Revise-Finetune (ARF) pipeline that enables smaller open-source language models (LLMs) to surpass substantially larger proprietary models in customer service summarization tasks. The pipeline first analyzes and categorizes common errors in summaries produced by a teacher model (GPT-3.5), then performs a targeted revision using a compact editor model (Llama 3.1 70B) to generate high-quality, refined training data. Fine-tuning smaller student models (e.g., Llama 3.1 8B, QWen3 4B) on this refined data resulted in superior summarization performance compared to GPT-3.5. The ARF pipeline improves cost efficiency and data privacy while maintaining competitive accuracy, illustrating a generalizable framework for enhancing open-source LLMs across diverse downstream applications.","short_abstract":"We introduce an Analyze-Revise-Finetune (ARF) pipeline that enables smaller open-source language models (LLMs) to surpass substantially larger proprietary models in customer service summarization tasks. The pipeline first analyzes and categorizes common errors in summaries produced by a teacher model (GPT-3.5), then pe...","url_abs":"https://arxiv.org/abs/2511.03005","url_pdf":"https://arxiv.org/pdf/2511.03005v3","authors":"[\"Hee-Jin Lee\",\"Zhen Guo\",\"Luchao Jin\",\"Morteza Moazami Goudarzi\"]","published":"2025-11-04T21:17:49Z","proceeding":"cs.CL","tasks":"[\"cs.CL\"]","methods":"[\"Large Language Model\",\"Language Model\"]","has_code":false}