{"ID":2867070,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.18864","arxiv_id":"2509.18864","title":"Conf-Profile: A Confidence-Driven Reasoning Paradigm for Label-Free User Profiling","abstract":"User profiling, as a core technique for user understanding, aims to infer structural attributes from user information. Large Language Models (LLMs) provide a promising avenue for user profiling, yet the progress is hindered by the lack of comprehensive benchmarks. To bridge this gap, we propose ProfileBench, an industrial benchmark derived from a real-world video platform, encompassing heterogeneous user data and a well-structured profiling taxonomy. However, the profiling task remains challenging due to the difficulty of collecting large-scale ground-truth labels, and the heterogeneous and noisy user information can compromise the reliability of LLMs. To approach label-free and reliable user profiling, we propose a Confidence-driven Profile reasoning framework Conf-Profile, featuring a two-stage paradigm. We first synthesize high-quality labels by leveraging advanced LLMs with confidence hints, followed by confidence-weighted voting for accuracy improvement and confidence calibration for a balanced distribution. The multiple profile results, rationales, and confidence scores are aggregated and distilled into a lightweight LLM. We further enhance the reasoning ability via confidence-guided unsupervised reinforcement learning, which exploits confidence for difficulty filtering, quasi-ground truth voting, and reward weighting. Experimental results demonstrate that Conf-Profile delivers substantial performance through the two-stage training, improving F1 by 13.97 on Qwen3-8B.","short_abstract":"User profiling, as a core technique for user understanding, aims to infer structural attributes from user information. Large Language Models (LLMs) provide a promising avenue for user profiling, yet the progress is hindered by the lack of comprehensive benchmarks. To bridge this gap, we propose ProfileBench, an industr...","url_abs":"https://arxiv.org/abs/2509.18864","url_pdf":"https://arxiv.org/pdf/2509.18864v1","authors":"[\"Yingxin Li\",\"Jianbo Zhao\",\"Xueyu Ren\",\"Jie Tang\",\"Wangjie You\",\"Xu Chen\",\"Kan Zhou\",\"Chao Feng\",\"Jiao Ran\",\"Yuan Meng\",\"Zhi Wang\"]","published":"2025-09-23T09:58:37Z","proceeding":"cs.AI","tasks":"[\"cs.AI\"]","methods":"[\"Reinforcement Learning\",\"Large Language Model\",\"Language Model\"]","has_code":false}