arXiv 2508.118732025-08-16cs.CYcs.AIcs.HCcs.MM

SimInterview：通过基于大语言模型的模拟多语言面试训练系统变革商业教育

SimInterview: Transforming Business Education through Large Language Model-Based Simulated Multilingual Interview Training System

Analytics Everywhere Lab, University of New Brunswick, Canada（新不伦瑞克大学分析 everywhere 实验室）；University of Foreign Language Studies, University of Danang, Vietnam（越南丹绒大学外语学院）；Faculty of Information Technology, University of Science, VNU-HCM, Vietnam（越南胡志明市大学信息科技学院）；Faculty of Economics and Accounting, Quy Nhon University, Vietnam（越南奎隆大学经济与会计学院）；Faculty of Natural Sciences, Quy Nhon University, Vietnam（越南奎隆大学自然科学学院）

Truong Thanh Hung Nguyen, Tran Diem Quynh Nguyen, Hoang Loc Cao, Thi Cam Thanh Tran, Thi Cam Mai Truong, Hung Cao

备注：: Published as a conference paper at ICEFM 2025

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

SimInterview利用大语言模型、RAG及语音与数字人技术构建多语言模拟面试训练系统，实验证明其能对齐职位要求、保留简历内容并获高满意度，同时提出可争议AI设计以应对监管需求。

摘要：

商务面试准备既需要扎实的理论基础，也需要精细的软技能，然而传统课堂教学方法很少能提供雇主当前所期望的个性化、具有文化意识的练习。本文介绍了SimInterview，一个基于大语言模型（LLM）的模拟多语言面试训练系统，专为进入人工智能变革劳动力市场的商务专业人士设计。我们的系统利用LLM智能体和合成人工智能技术创建逼真的虚拟招聘人员，能够进行个性化、实时的对话式面试。该框架使用检索增强生成（RAG）动态调整面试场景，以在多种语言中将个人简历与具体职位要求相匹配。系统构建于LLM（OpenAI o3、Llama 4 Maverick、Gemma 3）之上，并集成了Whisper语音识别、GPT-SoVITS语音合成、基于Ditto扩散的说话人头像生成模型以及ChromaDB向量数据库，显著提升了英语和日语市场的面试准备水平。针对大学水平候选人的实验表明，该系统始终使其评估与职位要求保持一致，忠实保留简历内容，并获得高满意度评分，其中轻量级Gemma 3模型产生了最具吸引力的对话。定性研究结果显示，标准化的日语简历格式改善了文档检索，而多样化的英语简历引入了额外的变异性，同时研究还揭示了文化规范如何塑造追问策略。最后，我们还概述了一种可争议的人工智能设计，该设计能够进行解释、检测偏见并保留人在回路，以满足新兴的监管期望。

英文摘要：

Business interview preparation demands both solid theoretical grounding and refined soft skills, yet conventional classroom methods rarely deliver the individualized, culturally aware practice employers currently expect. This paper introduces SimInterview, a large language model (LLM)-based simulated multilingual interview training system designed for business professionals entering the AI-transformed labor market. Our system leverages an LLM agent and synthetic AI technologies to create realistic virtual recruiters capable of conducting personalized, real-time conversational interviews. The framework dynamically adapts interview scenarios using retrieval-augmented generation (RAG) to match individual resumes with specific job requirements across multiple languages. Built on LLMs (OpenAI o3, Llama 4 Maverick, Gemma 3), integrated with Whisper speech recognition, GPT-SoVITS voice synthesis, Ditto diffusion-based talking head generation model, and ChromaDB vector databases, our system significantly improves interview readiness across English and Japanese markets. Experiments with university-level candidates show that the system consistently aligns its assessments with job requirements, faithfully preserves resume content, and earns high satisfaction ratings, with the lightweight Gemma 3 model producing the most engaging conversations. Qualitative findings revealed that the standardized Japanese resume format improved document retrieval while diverse English resumes introduced additional variability, and they highlighted how cultural norms shape follow-up questioning strategies. Finally, we also outlined a contestable AI design that can explain, detect bias, and preserve human-in-the-loop to meet emerging regulatory expectations.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。