InterviewBot: Real-Time End-to-End Dialogue System to Interview Students for College Admission
This work addresses the need for scalable and efficient assessment of academic and cultural readiness for college admissions, though it is incremental as it builds on existing transformer-based models with new methods for handling conversation history and topics.
The authors tackled the problem of automating college admission interviews by developing InterviewBot, an end-to-end dialogue system that conducts 10-minute hybrid-domain conversations with foreign students, achieving high satisfaction in fluency and context awareness based on real-time testing with professionals and students.
We present the InterviewBot that dynamically integrates conversation history and customized topics into a coherent embedding space to conduct 10 mins hybrid-domain (open and closed) conversations with foreign students applying to U.S. colleges for assessing their academic and cultural readiness. To build a neural-based end-to-end dialogue model, 7,361 audio recordings of human-to-human interviews are automatically transcribed, where 440 are manually corrected for finetuning and evaluation. To overcome the input/output size limit of a transformer-based encoder-decoder model, two new methods are proposed, context attention and topic storing, allowing the model to make relevant and consistent interactions. Our final model is tested both statistically by comparing its responses to the interview data and dynamically by inviting professional interviewers and various students to interact with it in real-time, finding it highly satisfactory in fluency and context awareness.