Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Large Language Models for Assisting American College Applications

Zhengliang Liu, Weihang You, Peng Shu, Junhao Chen, Yi Pan, Hanqi Jiang, Yiwei Li, Zhaojun Ding, Chao Cao, Xinliang Li, Yifan Zhou, Ruidong Zhang, Shaochen Xu, Wei Ruan, Huaqin Zhao, Dajiang Zhu, Tianming Liu

TL;DR

EZCollegeApp is presented, a large language model (LLM)-powered system that assists high-school students by structuring application forms, grounding suggested answers in authoritative admissions documents, and maintaining full human control over final responses.

Abstract

American college applications require students to navigate fragmented admissions policies, repetitive and conditional forms, and ambiguous questions that often demand cross-referencing multiple sources. We present EZCollegeApp, a large language model (LLM)-powered system that assists high-school students by structuring application forms, grounding suggested answers in authoritative admissions documents, and maintaining full human control over final responses. The system introduces a mapping-first paradigm that separates form understanding from answer generation, enabling consistent reasoning across heterogeneous application portals. EZCollegeApp integrates document ingestion from official admissions websites, retrieval-augmented question answering, and a human-in-the-loop chatbot interface that presents suggestions alongside application fields without automated submission. We describe the system architecture, data pipeline, internal representations, security and privacy measures, and evaluation through automated testing and human quality assessment. Our source code is released on GitHub (https://github.com/ezcollegeapp-public/ezcollegeapp-public) to facilitate the broader impact of this work.

Large Language Models for Assisting American College Applications

TL;DR

EZCollegeApp is presented, a large language model (LLM)-powered system that assists high-school students by structuring application forms, grounding suggested answers in authoritative admissions documents, and maintaining full human control over final responses.

Abstract

American college applications require students to navigate fragmented admissions policies, repetitive and conditional forms, and ambiguous questions that often demand cross-referencing multiple sources. We present EZCollegeApp, a large language model (LLM)-powered system that assists high-school students by structuring application forms, grounding suggested answers in authoritative admissions documents, and maintaining full human control over final responses. The system introduces a mapping-first paradigm that separates form understanding from answer generation, enabling consistent reasoning across heterogeneous application portals. EZCollegeApp integrates document ingestion from official admissions websites, retrieval-augmented question answering, and a human-in-the-loop chatbot interface that presents suggestions alongside application fields without automated submission. We describe the system architecture, data pipeline, internal representations, security and privacy measures, and evaluation through automated testing and human quality assessment. Our source code is released on GitHub (https://github.com/ezcollegeapp-public/ezcollegeapp-public) to facilitate the broader impact of this work.
Paper Structure (127 sections, 4 figures)

This paper contains 127 sections, 4 figures.

Table of Contents

  1. Introduction
  2. Problem Setting and Design Principles
  3. Challenges in American College Applications
  4. Heterogeneous Form Structures
  5. Conditional Logic and Dynamic Fields
  6. Ambiguous and Context-Dependent Questions
  7. Policy and Requirement Complexity
  8. High Stakes and Low Tolerance for Error
  9. Design Principles
  10. Grounded Assistance
  11. Human-in-the-Loop
  12. Structured Reasoning
  13. Practical Deployment
  14. Extraction Over Generation
  15. Security and Privacy Measures
  16. ...and 112 more sections

Figures (4)

  • Figure 1: Security and privacy architecture of EZCollegeApp...
  • Figure 2: System architecture and workflow of EZCollegeApp. The system consists of three main components: (1) a conversational front-end for document collection and user orchestration, (2) a back-end document processing and reasoning pipeline that extracts, structures, and indexes student information, and (3) an Application Copilot browser extension that provides real-time, human-supervised assistance during form completion. The workflow progresses from document upload through processing and indexing to contextual answer suggestion.
  • Figure 3: Suggestion popup displaying profile data with carousel navigation and copy functionality.
  • Figure 4: AI editing interface showing diff view with iterative refinement options.