# CoMAI: A Collaborative Multi-Agent Framework for Robust and Equitable Interview Evaluation

Source: [CoMAI: A Collaborative Multi-Agent Framework for Robust and Equitable Interview Evaluation](<https://www.sota2.com/research/paper/69ba31ae0636e5bf4afc0a16>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-03-19T05:00:05.030Z

Authors: Gengxin Sun, Ruihao Yu, Liangyi Yin, Yunqi Yang, Bin Zhang, Zhiwei Xu

Published: 2026-03-17T07:44:12.000Z

## Abstract

Ensuring robust and fair interview assessment remains a key challenge in AI-driven evaluation. This paper presents CoMAI, a general-purpose multi-agent interview framework designed for diverse assessment scenarios. In contrast to monolithic single-agent systems based on large language models (LLMs), CoMAI employs a modular task-decomposition architecture coordinated through a centralized finite-state machine. The system comprises four agents specialized in question generation, security, scoring, and summarization. These agents work collaboratively to provide multi-layered security defenses against prompt injection, support multidimensional evaluation with adaptive difficulty adjustment, and enable rubric-based structured scoring that reduces subjective bias. Experimental results demonstrate that CoMAI achieved 90.47% accuracy, 83.33% recall, and 84.41% candidate satisfaction. These results highlight CoMAI as a robust, fair, and interpretable paradigm for AI-driven interview assessment.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Adversarial Prompt Injection Defense on 500 adversarial samples](<https://www.sota2.com/research/sota/adversarial-prompt-injection-defense-on-500-adversarial-samples.md>): Adversarial Prompt Injection Defense; 500 adversarial samples; Defense Success Rate: 100.

- [Interview Assessment on University Admission Interview Dataset 55 candidates (test)](<https://www.sota2.com/research/sota/interview-assessment-on-university-admission-interview-dataset-55-candidates-test.md>): Interview Assessment; University Admission Interview Dataset 55 candidates (test); Recall: 95.45.

- [Interview Assessment on Interview User Study (55 participants)](<https://www.sota2.com/research/sota/interview-assessment-on-interview-user-study-55-participants.md>): Interview Assessment; Interview User Study (55 participants); Satisfaction: 84.41.

- [Interview Scoring and Admission Analysis on Interview Score Distribution (averaged across 55 participants)](<https://www.sota2.com/research/sota/interview-scoring-and-admission-analysis-on-interview-score-distribution-averaged-across-55-participants.md>): Interview Scoring and Admission Analysis; Interview Score Distribution (averaged across 55 participants); Mean Score: 62.92.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2603.16215>)

[PDF](<https://arxiv.org/pdf/2603.16215.pdf>)
