Publications

REPAIR-Bench: A Benchmark for Robot Error Perception and Interaction Recovery

Published in Submitted to IROS 2026, 2026

A benchmark of 214 human-robot interaction trials from 41 participants with synchronized multimodal signals (facial action units, head pose, speech, and post-interaction affect), formalizing three novel evaluation tasks across the full failure lifecycle. Hierarchical recurrent modeling improves failure detection over single-session baselines (strict F1: 0.80 vs. 0.68).

Recommended citation: G. Pioldi, Y. Batra, Y. Bai, P. Marur, A. Ibrayeva, Promise Ekpo, A. Taylor. "REPAIR-Bench: A Benchmark for Robot Error Perception and Interaction Recovery." Submitted to IROS, 2026.

RFM-HRI: A Multimodal Dataset of Medical Robot Failure, User Reaction, and Recovery Preferences for Item Retrieval Tasks

Published in ACM Transactions on Human-Robot Interaction (THRI), accepted 2026, 2026

A multimodal dataset capturing 214 human-robot interactions from 41 laypersons and healthcare workers responding to crash-cart robot failures in an item-retrieval task, with facial action units (OpenFace) and speech transcription (Whisper) to quantify reactions and preferred recovery modes in embodied AI.

Recommended citation: Y. Batra, Promise Ekpo, G. Pioldi, P. Marur, A. Ibrayeva, A. Taylor. "RFM-HRI: A Multimodal Dataset of Medical Robot Failure, User Reaction, and Recovery Preferences for Item Retrieval Tasks." THRI, 2026 (accepted).

A Generalized Nash Equilibrium-Seeking Scheme for Trauma Resuscitation

Published in IFAC World Congress, accepted 2026, 2026

A distributed generalized Nash equilibrium (GNE) algorithm for multi-agent coordination in trauma resuscitation, achieving state consensus for n=7 healthcare workers with heterogeneous dual variables reflecting network topology.

Recommended citation: Promise Ekpo, A. Taylor, L. Molu. "A Generalized Nash Equilibrium-Seeking Scheme for Trauma Resuscitation." IFAC World Congress, 2026 (accepted).

Integrated Artificial Intelligence in Healthcare and the Patient’s Experience of Care

Published in Scientific Reports (Nature), accepted, 2026

A study of how integrated AI systems in healthcare shape the patient experience of care.

Recommended citation: O. Ogundare, T. Owadokun, T. Ogundare, Promise Ekpo, H. L. Nguyen, S. Bello. "Integrated Artificial Intelligence in Healthcare and the Patients Experience of Care." Scientific Reports (Nature), accepted.

Skill-Aligned Fairness in Multi-Agent Learning for Collaboration in Healthcare (FairSkillMARL)

Published in Submitted to NeurIPS 2026 Datasets & Benchmarks Track, 2025

A skill-aligned fairness framework and the MARLHospital simulator for evaluating fairness-efficiency trade-offs in heterogeneous multi-agent teams, achieving 40-60% higher skill-task alignment than workload-only baselines with 70-85% reduction in workload-range disparity.

Recommended citation: Promise Ekpo, B. La, T. Wiener, S. Agarwal, A. Agrawal, G. Gonzalez-Pumariega, L. Molu, A. Taylor. "Skill-Aligned Fairness in Multi-Agent Learning for Collaboration in Healthcare." Submitted, NeurIPS 2026 Datasets & Benchmarks Track. https://arxiv.org/abs/2508.18708

AdaFair-MARL: Enforcing Adaptive Fairness Constraints in Multi-Agent Reinforcement Learning

Published in Submitted to IEEE Conference on Decision and Control (CDC) 2025, 2025

An adaptive fairness-constraint enforcement method for multi-agent reinforcement learning with heterogeneous teams, using automatic penalty-parameter learning via Lagrangian dual ascent.

Recommended citation: Promise Ekpo, S. Agarwal, F. Grimm, J. Liu, L. Molu, A. Taylor. "AdaFair-MARL: Enforcing Adaptive Fairness Constraints in Multi-Agent Reinforcement Learning." Submitted, IEEE CDC, 2025. https://arxiv.org/pdf/2511.14135

Human-Robot Teaming Field Deployments: A Comparison Between Verbal and Non-verbal Communication

Published in IEEE RO-MAN Workshop, 2025 (accepted), 2025

A field deployment of a robotic crash cart with 115 healthcare workers at Weill Cornell Medicine, comparing verbal vs. non-verbal robot communication with statistical analysis of workload and attitudes.

Recommended citation: T. Tanjim, Promise Ekpo, H. Cao, J. St. George, K. Ching, H. R. Lee, A. Taylor. "Human-Robot Teaming Field Deployments: A Comparison Between Verbal and Non-verbal Communication." IEEE RO-MAN Workshop, 2025. https://arxiv.org/pdf/2506.08890

Investigating Persuasiveness in Large Language Models

Published in M.A. Thesis, Princeton University, 2023

M.A. thesis quantifying the persuasiveness of GPT-3/4 using a game-theoretic framework, demonstrating 12-50% belief shift toward false statements and uncovering systematic social bias in generated arguments.

Recommended citation: Promise Ekpo. "Investigating Persuasiveness in Large Language Models." M.A. Thesis, Princeton University, 2023.

Promise Ekpo Osaine

Publications