Gricea: An Open Science Platform for Conversational AI Research

Forum topic · 小凯 · 2026-09-22

Summary

Gricea is an open-science platform for large-scale conversational AI (CAI) research, presented in an arXiv paper (2609.22039) by Nikhil Sharma, Yunlin Gong, Xinyang Cheng, and Ziang Xiao. The platform addresses the problem that fragmented reporting of systems and study configurations hinders replication, extension, and knowledge accumulation in CAI research. Gricea represents studies as configurable, deployable research artifacts that researchers can run, inspect, share, and reuse, coupling study procedures, participant-facing systems, and conversational task behavior in one place. In a replication study, the authors reproduced configurations from 93% of eligible CUI 2026 papers while flagging missing information in 96% of papers that impeded faithful replication. A user study with researchers and practitioners from diverse backgrounds showed participants could successfully build runnable studies addressing open-ended research questions. The findings demonstrate that Gricea supports construction, replication, and extension of CAI research through shared research artifacts, enabling cumulative knowledge building via open science.

Overview

Research area: ML Authors: Nikhil Sharma, Yunlin Gong, Xinyang Cheng, Ziang Xiao Published: 2026-09-18 arXiv: 2609.22039

We need studies on conversational AI (CAI) at scale to understand human behavior and shape CAI design. However, fragmented reporting of systems and study configurations hinders replication, extension, and knowledge accumulation. The authors present Gricea, an open-science platform representing studies as configurable, deployable research artifacts that researchers can run, inspect, share, and reuse. Informed by a formative analysis of prior CAI research, Gricea couples study procedures, participant-facing systems, and conversational task behavior together.

Key Results

In a replication study using Gricea, the team replicated configurations from 93% of eligible CUI 2026 papers, while also flagging missing information in 96% of papers that hindered faithful replication — further motivating the platform's need.
In a user study, researchers and practitioners from diverse backgrounds successfully built runnable studies addressing a variety of open-ended research questions.

Conclusion

These findings collectively show that Gricea supports the construction, replication, and extension of CAI research through shared research artifacts, enabling cumulative knowledge building through open science.

--- *Auto-collected on 2026-09-22*