Researchers behind a new paper on Hugging Face have introduced Gricea, described as an open-science platform for conversational AI (CAI) research. The work argues that understanding human behavior and informing the design of conversational systems requires studies conducted at scale, yet the field currently suffers from fragmented reporting of systems and study configurations.
According to the paper, that fragmentation obstructs replication, extension, and the accumulation of knowledge. Gricea addresses this by representing studies as configurable, deployable research artifacts that researchers can run, inspect, share, and reuse.
The platform was informed by a formative analysis of prior CAI research and couples study procedures, participant-facing systems, and conversational task behavior together. In a replication study using Gricea, the authors report replicating configurations from 93% of eligible CUI 2026 papers, while also flagging missing information in 96% of papers that would hinder faithful replication — a result the authors say further motivates the need for the platform.
A separate user study found that researchers and practitioners from diverse backgrounds were able to construct runnable studies addressing a range of open-ended research questions. The authors conclude that these findings demonstrate Gricea's support for constructing, reproducing, and extending CAI studies through shared research artifacts, enabling cumulative knowledge building through open science.