ComperDial: Commonsense Persona-grounded Dialogue Dataset and Benchmark
This provides a dataset and metric to improve evaluation for open-domain dialogue systems, addressing a specific need in the field.
The authors introduced ComperDial, a benchmark with human-scored responses for 10,395 dialogue turns from 99 agents, and CPDScore, a new automatic evaluation metric that shows higher correlation with human judgments than existing metrics.
We propose a new benchmark, ComperDial, which facilitates the training and evaluation of evaluation metrics for open-domain dialogue systems. ComperDial consists of human-scored responses for 10,395 dialogue turns in 1,485 conversations collected from 99 dialogue agents submitted to the Commonsense Persona-grounded Dialogue (CPD) challenge. As a result, for any dialogue, our benchmark includes multiple diverse responses with variety of characteristics to ensure more robust evaluation of learned dialogue metrics. In addition to single-turn response scores, ComperDial also contains dialogue-level human-annotated scores, enabling joint assessment of multi-turn model responses throughout a dialogue. Finally, building off ComperDial, we devise a new automatic evaluation metric to measure the general similarity of model-generated dialogues to human conversations. Our experimental results demonstrate that our novel metric, CPDScore is more correlated with human judgments than existing metrics. We release both ComperDial and CPDScore to the community to accelerate development of automatic evaluation metrics for open-domain dialogue systems.