It HAS to be Subjective: Human Annotator Simulation via Zero-shot Density Estimation
This work addresses the need for efficient and realistic human evaluation simulation in machine learning, though it is incremental as it builds on existing meta-learning and density estimation methods.
The paper tackles the problem of simulating human annotators for cost-effective evaluation by proposing a zero-shot density estimation framework that incorporates human variability, resulting in superior performance in predicting aggregated behaviors, matching annotation distributions, and simulating inter-annotator disagreements across three real-world tasks.
Human annotator simulation (HAS) serves as a cost-effective substitute for human evaluation such as data annotation and system assessment. Human perception and behaviour during human evaluation exhibit inherent variability due to diverse cognitive processes and subjective interpretations, which should be taken into account in modelling to better mimic the way people perceive and interact with the world. This paper introduces a novel meta-learning framework that treats HAS as a zero-shot density estimation problem, which incorporates human variability and allows for the efficient generation of human-like annotations for unlabelled test inputs. Under this framework, we propose two new model classes, conditional integer flows and conditional softmax flows, to account for ordinal and categorical annotations, respectively. The proposed method is evaluated on three real-world human evaluation tasks and shows superior capability and efficiency to predict the aggregated behaviours of human annotators, match the distribution of human annotations, and simulate the inter-annotator disagreements.