Most previous research in counterfactual generation has focused on the model side by either generating counterfactuals to improve the model's performance or explaining its behaviors post hoc.
any single sentence that compares and contrasts this work with prior work.
While SAT-based rendering supported human sensemaking in both Gero et al. [29] and Mocha, we also show that the combination of VT and SAT support the model's learning.
any single sentence that compares and contrasts this work with prior work.
This finding is consistent with previous work that supports users' sense-making of text, e.g., by modulating text saliency. Specifically, Gu et al. [32] and Gero et al. [29] both found improved reading efficiency and comprehension with saliency-modulating text renderings.
any single sentence that compares and contrasts this work with prior work.