cs.LGSep 7, 2026

AVCG: A Generalized Variational Framework for Counterfactual Generation under Hypothesis Distributions

Authors: Jamie DuellAlejandro Jimenez RodriguezMahault Albarracin

Organizations: School of Computing and Digital Technologies, Sheffield Hallam University · Centre of Excellence in AI and Robotics (CEAIR), Sheffield Hallam University · 3Laboratoire d’analyse cognitive de l’information (LANCI), Université du Québec à Montréal · 4Institut Santé et société (ISS), Université du Québec à Montréal · 5Institut de recherches et d’études féministes (IREF), Université du Québec à Montréal

Abstract

Counterfactual explanations formalize "what-if" scenarios by identifying modifications to an input instance that obtain a desired alternative prediction. Traditionally, whether generated via instance-specific optimization or amortized single pass models, these approaches rely on a single, deterministic point-estimate predictor. However, this ignores predictive uncertainty and hypothesis variability, leading to brittle explanations that frequently become invalid if the underlying model is retrained or updated. To address this fragility, we propose the Amortized Variational Counterfactual Generator (AVCG), a generalized optimization framework that formulates counterfactual generation as optimization over an arbitrary distribution of plausible predictive hypotheses rather than a single deterministic predictor. This formulation naturally accommodates Bayesian posteriors, Rashomon-restricted hypothesis spaces, and other uncertainty representations within a unified optimization framework. Evaluation across multiple benchmark datasets demonstrates that the AVCG framework produces counterfactual explanations that remain highly valid under predictive uncertainty and model changes, while maintaining competitive plausibility and single-pass runtime performance.

Explore similar work

CardsList