TIGTEC : Token Importance Guided TExt Counterfactuals
This provides a modular tool for explaining AI decisions in text applications, though it is incremental as it builds on existing counterfactual generation techniques.
The paper tackles the problem of generating counterfactual explanations for text classifiers by proposing TIGTEC, a method that uses token importance to edit text, resulting in efficient production of sparse, plausible, and diverse counterfactuals with high success rates.
Counterfactual examples explain a prediction by highlighting changes of instance that flip the outcome of a classifier. This paper proposes TIGTEC, an efficient and modular method for generating sparse, plausible and diverse counterfactual explanations for textual data. TIGTEC is a text editing heuristic that targets and modifies words with high contribution using local feature importance. A new attention-based local feature importance is proposed. Counterfactual candidates are generated and assessed with a cost function integrating semantic distance, while the solution space is efficiently explored in a beam search fashion. The conducted experiments show the relevance of TIGTEC in terms of success rate, sparsity, diversity and plausibility. This method can be used in both model-specific or model-agnostic way, which makes it very convenient for generating counterfactual explanations.