1
Counterfactual Explanations Under Concept Drift
探索概念漂移下如何生成稳健的反事实解释，机器学习可解释性前沿研究。
arXiv:2605.17651v1 Announce Type: new Abstract: Counterfactual explanations (CFEs) provide actionable recourse, but most methods assume a static frame…
2
When Bits Break Recourse: Counterfactual-Faithful Quantization
量化压缩如何破坏反事实解释的可信度？论文提出反事实忠诚量化方法，为模型压缩与可解释性的矛盾提供新解法。
arXiv:2605.17160v1 Announce Type: new Abstract: Quantization can preserve predictive accuracy under low-bit deployment while silently breaking algorit…