An iERF-centric framework unifies local, global, and mechanistic interpretability in vision models via SRD for saliency, CAFE for concept anchoring, and ICAT for interlayer attribution.
Learning important features through propagating activation differences
10 Pith papers cite this work. Polarity classification is still indexing.
10
Pith papers citing it
representative citing papers
XtrAIn shifts occlusion from input space to parameter space along the training trajectory to produce cleaner feature attributions than standard methods.
A triplet-based plateau search algorithm is proposed to adaptively determine a near-minimal number of trees for random forests by monitoring relative OOB score changes across forest size triplets, removing n_trees from the TPE search space.
Scaling vision models by depth and parameter count does not consistently improve localisation-based explanation quality across architectures, datasets, and post-hoc methods; smaller models often perform comparably or better.
New unsupervised method adapts the multivariate logrank statistic into a differentiable loss for training any neural network on any data modality to discover prognostically distinct patient clusters, demonstrated on myeloma lab data and lung cancer CT images with post-hoc explainability.
GraphPINE is a GNN architecture that initializes node importance from prior knowledge graphs and propagates updates via an importance propagation layer for interpretable drug response prediction on over 5,000 genes and 952 drugs.
CodeQ aggregates token rationales into code categories to enable global interpretability of LLMs, claiming over 50% entropy reduction and revealing model preference for syntactic cues plus human misalignment in a 37-person study.
3D CNNs predict elastic moduli of nanoporous metals with R²=0.955, outperforming descriptor-based models, and transfer learning works on smaller denser datasets for large-scale Pareto optimization.
Explores reference document choices for applying DeepSHAP to neural retrieval models and reports that its explanations differ substantially from those of LIME.
CNN trigger for QGP events reaches 83.7% accuracy on reconstructed Au+Au events at 30 AGeV after training on PHSD and cross-validation on UrQMD, with deployment via lightweight C++ package.
citing papers explorer
-
From Local to Global to Mechanistic: An iERF-Centered Unified Framework for Interpreting Vision Models
An iERF-centric framework unifies local, global, and mechanistic interpretability in vision models via SRD for saliency, CAFE for concept anchoring, and ICAT for interlayer attribution.
-
XtrAIn: Training-Guided Occlusion for Feature Attribution
XtrAIn shifts occlusion from input space to parameter space along the training trajectory to produce cleaner feature attributions than standard methods.
-
How Many Trees in a Random Forest? A Revisited Approach with Plateau Search and Optuna Integration
A triplet-based plateau search algorithm is proposed to adaptively determine a near-minimal number of trees for random forests by monitoring relative OOB score changes across forest size triplets, removing n_trees from the TPE search space.
-
Scaling Vision Models Does Not Consistently Improve Localisation-Based Explanation Quality
Scaling vision models by depth and parameter count does not consistently improve localisation-based explanation quality across architectures, datasets, and post-hoc methods; smaller models often perform comparably or better.
-
Unsupervised risk factor identification across cancer types and data modalities via explainable artificial intelligence
New unsupervised method adapts the multivariate logrank statistic into a differentiable loss for training any neural network on any data modality to discover prognostically distinct patient clusters, demonstrated on myeloma lab data and lung cancer CT images with post-hoc explainability.
-
GraphPINE: Graph Importance Propagation for Interpretable Drug Response Prediction
GraphPINE is a GNN architecture that initializes node importance from prior knowledge graphs and propagates updates via an importance propagation layer for interpretable drug response prediction on over 5,000 genes and 952 drugs.
-
Enabling Global, Human-Centered Explanations for LLMs:From Tokens to Interpretable Code and Test Generation
CodeQ aggregates token rationales into code categories to enable global interpretability of LLMs, claiming over 50% entropy reduction and revealing model preference for syntactic cues plus human misalignment in a 37-person study.
-
Transferable 3D Convolutional Neural Networks for Elastic Constants Prediction in Nanoporous Metals
3D CNNs predict elastic moduli of nanoporous metals with R²=0.955, outperforming descriptor-based models, and transfer learning works on smaller denser datasets for large-scale Pareto optimization.
-
A study on the Interpretability of Neural Retrieval Models using DeepSHAP
Explores reference document choices for applying DeepSHAP to neural retrieval models and reports that its explanations differ substantially from those of LIME.
-
CNN-Based Online Trigger for QGP Event Selection
CNN trigger for QGP events reaches 83.7% accuracy on reconstructed Au+Au events at 30 AGeV after training on PHSD and cross-validation on UrQMD, with deployment via lightweight C++ package.