OceanCBM is the first concept bottleneck model for spatiotemporal ocean prediction that uses mixed supervision on physical concepts and a free concept to deliver consistent mechanistic representations for mixed layer heat content forecasts.
Concept bottleneck models
7 Pith papers cite this work. Polarity classification is still indexing.
7
Pith papers citing it
citation-role summary
background 1
extension 1
method 1
citation-polarity summary
years
2026 7verdicts
UNVERDICTED 7representative citing papers
HypCBM reformulates concept activations as geometric containment in hyperbolic space to produce sparse, hierarchy-aware signals that match Euclidean models trained on 20 times more data.
ShapShift explains prediction shifts by attributing them to changes in conditional probabilities of tree-defined subgroups via conditional Shapley values, with exact computation for single trees and surrogate extensions for other models.
YUV20K is a complexity-driven VCOD benchmark with 24k annotated frames, paired with a model using Motion Feature Stabilization via semantic primitives and Trajectory-Aware Alignment via deformable sampling that outperforms prior methods.
SoftSAE replaces fixed-K sparsity in autoencoders with a learned, input-dependent number of active features via a soft top-k operator.
Benchmark shows that combining data rebalancing with feature disentanglement mitigates shortcut learning more effectively than rebalancing alone in medical imaging models.
HTAF is a sigmoid-tanh composite that approximates the Heaviside function to allow stable gradient training of binary activation networks, yielding ICBMs with stable discretization and competitive performance on image tasks.
citing papers explorer
-
OceanCBM: A Concept Bottleneck Model for Mechanistic Interpretability in Ocean Forecasting
OceanCBM is the first concept bottleneck model for spatiotemporal ocean prediction that uses mixed supervision on physical concepts and a free concept to deliver consistent mechanistic representations for mixed layer heat content forecasts.
-
Hyperbolic Concept Bottleneck Models
HypCBM reformulates concept activations as geometric containment in hyperbolic space to produce sparse, hierarchy-aware signals that match Euclidean models trained on 20 times more data.
-
ShapShift: Explaining Model Prediction Shifts with Subgroup Conditional Shapley Values
ShapShift explains prediction shifts by attributing them to changes in conditional probabilities of tree-defined subgroups via conditional Shapley values, with exact computation for single trees and surrogate extensions for other models.
-
YUV20K: A Complexity-Driven Benchmark and Trajectory-Aware Alignment Model for Video Camouflaged Object Detection
YUV20K is a complexity-driven VCOD benchmark with 24k annotated frames, paired with a model using Motion Feature Stabilization via semantic primitives and Trajectory-Aware Alignment via deformable sampling that outperforms prior methods.
-
SoftSAE: Dynamic Top-K Selection for Adaptive Sparse Autoencoders
SoftSAE replaces fixed-K sparsity in autoencoders with a learned, input-dependent number of active features via a soft top-k operator.
-
Mitigating Shortcut Learning via Feature Disentanglement in Medical Imaging: A Benchmark Study
Benchmark shows that combining data rebalancing with feature disentanglement mitigates shortcut learning more effectively than rebalancing alone in medical imaging models.
-
A Composite Activation Function for Learning Stable Binary Representations
HTAF is a sigmoid-tanh composite that approximates the Heaviside function to allow stable gradient training of binary activation networks, yielding ICBMs with stable discretization and competitive performance on image tasks.