SPA unlocks patch-level features in CLIP for class-incremental learning via semantic-guided selection and optimal transport alignment with class descriptions, plus projectors and pseudo-feature replay to reduce forgetting.
hub
Sinkhorn distances: Lightspeed computation of optimal transport.Advances in neural information processing systems, 26
11 Pith papers cite this work. Polarity classification is still indexing.
11
Pith papers citing it
hub tools
citation-role summary
method 2
citation-polarity summary
roles
method 2polarities
use method 2representative citing papers
Contrastive predictive coding pretraining combined with structured state space models yields the strongest ECG foundation models, with continued gains from scaling data to 11 million samples.
Differentiable nonconformity scores induce flows that sample conformal prediction set boundaries, and mixing flows across levels produces conformal predictive distributions whose quantiles match the sets.
4D-GSW introduces a kinematic-aware spatio-temporal watermarking framework for 4D Gaussian Splatting that uses a Spatio-Temporal Curvature metric and HMM-MRF model to maintain consistency under attacks.
OTT-Vid uses optimal transport with non-uniform token mass and locality-aware costs to dynamically allocate compression budgets across video frames, retaining 95.8% VQA and 73.9% VTG performance at 10% token retention.
CapsID uses probabilistic capsule routing and confidence-based termination to generate variable-length semantic IDs, improving recall by 9.6% over strong baselines with half the latency of dual-representation systems.
Branched Normalizing Flow improves conditional coverage robustness of conformal prediction under distribution shift by normalizing test inputs to the calibration distribution and mapping prediction sets back.
Span-level Wasserstein distances between hidden-state distributions of correct and incorrect rollouts provide a self-supervised signal to reweight advantages in GRPO, improving fine-grained credit assignment on math and code tasks.
A reduced-order model for parametrized optimal transport problems using low-dimensional cone or subspace constraints and EIM-based error estimation.
EnFlow integrates flow-based conformer generation with energy landscape modeling to enable joint ensemble generation and ground-state identification using only 1-2 ODE steps.
A new benchmarking framework shows virtual cell models overestimate performance on standard tests, drop sharply on unseen contexts and perturbations, and produce inconsistent rankings across metrics.
citing papers explorer
-
Unlocking Patch-Level Features for CLIP-Based Class-Incremental Learning
SPA unlocks patch-level features in CLIP for class-incremental learning via semantic-guided selection and optimal transport alignment with class descriptions, plus projectors and pseudo-feature replay to reduce forgetting.
-
Pretraining Strategies and Scaling for ECG Foundation Models: A Systematic Study
Contrastive predictive coding pretraining combined with structured state space models yields the strongest ECG foundation models, with continued gains from scaling data to 11 million samples.
-
Flow-Based Conformal Predictive Distributions
Differentiable nonconformity scores induce flows that sample conformal prediction set boundaries, and mixing flows across levels produces conformal predictive distributions whose quantiles match the sets.
-
4D-GSW: Kinematic-Aware Spatio-Temporal Consistent Watermarking for 4D Gaussian Splatting
4D-GSW introduces a kinematic-aware spatio-temporal watermarking framework for 4D Gaussian Splatting that uses a Spatio-Temporal Curvature metric and HMM-MRF model to maintain consistency under attacks.
-
OTT-Vid: Optimal Transport Temporal Token Compression for Video Large Language Models
OTT-Vid uses optimal transport with non-uniform token mass and locality-aware costs to dynamically allocate compression budgets across video frames, retaining 95.8% VQA and 73.9% VTG performance at 10% token retention.
-
CapsID: Soft-Routed Variable-Length Semantic IDs for Generative Recommendation
CapsID uses probabilistic capsule routing and confidence-based termination to generate variable-length semantic IDs, improving recall by 9.6% over strong baselines with half the latency of dual-representation systems.
-
Robust Conditional Conformal Prediction via Branched Normalizing Flow
Branched Normalizing Flow improves conditional coverage robustness of conformal prediction under distribution shift by normalizing test inputs to the calibration distribution and mapping prediction sets back.
-
Hidden States Know Where Reasoning Diverges: Credit Assignment via Span-Level Wasserstein Distance
Span-level Wasserstein distances between hidden-state distributions of correct and incorrect rollouts provide a self-supervised signal to reweight advantages in GRPO, improving fine-grained credit assignment on math and code tasks.
-
A reduced-order model for parametrized Optimal Transport problems
A reduced-order model for parametrized optimal transport problems using low-dimensional cone or subspace constraints and EIM-based error estimation.
-
Energy-Guided Generative Modeling for Low-Energy Molecular Structure Discovery
EnFlow integrates flow-based conformer generation with energy landscape modeling to enable joint ensemble generation and ground-state identification using only 1-2 ODE steps.
-
Benchmarking virtual cell models for in-the-wild perturbation response
A new benchmarking framework shows virtual cell models overestimate performance on standard tests, drop sharply on unseen contexts and perturbations, and produce inconsistent rankings across metrics.