Channels - Transformers Efficiently Perform In-Context Logistic Regression via Normalized Gradient Descent :: FRELIP Discovery

Similar Items: Transformers Efficiently Perform In-Context Logistic Regression via Normalized Gradient Descent

Quick Look
Complex Equation Learner: Rational Symbolic Regression with Gradient Descent in Complex Domain
Quick Look
Understanding In-Context Learning for Nonlinear Regression with Transformers: Attention as Featurizer
Quick Look
Normalizing Trajectory Models
Quick Look
Randomized Subspace Nesterov Accelerated Gradient
Quick Look
Decentralized Proximal Stochastic Gradient Langevin Dynamics
Quick Look
On the Wasserstein Gradient Flow Interpretation of Drifting Models
Quick Look
A decoupled diffusion planner that adapts to changing cost limits by using cost-conditioned generation for safety and reward gradients for performance
Quick Look
STARFlow2: Bridging Language Models and Normalizing Flows for Unified Multimodal Generation
Quick Look
Auto-FlexSwitch: Efficient Dynamic Model Merging via Learnable Task Vector Compression
Quick Look
Label-Efficient School Detection from Aerial Imagery via Weakly Supervised Pretraining and Fine-Tuning
Quick Look
Spiking Sequence Machines and Transformers
Quick Look
Fast Byte Latent Transformer
Quick Look
Transformers with Selective Access to Early Representations
Quick Look
Taming Outlier Tokens in Diffusion Transformers
Quick Look
Joint Treatment Effect Estimation from Incomplete Healthcare Data: Temporal Causal Normalizing Flows with LLM-driven Evolutionary MNAR Imputation
Quick Look
Transformed Latent Variable Multi-Output Gaussian Processes
Quick Look
Physiologically Grounded Driver Behavior Classification: SHAP-Driven Elite Feature Selection and Hybrid Gradient Boosting for Multimodal Physiological Signals
Quick Look
DEFault++: Automated Fault Detection, Categorization, and Diagnosis for Transformer Architectures
Quick Look
Memory-Efficient Continual Learning with CLIP Models
Quick Look
Trust, but Verify: Peeling Low-Bit Transformer Networks for Training Monitoring
Quick Look
Semiparametric Efficient Test for Interpretable Distributional Treatment Effects
Quick Look
Beyond Gaussian Bottlenecks: Topologically Aligned Encoding of Vision-Transformer Feature Spaces
Quick Look
Efficient Multivector Retrieval with Token-Aware Clustering and Hierarchical Indexing
Quick Look
Superposition Is Not Necessary: A Mechanistic Interpretability Analysis of Transformer Representations for Time Series Forecasting