archive
Every paper Pith has read. Search by title, abstract, or pith.
14903 papers in cs.LG · page 11
-
Mean UA-RAO with ensembles beats deterministic baselines in PQD localization
A Unified Framework for Uncertainty-Aware Explainable Artificial Intelligence: A Case Study in Power Quality Disturbance Classification
-
Parallel Monte Carlo trains deep state space models 10x faster
Efficient Learning of Deep State Space Models via Importance Smoothing
-
Projection algorithm reduces constraint violations to O(log T)
Improved Guarantees for Constrained Online Convex Optimization via Self-Contraction
-
Hyperbolic operator adds L1 bias to stable sparse transformer training
HORST: Composing Optimizer Geometries for Sparse Transformer Training
-
FL programs factor through fixed shared state
A Typed Tensor Language for Federated Learning
-
Unbalanced OT learns unique map from noisy to clean images without pairs
UOTIP: Unbalanced Optimal Transport Map for Unpaired Inverse Problems
-
Separate corrector cuts error buildup in deep forecasts
Reviving Error Correction in Modern Deep Time-Series Forecasting
-
Decoupled messages sustain MARL performance at low bandwidth
Decoupling Communication from Policy: Robust MARL under Bandwidth Constraints
-
Blueprint couples materials and biomedical data into governed AI workflows
AIMBio-Mat: An AI-Native FAIR Platform for Closed-Loop Materials Discovery and Biomedical Translation
-
Music attention uses metadata to cut repetition in generated melodies
Musical Attention Transformer: Music Generation Using a Music-Specific Attention Model
-
New transformer fuses hyperspectral imagery with other EO sensors
SpectralEarth-FM: Bringing Hyperspectral Imagery into Multimodal Earth Observation Pretraining
-
Self-pretraining uncovers attention patterns labels cannot reach
Towards Understanding Self-Pretraining for Sequence Classification
-
Personalized bounds deconfound recommendations without RCTs
Robust Personalized Recommendation under Hidden Confounding in MNAR
-
Token prioritization lifts task accuracy in wireless systems
TONIC: Token-Centric Semantic Communication for Task-Oriented Wireless Systems
-
Expectation consistency suffices for calibration under covariate shift
Expectation Consistency Loss: Rethink Confidence Calibration under Covariate Shift
-
Vector quantization builds local calibration maps for multiclass models
Divide et Calibra: Multiclass Local Calibration via Vector Quantization
-
Pairwise data trains multimodal LLMs without full joint alignments
Multimodal LLMs under Pairwise Modalities
-
Causal constraints' power depends on the tasks they accompany
A Dialogue between Causal and Traditional Representation Learning: Toward Mutual Benefits in a Unified Formulation
-
Transformer mutation evolves improved approximate multipliers
Genetic Programming with Transformer-Based Mutation for Approximate Circuit Design
-
Diffusion link lets GPs condition on text or physics
Conditioning Gaussian Processes on Almost Anything
-
Unified model links peak timing and intensity in electricity forecasts
PeakFocus: Bridging Peak Localization and Intensity Regression via a Unified Multi-Scale Framework for Electricity Load Forecasting
-
Dynamic programming computes exact Banzhaf values for kNN
Efficient Banzhaf-Based Data Valuation for $k$-Nearest Neighbors Classification
-
Local boundary finds valid adjustment sets for causal effects
Local Covariate Selection for Average Causal Effect Estimation without Pretreatment and Causal Sufficiency Assumptions
-
Off-the-shelf persona vectors rival targeted sycophancy steering
Playing Devil's Advocate: Off-the-Shelf Persona Vectors Rival Targeted Steering for Sycophancy
-
SA error tails range from sub-Gaussian to near-Pareto with Markov noise
Concentration of General Stochastic Approximation Under Heavy-Tailed Markovian Noise
-
Landsat addition cuts TanDEM-X forest height RMSE by 13.5%
Hybrid Machine Learning Model for Forest Height Estimation from TanDEM-X and Landsat Data
-
Pontryagin framework optimizes policies for non-exponential discounts
Beyond the Bellman Recursion: A Pontryagin-Guided Framework for Non-Exponential Discounting
-
Latent GP and optimal transport track cell changes over time
Modeling Temporal scRNA-seq Data with Latent Gaussian Process and Optimal Transport
-
Flat minima enable non-vacuous bounds for transformers on sparse boolean tasks
A Sharper Picture of Generalization in Transformers
-
Routing imbalance in MoE stays fixed when expert parallelism scales
Diagnosing Overhead in Dispatch Operations: Cross-architecture Observatory
-
Point cloud sequences adapt simulators to new materials
Point Cloud Sequence Encoding for Material-conditioned Graph Network Simulators
-
Private mutual information selects better client groups for federated learning
Choose Wisely and Privately: Proactive Client Selection for Fair and Efficient Federated Learning
-
Proactive client choice cuts rounds and boosts fairness in federated learning
Choose Wisely and Privately: Proactive Client Selection for Fair and Efficient Federated Learning
-
TabPFN tops NIR regression calibration benchmarks
Tabular foundation models for robust calibration of near-infrared chemical sensing data
-
Conformal triage releases some event-positive cases at lower review
A Deployment Audit of Release-Side Risk in Conformal Triage under Prevalence Shift
-
DASH discovers strong hybrid attention for LLMs in 20 minutes on one GPU
DASH: Fast Differentiable Architecture Search for Hybrid Attention in Minutes on a Single GPU
-
Conformal method controls contamination in multi-LLM benchmarks
Provable Joint Decontamination for Benchmarking Multiple Large Language Models
-
Neural gate turns entity proxies into structural lag outputs
Discovering Entity-Conditioned Lag Heterogeneity: A Lag-Gated Neural Audit Framework for Panel Time Series
-
Oscillatory network scales to ImageNet with high efficiency
Winfree Oscillatory Neural Network
-
One program decodes bundles at 100% on four frozen embeddings
Sutra: Tensor-Op RNNs as a Compilation Target for Vector Symbolic Architectures
-
Sutra compiles VSA programs to tensor graphs with exact decoding
Sutra: Tensor-Op RNNs as a Compilation Target for Vector Symbolic Architectures
-
Unlearned models keep low calibration but lean on shortcuts
Calibration vs Decision Making: Revisiting the Reliability Paradox in Unlearned Language Models
-
Fighting game AIs learn how long to hold each move
For How Long Should We Be Punching? Learning Action Duration in Fighting Games
-
Agent surfaces novel threats in 15% of security incidents
GenAI-Driven Threat Detection with Microsoft Security Copilot
-
Agent finds hidden threats in 15% of security incidents
GenAI-Driven Threat Detection with Microsoft Security Copilot
-
Mechanism stratification lifts kinase inhibitor predictions
Training distribution determines the ceiling of drug-blind cancer sensitivity prediction
-
Optimal transport learns fMRI dictionaries across individual brain shapes
Learning fMRI activations dictionaries across individual geometries via optimal transport
-
Neighbor variance spots graph anomalies with no training
NeighborDiv: Training-free Zero-shot Generalist Graph Anomaly Detection via Neighbor Diversity
-
CIG reward unifies lifelong and episodic exploration signals
CIG: Exploration via Conditional Information Gain
-
Frequency regularization lifts attack transfer to closed MLLMs
Frequency-Domain Regularized Adversarial Alignment for Transferable Attacks against Closed-Source MLLMs