29d ago
Learning from Synthetic Data via Provenance-Based Input Gradient Guidance
Learning from synthetic data using provenance-based input gradient guidance to teach models discriminative regions.
29d ago
Inversion-Free Natural Gradient Descent on Riemannian Manifolds
Inversion-free stochastic natural gradient method for optimization on Riemannian manifolds with implicit parameter constraints.
29d ago
Learning Contractive Integral Operators with Fredholm Integral Neural Operators
Fredholm Integral Neural Operators framework for learning non-expansive integral operators with universal approximation proofs.
29d ago
SkillRT: Compiling Skills for Efficient Execution Everywhere
SkillRT: Compiler-inspired system treating LLM agent skills as code for consistent, efficient execution across diverse platforms.
29d ago
Characterization of Gaussian Universality Breakdown in High-Dimensional Empirical Risk Minimization
Theoretical characterization of Gaussian universality breakdown in high-dimensional convex empirical risk minimization under non-Gaussian data.
29d ago
The Compression Gap: Why Discrete Tokenization Limits Vision-Language-Action Model Scaling
Analysis of why discrete tokenization limits Vision-Language-Action model scaling; introduces Compression Gap information-theoretic principle.
29d ago
Learning the Signature of Memorization in Autoregressive Language Models
First transferable learned membership inference attack on fine-tuned LLMs using unlimited labeled data from fine-tuning process itself.
29d ago
PR3DICTR: A modular AI framework for medical 3D image-based detection and outcome prediction
PR3DICTR: PyTorch/MONAI-based open framework for 3D medical image classification and standardized deep learning model training.
29d ago
A Tsetlin Machine-driven Intrusion Detection System for Next-Generation IoMT Security
Tsetlin Machine-based intrusion detection system for IoMT networks addressing cybersecurity vulnerabilities in medical device systems.
29d ago
Efficient Causal Graph Discovery Using Large Language Models
LLM framework for causal graph discovery using breadth-first search queries, reducing complexity from quadratic to linear query requirements.
29d ago
Output-Constrained Decision Trees
Methods for training constrained regression trees incorporating domain-specific output constraints using mixed-integer programming and other approaches.
29d ago
Supplementary Materials to Graph Convolutional Branch and Bound
Integration of neural networks into combinatorial optimization for NP-hard problems, learning heuristics and optimality scores via graph convolutional networks.
29d ago
Amortized Inference of Causal Models via Conditional Fixed-Point Iterations
Amortized inference framework training single model to predict causal mechanisms across multiple datasets for out-of-distribution generalization.
29d ago
Distributional Statistics Restore Training Data Auditability in One-step Distilled Diffusion Models
Method for detecting unauthorized training data in one-step distilled diffusion models using distributional statistics instead of memorization detection.
29d ago
Zero-shot Concept Bottleneck Models
Zero-shot concept bottleneck models enabling interpretable predictions without target task training by leveraging pre-trained vision-language models.
29d ago
A Unified Approach to Analysis and Design of Denoising Markov Models
Mathematical framework unifying diffusion and flow-based generative models as denoising Markov processes with rigorous theoretical foundation.
29d ago
Accelerated Learning with Linear Temporal Logic using Differentiable Simulation
Integration of linear temporal logic specifications into RL using differentiable simulation for safe, correct-by-construction controller synthesis.
29d ago
Beyond Noisy-TVs: Noise-Robust Exploration Via Learning Progress Monitoring
Noise-robust exploration method for RL using learning progress monitoring to escape unlearnable noise sources with improved sample efficiency.
29d ago
ARMOR: High-Performance Semi-Structured Pruning via Adaptive Matrix Factorization
ARMOR: one-shot post-training pruning algorithm for LLMs achieving 2:4 semi-structured sparsity with minimal performance degradation for efficient deployment.
29d ago
High-probability Convergence Guarantees of Decentralized SGD
Theoretical analysis of convergence guarantees for decentralized stochastic gradient descent with high-probability bounds and reduced assumptions.
29d ago
Local Reinforcement Learning with Action-Conditioned Root Mean Squared Q-Functions
Extension of Forward-Forward algorithm to reinforcement learning with action-conditioned Q-functions, replacing backpropagation with local learning.
29d ago
f-INE: A Hypothesis Testing Framework for Estimating Influence under Training Randomness
f-INE framework for stable influence estimation under training randomness, addressing instability in sample-level impact estimation for data curation.
29d ago
Diffusion Models as Dataset Distillation Priors
Dataset distillation method leveraging diffusion models as priors to synthesize compact, representative datasets with improved diversity and generalization.
29d ago
Towards best practices in low-dimensional semi-supervised latent Bayesian optimization for the design of antimicrobial peptides
Deep learning approach for antimicrobial peptide design using semi-supervised latent Bayesian optimization with improved interpretability.
29d ago
Steering Autoregressive Music Generation with Recursive Feature Machines
MusicRFM framework enabling fine-grained control over pre-trained music generation models by steering internal activations via Recursive Feature Machines.
29d ago
Fast and Robust Simulation-Based Inference With Optimization Monte Carlo
Bayesian parameter inference method for complex stochastic simulators using differentiable approaches to reduce simulation costs in high-dimensional spaces.
29d ago
Goal-Driven Reward by Video Diffusion Models for Reinforcement Learning
Goal-driven reward signals from pretrained video diffusion models for reinforcement learning agent training.
29d ago
Pushing the Limits of Distillation-Based Continual Learning via Classifier-Proximal Lightweight Plugins
Distillation-based continual learning with classifier-proximal plugins addressing stability-plasticity tradeoff.
29d ago
Resting Neurons, Active Insights: Robustify Activation Sparsity for Large Language Models
Method to robustify activation sparsity in LLMs by addressing representational instability during inference acceleration.
29d ago
Community-Based Early-Stage Chronic Kidney Disease Screening using Explainable Machine Learning for Low-Resource Settings
Machine learning approach for early chronic kidney disease screening in low-resource settings using explainable models.
29d ago
Forget Many, Forget Right: Scalable and Precise Concept Unlearning in Diffusion Models
Scalable multi-concept unlearning in text-to-image diffusion models addressing weight conflicts and collateral damage.
29d ago
On the Extreme Variance of Certified Local Robustness Across Model Seeds
Analysis of extreme variance in certified robustness verification across neural network model seeds.
29d ago
Textual Equilibrium Propagation for Deep Compound AI Systems
Textual Equilibrium Propagation for optimizing compound AI systems with multiple modules in long-horizon agentic workflows.
29d ago
Early Classification of Time Series in Non-Stationary Cost Regimes
Early classification of time series in non-stationary environments with uncertain and time-varying decision costs.
29d ago
ChronoSpike: An Adaptive Spiking Graph Neural Network for Dynamic Graphs
ChronoSpike: Adaptive spiking graph neural network for dynamic graph representation learning with event-driven efficiency.
29d ago
When RL Meets Adaptive Speculative Training: A Unified Training-Serving System
Unified training-serving system combining RL with adaptive speculative decoding for accelerated LLM inference.
29d ago
Infusion: Shaping Model Behavior by Editing Training Data via Influence Functions
Infusion: Framework using influence functions to craft training data perturbations that induce targeted model behavior changes.
29d ago
Equivariant Evidential Deep Learning for Interatomic Potentials
Uncertainty quantification for machine learning interatomic potentials using evidential deep learning.
29d ago
Low-Dimensional and Transversely Curved Optimization Dynamics in Grokking
Geometric analysis of optimization dynamics in transformers trained on modular arithmetic revealing low-dimensional subspaces.
29d ago
Early-Warning Signals of Grokking via Loss-Landscape Geometry
Study on early-warning signals of grokking via loss-landscape geometry on SCAN and Dyck-1 benchmarks.
29d ago
CeRA: Overcoming the Linear Ceiling of Low-Rank Adaptation via Capacity Expansion
CeRA: Parameter-efficient fine-tuning method extending LoRA with non-linear capacity expansion via gating and dropout.
29d ago
Learning Physical Operators using Neural Operators
Physics-informed neural operators for solving PDEs with improved generalization beyond training distributions.
29d ago
SafeSci: Safety Evaluation of Large Language Models in Science Domains and Beyond
SafeSci: Framework for evaluating safety of large language models in scientific domains with comprehensive benchmarks.
29d ago
CRISP: Compressed Reasoning via Iterative Self-Policy Distillation
CRISP: Method for teaching LLMs to reason more concisely via self-distillation with 'be concise' conditioning.
29d ago
Stock Market Prediction Using Node Transformer Architecture Integrated with BERT Sentiment Analysis
Stock market prediction using Node Transformer and BERT sentiment analysis for financial forecasting.
29d ago
Pretrained Video Models as Differentiable Physics Simulators for Urban Wind Flows
WinDiNet uses pretrained video diffusion model as differentiable physics simulator for urban wind flow prediction, replacing expensive CFD simulations.
29d ago
$\lambda$-GELU: Learning Gating Hardness for Controlled ReLU-ization in Deep Networks
λ-GELU parameterized gating function enabling controlled ReLU conversion while maintaining smooth activation properties for deployment.
29d ago
ERPO: Token-Level Entropy-Regulated Policy Optimization for Large Reasoning Models
ERPO method for token-level credit assignment in LLM reasoning models, addressing entropy collapse in GRPO through information heterogeneity.
29d ago
Temporal Credit Is Free
Recurrent network training without Jacobian propagation using hidden state temporal credit. Studies gradient normalization and online adaptation.
29d ago
The Spectral Edge Thesis: A Mathematical Framework for Intra-Signal Phase Transitions in Neural Network Training
Mathematical framework explaining phase transitions in neural network training via spectral gap of parameter update Gram matrices. Grokking and capability gains analysis.