Machine learning models for metabolic liver disease prediction from EHR data, comparing LASSO, random forests, and neural networks.
LLM-based flight delay prediction integrating textual aeronautical data and aircraft trajectories for air traffic management.
Graph neural network architecture using selective state space modeling to address over-smoothing in deep GNNs via node-specific representation evolution.
Optimization of continuous attractor neural networks for brain-inspired path integration, reducing computational redundancy in navigation systems.
Vision-Language-Action model with active visual attention for robotic manipulation, extending from Markov to partially observable decision processes.
Analysis of flow-based diffusion models revealing two-stage behavior through oracle velocity fields, focusing on memorization-generalization dynamics.
Multi-agent RL framework for adaptive traffic signal control, replacing static controllers with learning-based optimization for complex traffic dynamics.
Multi-agent RL for graph-based coordination with bandwidth constraints, addressing what information agents should transmit under communication limits.
Analysis of self-reflection emergence in LLMs through RL post-training, using gradient attribution to explain distinct solution generation and revision capabilities.
Imitation learning framework for combinatorial optimization problems, examining how expert demonstrations affect policy learning in sequential decision problems.
FP8 low-precision quantization for LLM reinforcement learning, addressing memory and compute bottlenecks in rollout generation with engineering and algorithmic solutions.
Demonstrates layer pruning limitations for LLM reasoning tasks, showing pruned models lose algorithmic capabilities despite compression on classification tasks.
dnaHNet foundation model for genomic sequence learning with tokenizer-free design preserving biological motifs while handling long contexts efficiently.
Adaptive model selection framework for demand forecasting addressing horizon-induced degradation across heterogeneous inventory portfolios.
Physics-guided neural network for high-resolution air quality prediction incorporating topography and wind direction as critical factors.
SubQuad pipeline for adaptive immune repertoire analysis combining subquadratic retrieval with learned multimodal fusion for clinical clonotype detection.
Reinforcement-aware knowledge distillation method for distilling RL-trained reasoning LLMs into smaller models while preserving chain-of-thought capability.
Distributed prompt caching technique for accelerating local LLM inference on resource-constrained edge devices via inter-device state sharing.
Analyzes implicit regularization of Deep LDA objective for scale-invariant discriminative metric learning.
Theoretical analysis explaining Adam's empirical advantage over SGD through second-moment normalization using stopping-time/martingale analysis.
Enables exact gradient computation for spiking neural networks via differentiable ODE solving in JAX, supporting arbitrary neuron models.
Proposes prototypical exemplar condensation for memory-efficient continual learning, reducing stored samples per class from 20+ to single digits.
ALMAB-DC framework combines active learning, multi-armed bandits, and distributed computing for expensive black-box optimization.
Investigates mechanisms of introspective awareness in LLMs, where models detect injected steering vectors with minimal false positives.
Adapts KGE metric for non-stationary geoscientific systems in water management applications.
Analyzes distributional reinforcement learning with applications to healthcare, moving beyond expectation-based objectives for uncertain domains.
Proposes hierarchical SVG tokenization approach for improved scalable vector graphics modeling with LLMs via geometric-aware token design.
ALTO system for adaptive hyperparameter tuning and orchestration of LoRA fine-tuning jobs across heterogeneous multi-tenant environments.
Introduces Gated-SwinRMT, a vision transformer combining Swin attention with Manhattan-distance decay for improved spatial modeling.
Proposes CMRM, a framework for improving classification under label noise without privileged knowledge, using quantile-calibrated regularization.
Combines LLMs with Graph Neural Networks to enhance fMRI brain network analysis by leveraging LLM representations.
Bias-constrained diffusion schedules for PDE emulation with improved reconstruction error and efficient unrolled training.
Temporal transfer learning approach for traffic optimization using real-time driving advisories for connected vehicles.
Uses cognitively motivated grammar models as behavioral biometrics for authorship verification in digital forensics.
Social recommendation model using condition-guided approach to mitigate popularity bias in recommendation systems.
Survey of detection and characterization methods for coordinated online behavior including disinformation and social movements.
Studies extrinsic gender bias in Bangla pretrained language models across four NLP classification tasks.
Method for disentangling visual concepts in image generation to enable multi-aspect creative content generation.
Automatic self-supervised learning framework for social recommendations that learns auxiliary tasks without manual design.
Extends conformal prediction framework to hierarchical classification with constrained representation complexity.
Method for constraining sequential editing of LLMs to prevent knowledge degradation using editing anchor compression.
Shows XL MIMO wireless systems exhibit universal approximation properties similar to neural networks for OTA classification.
Sparse Bayesian learning method with space power prior for block sparse signal recovery with unknown patterns.
Framework for extreme super-resolution using autoregressive chain of scale states with preference alignment.
Agentic system for generating and validating synthetic image data to address data scarcity and label noise in vision tasks.
Novel variational quantum error correction approach using state distinguishability maximization for near-term quantum devices.
Evaluates LLM reasoning capabilities in social deduction game Avalon using Bayesian inference with graph-informed models.
Neural network method for solving two-stage stochastic unit commitment optimization problems in power systems.
Benchmark for evaluating how well multimodal models describe structural properties of time series data.
arXiv paper using deep learning to infer exoplanet geometry from transit light curves.