Multi-agent reinforcement learning approach for unmanned aircraft separation assurance under adversarial GPS degradation and spoofing.
Theoretical analysis of minimum-norm interpolation under 2-uniform convexity assumptions for understanding generalization in overparameterized neural networks.
Multi-agent traffic simulation framework using self-supervised world models to scale autonomous driving system testing with unlabeled sensor data.
Model-based reinforcement learning approach using Pontryagin methods and Hamiltonian actor-critic to address compounding model errors in long-horizon value estimation.
Mimosa: evolving multi-agent framework for autonomous scientific research that synthesizes and refines LLM-based agent workflows.
Transfer learning approach for Bayesian optimization applied to aircraft design problems.
Topological analysis of persistent homology for detecting phase transitions in spin models.
PolarQuant: post-training weight quantization method for LLM compression using Hadamard rotation and Gaussian optimization.
Analysis of modality gaps in multi-modal models like CLIP from robustness perspective.
Neural network approximation for inverse delay mapping in time-varying delay systems via operator learning.
Theoretical analysis of cyclic block coordinate optimization methods for variational inequalities.
Medical imaging segmentation using foundation models like MedSAM for brain tissue classification from MRI data.
GNN-based model for software vulnerability detection that offers better scalability than LLM approaches for code analysis tasks.
LiteCoST framework for document QA using chain-of-structured-thought and fine-tuned small language models for high accuracy and low latency.
Thiomi: large-scale multimodal dataset with 600k+ text annotations and 385k+ audio recordings across 10 African languages.
MemRerank framework distilling user purchase history into preference signals for personalized LLM-based shopping agent product reranking.
CNN and LightGBM surrogate models for real-time electromagnetic transient prediction in inverter-based microgrids.
SABLE framework for semantically-aware backdoor attacks in federated learning using realistic, in-distribution triggers.
Method for generating rigorous, human-interpretable explanations for tree ensemble model predictions.
Hardware-software framework for automatic task partitioning of deep reinforcement learning on Xilinx Versal ACAP.
Deep learning-based cryptanalytic techniques for differential fault attacks on lightweight stream ciphers in IoT.
Fine-tuning framework (AGFT) for improving zero-shot adversarial robustness of vision-language models while preserving alignment.
Multi-agent RL approach for cooperative AUV target tracking using diffusion models to address non-stationarity and coordination challenges.
Framework for high-quality dataset generation from closed-loop automotive data collection for ML model development.
Novel neural architecture (Metriplector) based on metriplectic field dynamics enabling gradient-free computation.
Combines model predictive path integral control with learning-based methods for gradient-free path following optimization.
Framework (PRoSFI) for generating verifiable step-by-step reasoning in LLMs using structured formal intermediaries and process rewards.
Benchmarks language models on child-scale datasets to understand data efficiency and linguistic knowledge emergence.
Statistical mechanics analysis of transformer parameter space for protein structure prediction with temperature-based sampling.
Data-driven framework for constructing LPV surrogate models with uncertainty quantification for nonlinear systems.
Uses small-scale language models to study multilingual language acquisition in children through computational modeling.
Agentic system for automated medical coding from clinical text using scalable, explainable approach that adapts to new codes.
Statistical learning approach for unbounded density ratio estimation and covariate shift adaptation without assuming bounded ratios.
mlr3mbo: modular R toolbox for Bayesian optimization supporting single/multi-objective, parallelization, and custom algorithm construction.
Learning-based method for designing Kazantzis-Kravaris/Luenberger observers for non-autonomous nonlinear systems using hypernetworks for input conditioning.
Reasoning-driven approach for generating synthetic multi-modal training data without manual prompts, addressing scarcity of specialized AI training datasets.
Methods for processing and analyzing petabyte-scale whole-brain 3D microscopy data from light-sheet fluorescence microscopy for neuroscience research.
DIAL proposes decoupling intent and action in Vision-Language-Action models via latent world modeling to improve decision-making and training stability in end-to-end robotic control.
GENIE method for editing Implicit Neural Representations via Gram-eigenmode deformations without retraining.
Hybrid machine learning framework for graduate admission prediction and university-program recommendation using 13,000 GradCafe records.
Method using epistemic uncertainty to identify unreliable explanations in post-hoc XAI methods, reducing explanation generation costs.
Deep learning framework combining segmentation and dual-mode compression for efficient wind turbine inspection imagery transfer.
Uncertainty quantification method for image segmentation using spatially-aware aggregation for out-of-distribution detection.
Proposes adaptive reasoning allocation during code generation for LLMs, addressing limitations of upfront thinking approaches in handling code complexity.
Statistical methods research on Oaxaca-Blinder decomposition showing reference group choice can reverse conclusions about group differences.
Early exiting predictive coding neural networks optimized for edge AI devices with resource constraints and privacy requirements.
GenOL framework for online learning with only concept names (name-only setup) enabling real-time adaptation to data distribution shifts in continual learning scenarios.
Introduces WEATHER-5K dataset and benchmarks physics-informed time-series forecasting models for global weather prediction.
Control-theoretic approach to reinforcement learning with convergence guarantees, new gradient theorem, and gradient ascent algorithm.
Information-theoretic analysis of transformer in-context learning on variable-order Markov chains with finite-sample accuracy bounds.