Scalable Sample-Level Causal Discovery in Event Sequences via Autoregressive Density Estimation
TRACE framework uses autoregressive density estimation for causal discovery in single event sequences.
TRACE framework uses autoregressive density estimation for causal discovery in single event sequences.
MetaDOAR meta-controller applies multi-agent reinforcement learning to large-scale cyber-network security games.
TRC² architecture enables LLMs to continually learn and adapt without catastrophic forgetting through specialized decoder design.
Framework combining ML and contextual stochastic optimization for transit network design under demand uncertainty.
AOI framework enabling LLM agents to improve from failed cloud diagnosis trajectories in SRE automation with safety constraints.
KV cache optimization using low-dimensional attention selection to reduce transformer memory with O(log N) key dimensions.
Study examining many-shot prompting for test-time LLM adaptation, analyzing reliability and limits of in-context learning scaling.
Active feature acquisition method for biomedical applications optimizing measurement selection under temporal and cost constraints.
Theoretical analysis proving attention sinks are functionally necessary in softmax transformers for certain tasks.
Research addressing multimodal model underperformance in context-aided forecasting via improved context quality assessment.
Machine learning method using hypergraph pre-training to improve atrial fibrillation prediction in stroke patients.
Interactive benchmark environment for synthesizing flat-foldable origamis, testing AI systems' planning and causal reasoning in physical domains.
AI-ECG system using foundation models for non-invasive hyperkalemia detection and handheld deployment in clinical settings.
Neural compression framework using SIREN auto-decoders for high-fidelity compression of multi-structural seismic velocity models.
Graph-based verifier for LLM task planning that identifies and corrects hallucinations and flaws in agent-generated plans.
Multimodal framework combining weather foundation model with satellite data for fine-grained solar irradiance forecasting.
Post-hoc model-agnostic explanation method using informative perturbation selection for uncertainty-aware interpretability of black-box ML models.
Convergence analysis of Muon optimizer under heavy-tailed noise for nonconvex optimization in large-scale deep neural network training.
Analysis showing wider beam search in LLMs can degrade output quality due to overestimation bias in noisy scorer outputs, with theoretical grounding.
Large-scale benchmark for AI agents combining partial observability, game-theoretic reasoning, and long-horizon planning in Pokemon battle environment.
Physics-informed neural networks and neural operators for simulating EUV electromagnetic wave diffraction in lithography mask applications.
Unsupervised machine learning approach for automated geometric parsing and classification of atomic octahedral networks in crystalline materials.
Theoretical analysis proving convergence bounds for Sequential Monte Carlo algorithm on multimodal distributions using soft decomposition.
Scalable statistical method for learning multi-axis Gaussian graphical models on biological network datasets with millions of cells.
Deep learning emulator for Earth system climate modeling that reduces computational cost while maintaining stability in multi-decadal projections.
Annotation-free method for reconstructing controllable 3D Gaussian splats of articulated objects from monocular video using flow derivatives.
Data synthesis engine using scene graphs to improve compositional generalization and semantic alignment in text-to-image generation models.
Legal analysis of copyright issues in generative AI training data collection and fair use doctrine across jurisdictions.
Inference framework for DNN estimators in generalized nonparametric models, enabling statistical inference on deep neural network predictions for categorical outcomes.
CHARM method calibrating reward models using Chatbot Arena scores to mitigate model preference bias, improving alignment of LLMs through RLHF.
PhysioOmni foundation model for multimodal physiological signals handling arbitrary missing modalities across EEG, ECG, EOG, EMG for healthcare and brain-computer interfaces.
NeuroSim V1.5 benchmarking software for compute-in-memory accelerators with non-ideal device and circuit modeling, addressing energy bottlenecks in AI hardware.
ALPCAH subspace learning method for heteroscedastic data with sample-wise varying noise, extending PCA for mixed-quality data dimensionality reduction.
Algorithm for unbiased low-rank matrix approximation minimizing expected Frobenius norm error while maintaining rank constraints and unbiased expectation.
BiomedSQL benchmark for text-to-SQL generation requiring scientific reasoning over biomedical knowledge bases, evaluating LLM capability for complex analytical tasks.
Clustering algorithm for trajectory data from mixture of unknown Markov chains with instance-dependent error bounds using spectral methods and injective Euclidean embeddings.
DP-Powered LLMs for privacy-preserving radiology report classification, enabling differential privacy in healthcare diagnosis and abnormality classification workflows.
TempCore benchmark analyzing whether video QA models genuinely require temporal frame selection, introducing Frame Selection Sensitivity metric for VLM diagnostic evaluation.
CardioComposer framework using differentiable geometry for controllable 3D cardiovascular anatomy generation from ellipsoidal primitives, balancing geometric control and realism.
Comparison of statistical and logic-based XAI techniques for interpreting ML security alerts in 5G intrusion detection systems, enabling actionable incident response.
ERGO framework for efficient high-resolution image processing in vision-language models using coarse-to-fine reasoning pipeline with two-stage visual token reduction.
Hilbert system recursively building formal proofs by combining informal LLM reasoning with Lean 4 verification, bridging gap between mathematical reasoning and formal proof generation.
Novel inverse problem for ptychographic phase retrieval without position knowledge, recovering scan positions jointly with images using variational inference for X-ray imaging.
Zephyrus framework combining foundation models for weather forecasting with LLM reasoning to enable language-based scientific workflows on meteorological datasets.
Security research on backdoor attacks in AI agent supply chains through poisoned interaction data collection, formalizing threat models for finetuned web browsing and tool-use agents.
Robotics research on learning from constrained demonstrations where expert interfaces limit optimal behavior demonstration, using techniques like kinesthetic teaching and sim-to-real transfer.
Game-theoretic model analyzing bias in meritocratic selection systems like admissions and hiring, examining how AI shapes perceived candidate value across socioeconomic groups.
Training-free diffusion model enabling layer-wise control in text-to-image generation through noise transplantation without fine-tuning or large datasets.
Deep Q-Network learns satellite weighting for CSI-free multi-satellite positioning in LEO constellations combined with weighted least squares estimation.
Joint audio-visual editing pipeline with video-to-audio generation model conditioning on source audio, target video, and text prompts for coherent edits.