All Articles
17020 articles total
Continual Learning with Elastic Regularization and Synthetic Replay for Federated MLLM Fine-Tuning
FedCMM is a framework for federated multimodal large language model fine-tuning that uses elastic regularization and synthetic replay to prevent catastrophic forgetting.
Toward Trustworthy Autonomous Science: A Two-Year Community Roadmap
An updated two-year community roadmap for trustworthy autonomous science emphasizes verification, reproducibility, and governance as critical for future progress.
GaitSpan: Growing Humanoid Locomotion from Walking to Running
GaitSpan is a framework that expands a basic walking policy into a continuous range of locomotion speeds (walking, jogging, running) for humanoid robots.
Never argue with your boss (2009)
An old Hacker News thread discussing interpersonal dynamics and management advice from 2009.
Learning to Discretize: Diffusion-Based Adaptive Mesh with Spectral Guidance
Proposes a two-stage diffusion framework to learn adaptive mesh discretization for neural partial differential equation (PDE) solvers.
Signal-Guided Optimization for Machine Unlearning
Introduces GSUO, a framework for machine unlearning that uses task-specific guidance signals to prevent over-unlearning or under-unlearning.
Gene Expression-Informed Jointly Controlled Generative Modeling for Precision Molecular Design
Presents JoPMol, a generative model for precision molecular design that integrates gene expression and molecular structure.
Evaluating Nonuniform Dependability Across Response Conditions: A Conditional Generalizability Framework Illustrated in Automated Essay Scoring
Discusses a framework for evaluating the dependability of automated scoring configurations in measurement-design settings.
Removable Defects: The Economics and Limits of Deliberate Deficiency
An economic and theoretical analysis of when it is profitable to design systems with deliberate, removable deficiencies.
Sparse Inter-Layer Dependencies of Transformer FFN Neurons
Introduces a training-free attribution method to identify sparse inter-layer dependencies in Transformer FFN neurons for better interpretability.
Mitigating The Effect of Class Imbalance in Data with Hierarchical and Dependable Structure
Proposes a Hierarchy-Aware RoBERTa framework to improve CWE vulnerability classification by incorporating structural information.
Are we Merging the Right Models? Impact of Expert Training Duration on Model Merging for LLMs
Studies how training duration of domain experts affects the quality of multi-task model merging for LLMs.
HPC-Enabled Video-based Coastal Wave Parameter Estimation Using V-JEPA and Deep Spatiotemporal Learning
An HPC-enabled deep learning framework for estimating coastal wave parameters from monocular coastal video.
Who's running all those tiny RPKI servers?
A community discussion on Hacker News regarding the operational infrastructure and entities responsible for maintaining the numerous small Resource Public Key Infrastructure (RPKI) servers.
BattVAE-GP: Generative Modeling of Long-Horizon Battery Degradation with Uncertainty Quantification
Introduction of BattVAE-GP, a hybrid physics-probabilistic learning framework for efficient, uncertainty-aware surrogate modeling of lithium-ion battery degradation trajectories.
Generalized Distribution-Free Semi-Supervised Learning with Risk Rewrite
A generalized distribution-free semi-supervised learning framework using risk rewriting to provide unbiased risk estimators for multiclass classification with reduced variance.
BAT-RM: A Boundary-Aware Transformer with Region-Aware Multi-Directional Mamba for Clinically Deployed Cervical Cancer Radiotherapy Auto-Contouring
BAT-RM, a boundary-aware Transformer and Mamba hybrid architecture, is deployed for cervical cancer radiotherapy auto-contouring, significantly reducing contouring time and improving consistency.
Scale-Aware Attention for Scarce Neural Data: An RG-Flow Transformer on Sleep-EDF EEG
Evaluation of an RG-Flow Transformer with renormalization-group inductive bias for EEG sleep staging, showing limited performance gains but improved interpretability of spectral exponents.
Graph-Constrained Policy Learning for Extreme Clinical Code Prediction
A graph-constrained traversal policy using language models to predict clinical ICD-10-CM codes, outperforming flat multi-label classification in extreme label spaces.
Exact and Certified Data Shapley for Weighted k-Nearest-Neighbor Regression and Soft-Label Prediction
Introduction of the first pseudo-polynomial-time exact algorithm for weighted KNN-regression Data Shapley, including a certified FPTAS and an open-source CPU-only library.