Reliable voice interaction is essential in environments with limited internet connectivity and strong privacy. However, most existing voice assistants depend on cloud-based services, which leads to latency issues, dependency on internet access, and privacy vulnerabilities. This…

arXiv

Sep 25

Threat-Aware Energy-Efficient Deployment for Dynamic UAV Networks: A Multi-Agent RL Approach

Ensuring operational safety in threat-prone environments remains a critical challenge for multi-UAV networks serving as aerial base stations. This paper proposes an efficient framework to maximize global energy efficiency (EE) while promoting safe operation through threat-aware…

arXiv

Sep 25

On the Limits of Univariate Deep Learning for Significant Wave Height Forecasting

This study conducts a systematic hyperparameter search across five deep learning architectures, DLinear, LSTM, PatchTST, ResAttLstm, and Mamba2, and nine context lengths (1-168 h) for single-station significant wave height (Hs) forecasting on NDBC buoy 41009, followed by…

arXiv

Sep 25

HCPN-GCN: Scaling Hierarchical Prototype Networks with Cone Geometry for Continual Graph Learning

Continual Graph Learning (CGL) aims to incrementally learn from graph-structured data while preserving knowledge acquired from previous tasks. A major challenge in this setting is catastrophic forgetting, where learning new tasks degrades performance on previously learned ones.…

This week in Evals

This week

Get the morning digest

Today's trail

Nvidia CEO Jensen Huang defends global chip competition against Anthropic's Amodei

Pulse

Multi-step SWE: Can Coding Agents Keep Building on Their Own Code Over Long Horizons?

Scientist World: Grow a Research World for AI-Scientists

XStack: A Benchmark for ML Repo Migration

Benchmarks as Software: Terminal-Bench and the Case for Continuous Benchmarks

T3R: Training-Free Two-Stage Token Refinement Towards Efficient and Robust VLA Models

Top picks

LUMO (Lightweight Unified Multilingual Orchestrator): A Privacy Preserving Offline Voice Assistant

Threat-Aware Energy-Efficient Deployment for Dynamic UAV Networks: A Multi-Agent RL Approach

On the Limits of Univariate Deep Learning for Significant Wave Height Forecasting

HCPN-GCN: Scaling Hierarchical Prototype Networks with Cone Geometry for Continual Graph Learning

Today's trail

This week in Evals

Multi-step SWE: Can Coding Agents Keep Building on Their Own Code Over Long Horizons?

Scientist World: Grow a Research World for AI-Scientists

XStack: A Benchmark for ML Repo Migration

Benchmarks as Software: Terminal-Bench and the Case for Continuous Benchmarks

T3R: Training-Free Two-Stage Token Refinement Towards Efficient and Robust VLA Models

Most active this week

Rising tools

Nvidia CEO Jensen Huang defends global chip competition against Anthropic's Amodei

LUMO (Lightweight Unified Multilingual Orchestrator): A Privacy Preserving Offline Voice Assistant

Threat-Aware Energy-Efficient Deployment for Dynamic UAV Networks: A Multi-Agent RL Approach

On the Limits of Univariate Deep Learning for Significant Wave Height Forecasting

HCPN-GCN: Scaling Hierarchical Prototype Networks with Cone Geometry for Continual Graph Learning