ECoLAD: Efficient Automotive Time-Series Anomaly Detection
Discover ECoLAD, a deployment-focused evaluation method enhancing automotive time-series anomaly detection under real-world constraints.
Cloudless-Training: Boost Geo-Distributed ML Efficiency
Discover Cloudless-Training, a framework enhancing geo-distributed ML training with adaptive scheduling and reduced communication overhead.
Temporal & Semantic Rotary Encoding for Sequential Models
Explore SIREN-RoPE, a novel temporal and semantic rotary encoding method that enhances Transformer models for sequential data with improved accuracy and ef...
Clinical AI Evaluation Using Case-Specific Rubrics & LLMs
Explore case-specific rubrics for clinical AI evaluation, validating LLM-clinician agreement across 823 encounters to improve healthcare outcomes.
Can AI Close the Discovery-to-Application Gap? Minecraft Case Study
Explore how AI agents struggle to bridge discovery and application gaps using SciCrafter in Minecraft, revealing key challenges for future intelligence mod...