Meta-Learning & Meta-RL: DeepMind’s Adaptive Agent Path
Explore meta-learning and meta-reinforcement learning techniques driving DeepMind's Adaptive Agent for rapid AI adaptation and improved task performance.
Deep RL for Finite-State Controllers in POMDPs
Discover how deep reinforcement learning enhances finite-state controllers for robust decision-making in POMDPs and hidden-model POMDPs.
ReasonMark: Semantic Watermarking for Large Reasoning Models
Discover ReasonMark, a novel semantic watermarking method that preserves reasoning integrity in large language models with minimal latency.
DR-LoRA: Adaptive Fine-Tuning for Mixture-of-Experts Models
Discover DR-LoRA, a dynamic rank LoRA method that optimizes fine-tuning of Mixture-of-Experts models for improved AI performance and efficiency.
EHRStruct: Benchmarking LLMs on Structured EHR Tasks
Discover EHRStruct, a benchmark framework to evaluate large language models on structured electronic health record tasks with 11 clinical challenges.