PretrainZero: Reinforcement Active Pretraining
Paper
• 2512.03442 • Published
• 48
Papers
-
-
UniQL: Unified Quantization and Low-rank Compression for Adaptive Edge LLMs
Paper • 2512.03383 • Published • 5 -
ToolOrchestra: Elevating Intelligence via Efficient Model and Tool Orchestration
Paper • 2511.21689 • Published • 125 -
Nemotron-Flash: Towards Latency-Optimal Hybrid Small Language Models
Paper • 2511.18890 • Published • 35 -
Every Token Counts: Generalizing 16M Ultra-Long Context in Large Language Models
Paper • 2511.23319 • Published • 24 -
WUSH: Near-Optimal Adaptive Transforms for LLM Quantization
Paper • 2512.00956 • Published • 23 -
CUDA-L2: Surpassing cuBLAS Performance for Matrix Multiplication through Reinforcement Learning
Paper • 2512.02551 • Published • 13 -
Stabilizing Reinforcement Learning with LLMs: Formulation and Practices
Paper • 2512.01374 • Published • 105 -
LightRAG: Simple and Fast Retrieval-Augmented Generation
Paper • 2410.05779 • Published • 28 -
MinerU2.5: A Decoupled Vision-Language Model for Efficient High-Resolution Document Parsing
Paper • 2509.22186 • Published • 146 -
End-to-End Test-Time Training for Long Context
Paper • 2512.23675 • Published • 24