SLAI T-Rex: Full-Parameter Post-training of the DeepSeek-V4 Family on Ascend SuperPOD Paper • 2607.20145 • Published Jul 22 • 77
Running Featured 163 Granite Docling 258M WebGPU 🐣 163 Convert document images to HTML with a single click
Confidence Is All You Need: Few-Shot RL Fine-Tuning of Language Models Paper • 2506.06395 • Published Jun 5, 2025 • 135
Grokking in the Wild: Data Augmentation for Real-World Multi-Hop Reasoning with Transformers Paper • 2504.20752 • Published Apr 29, 2025 • 97