Sobre este pódcast
AI-generated podcast where hosts Hal Turing and Dr. Ada Shannon discuss the latest research papers and reports in machine learning, AI systems, and optimization. Featuring honest critical analysis, proper citations, and nerdy humor.
891 episodios · technology
AI Post Transformers
Último episodioPiper: Fixing GPU Utilization in MoE Training on Frontier
Sobre este pódcast
AI-generated podcast where hosts Hal Turing and Dr. Ada Shannon discuss the latest research papers and reports in machine learning, AI systems, and optimization. Featuring honest critical analysis, proper citations, and nerdy humor.
Episodios recientes
- MoM: Mixture-of-Memories Fixes Linear Attention Recallhace 6 días
- Photonic Interconnects and the Race to Cut Prefill Latency8 sept
- The Case for Learned Index Structures Revisited7 sept
- AI Coaching That Preserves Human Skill Development6 sept
- The Great Debate: Bounded Refinement vs Open-Ended Self-Improvement6 sept
- NVIDIA Ultra-CC Beats Top Human at IOI 20266 sept
- MegaTrain: Training 120B-Parameter Models on a Single GPU5 sept
- KV Cache Management Faces Its First Head-to-Head Test4 sept
- Language Models Learn to Fetch Only What They Need4 sept
- Disaggregating Prefill, Decode, Attention, and FFN for Agentic Inference4 sept
podlinkr.com/es/ai-post-transformers
Para un cartel o una diapositiva.