Universal Transformers Need Memory: Depth-State Trade-Offs in Adaptive Recursivearxiv.org·1 pts·che_shr_cat·0
Embedded Universal Predictive Intelligence: a coherent framework for multi-agentarxiviq.substack.com·1 pts·che_shr_cat·0
NeurIPS 2025 Best Papers in Comics: From Artificial Hivemind to 1000-Layer RLgonzoml.substack.com·3 pts·che_shr_cat·0
Visualizing Research: How I Use Gemini 3.0 to Turn Papers into Comicsgonzoml.substack.com·1 pts·che_shr_cat·0
AlphaResearch: Accelerating New Algorithm Discovery with Language Modelsarxiviq.substack.com·1 pts·che_shr_cat·0
LeJEPA: Provable and Scalable Self-Supervised Learning Without the Heuristicsarxiviq.substack.com·2 pts·che_shr_cat·1
CaT Replaces CoT-SC / Compute as Teacher: Turning Inference Compute Intoarxiviq.substack.com·2 pts·che_shr_cat·0
Tiny Recursive Model (TRM) vs. Hierarchical Reasoning Model (HRM)gonzoml.substack.com·2 pts·che_shr_cat·0
Evolution Strategies at Scale: LLM Fine-Tuning Beyond Reinforcement Learningarxiviq.substack.com·2 pts·che_shr_cat·0
Autoreview: The Dragon Hatchling – The Missing Link Between the Transformer andarxiviq.substack.com·2 pts·che_shr_cat·0
LLM-JEPA: Large Language Models Meet Joint Embedding Predictive Architecturesarxiviq.substack.com·1 pts·che_shr_cat·0
Review: SpikingBrain Technical Spiking Brain-Inspired Large Modelsarxiviq.substack.com·2 pts·che_shr_cat·0
Canaries in the Coal Mine? Six Facts about the Recent Employment Effects of AIarxiviq.substack.com·1 pts·che_shr_cat·0
DeepConf: Scaling LLM reasoning with confidence, not just computearxiviq.substack.com·98 pts·che_shr_cat·35
Speed Always Wins: A Survey on Efficient Architectures for Large Language Modelsarxiviq.substack.com·3 pts·che_shr_cat·0