Performance Archaeology, Access Patterns, Memory Hierarchies, Optimization Forensics
School of Reward Hacks: Hacking harmless tasks generalizes to misaligned behavior in LLMs
arxiv.org·1d
DualSparse-MoE: Coordinating Tensor/Neuron-Level Sparsity with Expert Partition and Reconstruction
arxiv.org·57m
Google’s URL Context Grounding: Another Nail in RAG’s Coffin?
towardsdatascience.com·15h
Literature Review of the Effect of Quantum Computing on Cryptocurrencies using Blockchain Technology
arxiv.org·1d
LLM System Design and Model Selection
oreilly.com·18h
Loading...Loading more...