O alinhamento de segurança deve ser feito em mais do que apenas alguns tokens de profundidade (Pa...
36:15
2024 12 24 15 43 12
53:02
Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters (Paper)
31:22
The Trillion Dollar Equation
46:22
It's Not About Scale, It's About Abstraction
37:17
Leave No Context Behind: Efficient Infinite Context Transformers with Infini-attention
32:44
The Closest We’ve Come to a Theory of Everything
17:38
The moment we stopped understanding AI [AlexNet]
49:45