Anonymous
8/3/2025, 1:48:03 AM
No.106121431
>>106121398
>>106121322
I mean there's more that can be done here. Companies up till now just haven't really prioritized it. You can certainly tune and more probably use RL to make a model slop less. Even LeCun suggested that RL can be used for adjusting the world model, even if it sucks in terms of efficiency.
>>106121322
I mean there's more that can be done here. Companies up till now just haven't really prioritized it. You can certainly tune and more probably use RL to make a model slop less. Even LeCun suggested that RL can be used for adjusting the world model, even if it sucks in terms of efficiency.