[HN Gopher] New Meta FAIR Research and Models
       ___________________________________________________________________
        
       New Meta FAIR Research and Models
        
       Author : ilaksh
       Score  : 42 points
       Date   : 2024-12-13 21:07 UTC (1 hours ago)
        
 (HTM) web link (ai.meta.com)
 (TXT) w3m dump (ai.meta.com)
        
       | cube2222 wrote:
       | There's honestly so much interesting stuff here, esp. the llm-
       | related things - large concept models (operating on and
       | predicting concepts, not tokens), dynamic byte latent
       | transformers (byte-level alternative to standard tokenization),
       | sparse memory layers (successfully scaling key-value memory
       | layers without an increase in computational requirements).
       | 
       | Here they are presented as separate things, each of which
       | apparently improves quality / efficiency. I wonder what the
       | quality / efficiency increase is of all those methods put
       | together? Maybe that's what Llama 4 will be?
       | 
       | This looks like a lot of innovation is happening at Meta in those
       | areas, really cool!
        
       | airstrike wrote:
       | [delayed]
        
       | pkkkzip wrote:
       | meta has certainly redeemed itself and helping AI become moat-
       | free
        
       | vouaobrasil wrote:
       | > Today, Meta FAIR is releasing several new research artifacts
       | that highlight our recent innovations in developing agents,
       | robustness and safety, and architectures that facilitate machine
       | learning.
       | 
       | I sincerely doubt Meta cares about safety. It only cares about
       | the appearance of safety and complying with the law while
       | effectively skirting the law in principle.
        
       ___________________________________________________________________
       (page generated 2024-12-13 23:00 UTC)