🚀 Introducing DiscoveryRSI: Recursively Learning to Discover Valuable Hypotheses
@LingYang_PU5K followers
55,712 views
17 Comments
Comments from the original post on X (Twitter)

@MANIC_PANIC777
@LingYang_PU It'd be super cool if you could actually build a hypergraph memory system with IA³ adapters, coupled with this . That way you could have a full autodidactic lifetime learning model.
1 like

@deeponailabs
Check how often random hypotheses match later papers, that sets your baseline
1 like

@b4lee1981
huge release 🔥

@vikneshmanohar
Wait how are you testing against GPT-6?

@samjo_00
recursive learning for scientific discovery is the right direction imo

@BreezeOg1
hypotheses that actually got published later 👀 ok that's cool

@thebasedcapital
the 6.3% number is cool but what about the other 93.7%. do the misses just get tossed or fed back as the bad ones?

@n_guseva67
Congrats Ling!

@ShardulAggarwal
Very interesting! Congrats

@callum_mtb
6.3% vs 2.5% is pretty wild if real

@Cattch11Tsot
thought this was hype until I read the paper, benchmark setup is actually solid

@RuthWinstanley
27B beating Opus on hypothesis generation is insane

@Titan1Beast
Hmmmmmmm

@FaijanK21499117
Ok the 2.3% baseline for claude opus makes sense given how conservative it is

@Ivrqen
filtering the 94% noise is where all the actual human time goes

@Tr_Boladoo
what paper dataset was used for validation?

@Allmight_1919
how do you define matching later published work though
