In case you were wondering why there hasn't been a new @BedsideRounds in a while ...

@Dr_Oubre @noahkaufmanmd Curious to hear your thoughts! It's incredibly popular here at Beth israel. Probably all of my residents currently use it, and I think a lot of them use it more frequently than up to date.

@saigarikapati19 @DrDatta_AIIMS Yes :) (though to get at your question, human reasoning is much weirder and constrained than I think most people generally imagine)

@Dr_Oubre @noahkaufmanmd Have you used Open Evidence?

@noahkaufmanmd @Dr_Oubre OpenEvidence does a pretty good job already at providing RAG CDS in the encounter itself. Chart ingestion into a usable context window at scale still isn't possible without a TON of processing. We'll get there, but I think it's still going to be a little longer.

@byrd_nick This is an elegant study (loved the preprint), but it's hard to 100% what to make of this without a human baseline for QA tasks, or how this relates to more generalized reasoning tasks.

RT @howardm19: I'm not really surprised...

RT @deedydas: New research shows that LLMs don't perform well on long context. Perfect needle-in-the-haystack scores are easy—attention me…

@reverendofdoubt My guess would be this funding cycle (so over the next couple of months). I haven't heard anything concrete either.

RT @OldSchoolBoston: At Dusk (Boston Common at Twilight) by Frederick Childe Hassam, 1885-86.

@andrewparsonsMD I love this piece!

@raj_mehta An interesting perspective! So like NLP extraction of symptoms to identify new nosologies that we wouldn't have thought of otherwise?

@iandarin Yes! What you're getting at is the idea of a machine nosology, which would be the counter force to this (ie linking of symptoms/findings directly to outcomes using large data sets that can be shown to empirically work, but isn't understandable by us)

@CoryRohlfsen @DavidDeutschOxf @ToKTeacher @perigean @techczech IMO that's 100% the case, especially when you take a longer view at the multiple nosologies we've had in human history

@harvardgidoc @uofk_kmsa Sure it was a great talk! What topics did you cover?

@reverendofdoubt I don't understand it either 🤣 I'm an accidental researcher... Still a clinician/educator at heart!

@reverendofdoubt (suspect your research is like mine, with little overhead other than compute and personnel)