docere

5 ms·

1.
by docere 7mo ago
I agree that the necessity to design complex edge cases to find AI reasoning weaknesses indicates how far their capabilities have come. However, from a different point of view, failures of these types of edge cases which can be solved via &
2.
by docere 7mo ago
The issue with "adversarial" questions like the blood pressure one (which is open-sourced and published 1 year ago) is that they are eventually are ingested into model training data.
3.
by docere 7mo ago
Similar "broken" common-sense reasoning also occurs in medical edge-case reasoning ( https://www.nature.com/articles/s41598-025-22940-0 ); e.g. LLMs (o1) gets the following type of question wrong: A 4-year-old
4.
by docere 7mo ago
The child did not die, and suffered only minor injuries: https://abc7.com/post/california-teamsters-call-suspension-w... Under the same circumstances (kid suddenly emerging between two parked cars and running out onto
5.
MedEvalArena: Peer-judged LLM medical reasoning benchmark(danbernardo.substack.com)
1 points by docere 8mo ago|0 comments
6.
LLM Failure Modes in Medical QA Arising from Inflexible Reasoning(arxiv.org)
3 points by docere 2y ago|0 comments
7.
EEG-GPT(arxiv.org)
4 points by docere 3y ago|0 comments
8.
by docere 5y ago
You have a point. Medically induced coma such as with pentobarbital can “flatline” the brain, eliminating electrical fields. After weaning the pentobarbital patients have their memories intact.
9.
by docere 5y ago
People not uncommonly pass away while having brain activity recorded via EEG (source: I read EEGs in the US). I’m not sure why this article was published.
10.
by docere 7y ago
Thanks for sharing! It's very intriguing work. Is each node intended to model the local field potential of a given brain region? How 'large' or what scale are the individual nodes? (e.g. a few hundred neurons, a handful of ne
11.
by docere 7y ago
I’m a physician (epileptologist)... Started out comp sci and made the switch to medicine to help people. Currently doing programming/academic medicine roughly 50/50; the former has yet to provide any significant money or grants, b
12.
by docere 9y ago
Is this promising tech? EEG may be sensitive but not very specific.
13.
by docere 9y ago
Perhaps this has something to do with completion of myelination to the frontal lobes.