Aurametrix, Inc. ⚙️🦠🧬

3.4K posts

Aurametrix, Inc. ⚙️🦠🧬 banner
Aurametrix, Inc. ⚙️🦠🧬

Aurametrix, Inc. ⚙️🦠🧬

@Aurametrix

Connect the dots to better health! Aurametrix helps you identify the subtle cause-effect relationships that contribute to how you feel--any time, any day.

Silicon Valley, CA Katılım Şubat 2009
22.1K Takip Edilen21.5K Takipçiler
Aurametrix, Inc. ⚙️🦠🧬 retweetledi
Timothy Gowers @wtgowers
Timothy Gowers @wtgowers@wtgowers·
the time it would have taken me would probably have been of order of magnitude an hour (an estimate that comes with quite wide error bars). So it looks as though we have entered the brief but enjoyable era where our research is greatly sped up by AI but AI still needs us. 3/3
English
18
35
1.1K
101.3K
Aurametrix, Inc. ⚙️🦠🧬 retweetledi
Deedy
Deedy@deedydas·
"GPT-5 moves from human-comparable to above human-expert performance" GPT-5 outperforms licensed human experts by 25-30% and achieves SOTA results on the US medical licensing exam and the MedQA benchmark. I sound like a broken record, but AI models are better than most doctors.
Deedy tweet media
English
94
160
1.3K
219K
Aurametrix, Inc. ⚙️🦠🧬 retweetledi
Ethan Mollick
Ethan Mollick@emollick·
GPT-4o was below the level of medical professionals on medical reasoning benchmarks GPT-5 (apparently Thinking medium) now far exceeds them. (Usual benchmark caveats apply)
Ethan Mollick tweet media
elvis@omarsar0

GPT-5 on Multimodal Medical Reasoning On MedXpertQA MM, GPT-5 improves reasoning and understanding scores by +29.62% and +36.18% over GPT-4o. It surpasses pre-licensed human experts by +24.23% in reasoning and +29.40% in understanding.

English
18
107
828
105.4K
Aurametrix, Inc. ⚙️🦠🧬 retweetledi
Arena.ai
Arena.ai@arena·
In WebDev, Claude Opus 4.1 is sitting at #2, tied with Gemini-2.5-Pro. See the WebDev Arena rankings: lmarena.ai/leaderboard/we…
Arena.ai tweet media
English
4
3
93
13.3K
Aurametrix, Inc. ⚙️🦠🧬 retweetledi
Daniel Kang
Daniel Kang@ddkang·
The prevailing wisdom is that compute is the most important factor for frontier AI training. We think this is wrong: data is the most costly and important component of AI training. We collected estimates of revenue for major data labeling companies and compared them with the marginal compute cost for training top models in 2024. Our estimates show that data labeling is ~3x higher than the marginal training compute. 1/8
Daniel Kang tweet media
English
51
163
1.2K
154.2K
Aurametrix, Inc. ⚙️🦠🧬 retweetledi
Peter Tierney
Peter Tierney@drpetertierney·
WEARABLE TERNARY PLOT First try at plotting where certain wearables sit. If a product sits in the centre of the triangle, it could be considered an "all-rounder". If a product sits in the corner, it is a specialist at that particular metric.
Peter Tierney tweet media
English
1
9
37
5.2K