Sean Ward retweeted
Our VibeCodeBench evaluations affirm what @Anthropic just announced: Claude Sonnet 4 excels at autonomous multi-feature development. We've seen codebase navigation errors drop from 20% to near zero and strategic refactoring that saves ~500k tokens on multi stage, complex tasks. Proud to power Maestro with this breakthrough.
Introducing the next generation: Claude Opus 4 and Claude Sonnet 4.
Claude Opus 4 is our most powerful model yet, and the world’s best coding model.
Claude Sonnet 4 is a significant upgrade from its predecessor, delivering superior coding and reasoning.
ALT A benchmarking table titled Claude 4 benchmarks comparing performance metrics across various capabilities including coding, reasoning, tool use, multilingual Q&A, visual reasoning, and mathematics.
Sean Ward retweeted
Hottest week for London AI so far 🔥
Dev Day yesterday and AI Tinkerers tonight!
london.aitinkerers.org/p/ai-…
@monzo @tortus_AI @_lucas_godfrey @QuotientAI @samshapley @lukeharries @stephenbtl
Sean Ward retweeted
So it turns out that SARS-CoV-2 RNA is remarkably stable in saliva samples, whether stored frozen then tested or just left sitting around in warmer conditions...
Sean Ward retweeted
Very excited to share our new preprint "#SARSCoV2 viral load dynamics, duration of viral shedding and infectiousness: a living systematic review and meta-analysis" #IDTwitter
A huge team effort screening and synthesising the accumulating evidence. 1/6
medrxiv.org/content/10.1101/…
Sean Ward retweeted
The $5 mill XPRIZE RAPID COVID TESTING starts now. 🙌 The mission: develop COVID-19 testing solutions that are frequent, fast, cheap and easy, so the world can safely move forward. 🧪 🌎
REGISTER NOW 👉 bit.ly/307AeQJ
#covidtesting #takeactionnow #prizelaunch
Sean Ward retweeted
Really excited to share our new preprint modelling the effectiveness and surveying the burden of 9 different non-pharmaceutical interventions against #COVID19 transmission, using data collected across 41 different countries.
Full Preprint: doi.org/10.1101/2020.05.28.2…
1/8
Sean Ward retweeted
Can tracing alone control COVID?
Our model: it's possible if we >double efficacy by changing how we trace & use digital apps.
Thanks to @willbradbio, @EthanAlley, @jhhhuggins, & epidemiologist @alun_l !
doi.org/10.1101/2020.05.06.2…
1/n
Sean Ward retweeted
Let's say 1 person who's convinced of misinformation, on average, convinces 3 others. This means R0 = 3.
But! If you teach just 67% of people basic epidemiology/statistics, you can get R < 1, containing the info-demic!
This is called "nerd immunity"
Sean Ward retweeted
To help with scalable testing, @jgooten and I, in collaboration with @zhangf, are releasing STOPCovid, a single-reaction CRISPR-based COVID-19 detection protocol. We are sharing whitepaper, data, and requests for reagents on STOPCovid.science. 1/4
Sean Ward retweeted
New challenge: alternatives to swabs for sampling wanted and novel solutions to increase #Covid-19 #Testingmethods2020 across the UK - testingmethods.crowdicity.co…
@BIA_UK @DHSCgovuk @BIVDA @RCPath
Sean Ward retweeted
I seem to have launched another magazine... to share stories from the VOYAGERS.io community of good people helping each other. It's at medium.com/@VOYAGERS. I wrote about what running WIRED taught me about building a purpose-led community. is.gd/28lZ2Z
Sean Ward retweeted
US can get to millions of tests/day & come back to work safely (written w/ @zhangf @jrkelly & @JShendure): (1) Mass screening for work should be treated differently than clinical testing (2) Will need different regulation & logistics, tech is largely here
technologyreview.com/2020/04…
Sean Ward retweeted
Our paper about glowing plants is out in Nature Biotechnology today — but first, watch this:
1/5
Sean Ward retweeted
Important new study:
Analysis by @PHanlon17, @JonMinton and co estimates more than 10 years of life are lost for each UK Covid death on average, and much more in some cases.
More evidence against "they were going to die soon anyway" line.
Full paper: wellcomeopenresearch.org/art…
Sean Ward retweeted
Very interesting new study show saliva is more sensitive for #SARSCoV2 detection in #COVID19 patients than nasopharyngeal swabs
medrxiv.org/content/10.1101/…
This is great news for increasing testing capability as it removes the bottleneck of supply chain issue of swabs.
Sean Ward retweeted
Here is the link to the pre-pub of the work that John Bell cited in last week's government webinar on antibody testing - published today medrxiv.org/content/10.1101/…
Sean Ward retweeted
THREAD: The government has worked to improve testing capacity, including supporting the validation of synthetic swabs that can be manufactured at scale and increasing payment to labs running high throughput tests for COVID-19. There is more we can do to increase access. 1/n
Sean Ward retweeted
Great resource at @FINDdx on the test accuracy of serological tests for #Covid_19. Note confidence intervals on specificity...
finddx.shinyapps.io/COVID19D…
Then calculate #FalsePositiveRate for the expected low prevalence here:
vassarstats.net/
Then get sober.
Sean Ward retweeted
Arguing we should relax social distancing now like arguing the parachute slowed my fall so I can take it off now crankyuncle.com/critical-thi… @greghamra
