a_bonobo

12 ms·

121.
by a_bonobo 1y ago
Validation of biological labels easily takes years - in the OP's example it was a 'lucky' (huge!) coincidence that somebody already had spent years on one of the predicted proteins' labels. Nobody is going to stake 3-5
122.
by a_bonobo 1y ago
I mean, if you can solve 9% of Github issues automatically that's a fairly huge load of work you can automate. Then again you'd have to manually identify which 9% of issues.
123.
by a_bonobo 1y ago
I find the Konwinski Prize to be very interesting in this context. 1 million dollars to whoever's open source LLM solves >90% of a set of novel Github issues. https://www.kaggle.com/competitions/konwinski-prize&
124.
by a_bonobo 1y ago
Thanks for you answer! I agree, yours looks far easier to use - it'll be far easier to call your cmd client than Python package
125.
by a_bonobo 1y ago
How does this differ from langchain's tool calling?
126.
by a_bonobo 1y ago
I think this is where it's going to end up. The masses get the cheap AI education. The elite get the expensive, small class, analog education. There won't be a middle class of education, as in the current system - too expensive fo
127.
by a_bonobo 1y ago
Yes!!! My favorite recent think pieces have been on LLMs as a 'normal' technology. Nothing revolutionary, no AGI, just a normal tool. An accountant knows how to do math but they use the math in Excel to do it for them. They 100%
128.
by a_bonobo 1y ago
Anthropic has a nice press release that summarises it in simpler terms: https://www.anthropic.com/research/reasoning-models-dont-say...
129.
by a_bonobo 1y ago
Could the exclusion of CoT that be because of this recent Anthropic paper? https://assets.anthropic.com/m/71876fabef0f0ed4/original/rea... >We evaluate CoT faithfulness of state-of-the-art reasoning models
130.
by a_bonobo 1y ago
'every extension is also an amputation' - Marshal McLuhan (since Neil Stephenson's recent essay brought that quote up)
131.
by a_bonobo 1y ago
It's interesting that the article does not mention US-based Illumina. Illumina and BGI are the two biggest DNA sequencing companies in the world right now, and Illumina has been struggling immensely after China singled it out for reper
132.
by a_bonobo 1y ago
This happens again and again in research - I'm just reminded of the stem cell scandal around Obakata. Before the fraud was uncovered: dozens of senior researchers supporting the research, using the glory for their own gains. After the
133.
by a_bonobo 1y ago
> putting many protein-folding academics out of jobs. Has this actually happened? I'd wager it's more like in OP's article, AI as a helper that speeds up research, enabling bigger or different research questions. Reliable
134.
by a_bonobo 1y ago
Yes, human coders are predictably 'bad' - it scales roughly with the complexity of the task and the years of their experience, along with some randomness ('brain farts'). LLMs are completely unpredictably 'bad'
135.
by a_bonobo 1y ago
As a biologist all the field wants is p < 0.05. What it actually means is unnecessary. It's a hurdle to pass to have another paper on your CV.
136.
by a_bonobo 1y ago
Given that the highest-earning income tax bracket's tax rate in the US fell from ~70% to 37% since the 70s, one could argue that VC money used to be taxpayer's money. Nice visualisation here: https://www.fidelity.com&#x
137.
by a_bonobo 1y ago
I think this is the reason why articles about recent research in 99% of cases don't cite the research itself past something vague like 'Researchers from Harvard University have discovered..'. It's maddening when you then
138.
by a_bonobo 1y ago
At the same time, NIH just announced that all grants involving foreign researcher are shut. https://grants.nih.gov/grants/guide/notice-files/NOT-OD-25-1... >Effective with the date of this notice and until
139.
by a_bonobo 1y ago
I've recently had this problem where the important information (number of study participants, and how many were filtered out during which step) were only encoded in figures , not in the text. Maddening.
140.
by a_bonobo 1y ago
I'm already seeing this in the life-sciences. Illumina, who used to own the game of DNA sequencing, has been laying people off left and right. All my contacts within the company are laid off. Illumina is a special edge case as they&#x2
141.
by a_bonobo 1y ago
My guess is that once they're entrenched firmly enough, OpenAI will ask for money. First sell the free thing then once people rely on it enough, start charging.
142.
by a_bonobo 1y ago
Anthropic will likely go a similar way - what moat, what funding source does Anthropic have that OpenAI does not?
143.
by a_bonobo 1y ago
It's kind of like Russians throwing their oligarchs out of windows and then saying it's suicide: everybody knows who did it, and there won't be consequences. That in itself sends a signal: 'we can do this, what are you g
144.
by a_bonobo 1y ago
I agree, I'd even go so far and say it's one of the best places on the internet to discuss 'serious' books (within all the rampant troll posts). Book discussions on reddit are far too positive when it comes to terrible b
145.
by a_bonobo 2y ago
All decent people I know have deleted their Twitter accounts - the kind of people you now see on twitter in the mentions are... not good people.
146.
by a_bonobo 2y ago
Before Covid I thought I'd chuck a few hundred dollars into a trading app to understand that 'game' a little bit better. I chose a few companies I knew from work and some biotechs that had fun ticker names. One company had &#
147.
by a_bonobo 2y ago
The article makes it about IP theft but I'm sure this is to deter scammers, or at least signal an intent to do something about all the spammers and scammers using those APIs to build their bots. Lots of Twitter bots out there selling g
148.
by a_bonobo 2y ago
>We always have moved forward technologically despite doomers Have we? There's a de facto moratorium on gene editing in humans that all nations have so far adhered to (except for one person who promptly went to prison), there's
149.
by a_bonobo 2y ago
Is Trump personally disappearing these students of the streets? Is it Trump personally detaining and 'violently interrogating' people at the border? Is it Trump personally shutting down USAID? Is it Trump personally calculating th
150.
by a_bonobo 2y ago
>information itself must be entirely subsumed into an oppressive, proprietary, commercial apparatus I think that's the reason why I've (and probably many others?) have cooled down on general open source coding. Open source star

More ›