randomwalker

7 ms·

31.
On the Societal Impact of Open Foundation Models [pdf](crfm.stanford.edu)
2 points by randomwalker 3y ago|0 comments
32.
Will AI transform law? The hype is not supported by current evidence(aisnakeoil.com)
2 points by randomwalker 3y ago|0 comments
33.
Generative AI's end-run around copyright won't be resolved by the courts(aisnakeoil.com)
4 points by randomwalker 3y ago|2 comments
34.
Model alignment protects against accidental harms, not intentional ones(aisnakeoil.com)
1 points by randomwalker 3y ago|0 comments
35.
What the executive order means for openness in AI(aisnakeoil.com)
2 points by randomwalker 3y ago|0 comments
36.
The Foundation Model Transparency Index(crfm.stanford.edu)
47 points by randomwalker 3y ago|16 comments
37.
Evaluating LLMs Is a Minefield(cs.princeton.edu)
3 points by randomwalker 3y ago|0 comments
38.
Does ChatGPT have a liberal bias?(aisnakeoil.com)
4 points by randomwalker 3y ago|2 comments
39.
The REFORMS checklist for ML-based science(aisnakeoil.com)
2 points by randomwalker 3y ago|0 comments
40.
ML is useful for many things, but not for predicting scientific replicability(aisnakeoil.com)
116 points by randomwalker 3y ago|37 comments
41.
Is GPT-4 getting worse over time?(aisnakeoil.com)
7 points by randomwalker 3y ago|1 comments
42.
by randomwalker 3y ago
This paper is being misinterpreted. The degradations reported are somewhat peculiar to the authors' task selection and evaluation method and can easily result from fine tuning rather than intentionally degrading GPT-4's performanc
43.
Generative AI companies must publish transparency reports(knightcolumbia.org)
1 points by randomwalker 3y ago|0 comments
44.
by randomwalker 3y ago
OP here. Unfortunately this thread is mostly misinformation. There were a bunch of viral threads from the growth hacker / influencer crowd, including this one, within hours of the code release with a very superficial understanding of t
45.
Twitter showed us its algorithm – what does it tell us?(knightcolumbia.org)
225 points by randomwalker 3y ago|163 comments
46.
by randomwalker 4y ago
OP here. The CNET thing is actually pretty egregious, and not the kind of errors a human would make. These are the original investigations, if you'll excuse the tone: https://futurism.com/cnet-ai-errors https:/&#
47.
A misleading open letter about sci-fi AI dangers ignores the real risks(aisnakeoil.substack.com)
6 points by randomwalker 4y ago|1 comments
48.
by randomwalker 4y ago
Rebuttal: https://aisnakeoil.substack.com/p/a-misleading-open-letter-a... Summary: misinfo, labor impact, and safety are real dangers of LLMs. But in each case the letter invokes speculative, futuristic risks, ignoring
49.
by randomwalker 4y ago
We don't expect it to be free -- please read the article. That's not the issue at all. It's like if you subscribe to a product that you need to do your job, and one day the company tells you that the product is going away in
50.
by randomwalker 4y ago
Sure, but the article is talking about a completely different meaning of reproducibility, where a researcher uses an LLM as a tool to study some research question, and someone else comes along and wants to check whether the claims hold up.
51.
by randomwalker 4y ago
We're under no such misapprehension and we're keenly aware that this is an uphill battle. The issue is that LLMs have become part of the infrastructure of the Internet. Companies that build infrastructure have a responsibility to
52.
by randomwalker 4y ago
Addressed in the article: "OpenAI responded to the criticism by saying they'll allow researchers access to Codex. But the application process is opaque: researchers need to fill out a form, and the company decides who gets approve
53.
OpenAI’s policies hinder reproducible research on language models(aisnakeoil.substack.com)
610 points by randomwalker 4y ago|371 comments
54.
GPT-4 and professional benchmarks: the wrong answer to the wrong question(aisnakeoil.substack.com)
3 points by randomwalker 4y ago|0 comments
55.
URL to bibtex converter: GPT-4 code generation, GPT-3 back end(twitter.com)
1 points by randomwalker 4y ago|0 comments
56.
Understanding Social Media Recommendation Algorithms(knightcolumbia.org)
40 points by randomwalker 4y ago|2 comments
57.
Artists can now opt out of generative AI. It’s not enough(aisnakeoil.substack.com)
4 points by randomwalker 4y ago|0 comments
58.
The LLaMA is out of the bag. Should we expect a tidal wave of disinformation?(aisnakeoil.substack.com)
3 points by randomwalker 4y ago|0 comments
59.
by randomwalker 4y ago
OP here. Many people are reacting to the title of the paper. A few thoughts: * The paper is 35 pages long and it's hard to convey its message in any single title. We make clear in the text that our point is not that predictive optimiza
60.
Against predictive optimization(predictive-optimization.cs.princeton.edu)
162 points by randomwalker 4y ago|78 comments

More ›