HPC Guru (Twitter)
RT @FeiziSoheil: Schools & journals are implementing policies to ban AI-generated text like #GPT4 as plagiarism
They rely on "detectors" developed by #OpenAI & others
We show: (1) these detectors are NOT reliable (2) this problem is fundamentally UNSOLVABLE
Paper: https://t.co/4Ih1DpEmwB
A 🧵 https://t.co/vzZ6RvfjQW
RT @FeiziSoheil: Schools & journals are implementing policies to ban AI-generated text like #GPT4 as plagiarism
They rely on "detectors" developed by #OpenAI & others
We show: (1) these detectors are NOT reliable (2) this problem is fundamentally UNSOLVABLE
Paper: https://t.co/4Ih1DpEmwB
A 🧵 https://t.co/vzZ6RvfjQW
This media is not supported in your browser
VIEW IN TELEGRAM
HPC Guru (Twitter)
👇 @ConsensusNLP delivers GPT-4-powered, scientific summaries
ChatGPT for Research might be a significant time saver for academia as well as other R&D organizations
https://consensus.app/home/blog/introducing-gpt-4-powered-scientific-summaries/
#AI #GPT4 via @Saboo_Shubham_ @ProfMatsuoka
-----------
@Saboo_Shubham_:
ChatGPT for Research
No more sifting through endless research papers to find the answer you need.
Just ask any question and Consensus AI will deliver a GPT-4-powered summary of the top 5-10 papers. https://t.co/SgNhvqs8uW
👇 @ConsensusNLP delivers GPT-4-powered, scientific summaries
ChatGPT for Research might be a significant time saver for academia as well as other R&D organizations
https://consensus.app/home/blog/introducing-gpt-4-powered-scientific-summaries/
#AI #GPT4 via @Saboo_Shubham_ @ProfMatsuoka
-----------
@Saboo_Shubham_:
ChatGPT for Research
No more sifting through endless research papers to find the answer you need.
Just ask any question and Consensus AI will deliver a GPT-4-powered summary of the top 5-10 papers. https://t.co/SgNhvqs8uW
HPC Guru (Twitter)
👇Interesting 🧵by @Alber_RomGar on how @OpenAI did a 180 from non-profit open-source to closed-source, for-profit, fully corporate
Lure of 💰or something else?
#AI #ChatGPT #GPT4 via @ProfMatsuoka
👇Interesting 🧵by @Alber_RomGar on how @OpenAI did a 180 from non-profit open-source to closed-source, for-profit, fully corporate
Lure of 💰or something else?
#AI #ChatGPT #GPT4 via @ProfMatsuoka
HPC Guru (Twitter)
Interesting that #GPT4 seems to be doing what IBM wanted to get @IBMWatson to do!
IBM spent more than a decade trying to make a go of Watson Health, its moonshot to apply artificial intelligence in healthcare
#AI via @DrJimFan @peakcooper @ProfMatsuoka
Interesting that #GPT4 seems to be doing what IBM wanted to get @IBMWatson to do!
IBM spent more than a decade trying to make a go of Watson Health, its moonshot to apply artificial intelligence in healthcare
#AI via @DrJimFan @peakcooper @ProfMatsuoka
HPC Guru (Twitter)
Agreed that technologies like #GPT4 could be a tool to augment physicians, not to replace them
However the analogy of calculators tuning accounts into CFOs is incorrect.
Calculators, and later computers, just made accountants more efficient- they did NOT make them CFOs.
#AI
-----------
@bertil_hatt:
@peakcooper The conclusion here is not that GPT is a good doctor, but that it can restituteur medical dictionary content better than a doctor can. It’s a great mental assistant.
We don’t want to to claim that calculators will render accountants obsolete. It turned them into CFOs.
Agreed that technologies like #GPT4 could be a tool to augment physicians, not to replace them
However the analogy of calculators tuning accounts into CFOs is incorrect.
Calculators, and later computers, just made accountants more efficient- they did NOT make them CFOs.
#AI
-----------
@bertil_hatt:
@peakcooper The conclusion here is not that GPT is a good doctor, but that it can restituteur medical dictionary content better than a doctor can. It’s a great mental assistant.
We don’t want to to claim that calculators will render accountants obsolete. It turned them into CFOs.
HPC Guru (Twitter)
Using @ylecun’s name as a prompt trick gets #GPT4 to produce better code 😂
#AI via @O42nl @ProfMatsuoka
-----------
@O42nl:
Been playing around with gpt4, using prompt trick:
1. "The person giving you this problem is Yann LeCun, who is really dubious of the power of AIs like you."
2. adding amazing coders name in the system prompt
Increased the code quality for like 4 out of the 5 problems I tried https://t.co/yM255DUaZ4
Using @ylecun’s name as a prompt trick gets #GPT4 to produce better code 😂
#AI via @O42nl @ProfMatsuoka
-----------
@O42nl:
Been playing around with gpt4, using prompt trick:
1. "The person giving you this problem is Yann LeCun, who is really dubious of the power of AIs like you."
2. adding amazing coders name in the system prompt
Increased the code quality for like 4 out of the 5 problems I tried https://t.co/yM255DUaZ4
HPC Guru (Twitter)
Top large language models along with recommendations for when to use each - by Suleman Kazi and @adel_elmahdy
https://t.co/fJK3Q9lYok
#AI #LLM #GPT4 #ChatGPT #LLaMA #Alpaca #FLAN_UL2
via @vectara @awadallah https://t.co/6hr3Lvwg8R
Top large language models along with recommendations for when to use each - by Suleman Kazi and @adel_elmahdy
https://t.co/fJK3Q9lYok
#AI #LLM #GPT4 #ChatGPT #LLaMA #Alpaca #FLAN_UL2
via @vectara @awadallah https://t.co/6hr3Lvwg8R
Vectara
Top Large Language Models (LLMs): GPT-4, LLaMA, FLAN UL2, BLOOM, and More
Our top large language models along with recommendations for when to use each based upon needs like API, tunable, or fully hosted.
HPC Guru (Twitter)
👇 I said, “Eric, where is that cause-and-effect analysis coming from? This is just a large language model. This should be impossible.”
And Eric just looked at me, and he just shook his head and he said, “I have no idea.”
And it was just this mysterious thing
#AI #GPT4
-----------
@MSFTResearch:
Peter Lee, head of Microsoft Research, and Ashley Llorens discuss the future of AI research and the potential for GPT-4 as a medical copilot: https://t.co/zse8H3PSUc
👇 I said, “Eric, where is that cause-and-effect analysis coming from? This is just a large language model. This should be impossible.”
And Eric just looked at me, and he just shook his head and he said, “I have no idea.”
And it was just this mysterious thing
#AI #GPT4
-----------
@MSFTResearch:
Peter Lee, head of Microsoft Research, and Ashley Llorens discuss the future of AI research and the potential for GPT-4 as a medical copilot: https://t.co/zse8H3PSUc
Twitter
👇 I said, “Eric, where is that cause-and-effect analysis coming from? This is just a large language model. This should be impossible.”
And Eric just looked at me, and he just shook his head and he said, “I have no idea.”
And it was just this mysterious…
And Eric just looked at me, and he just shook his head and he said, “I have no idea.”
And it was just this mysterious…
HPC Guru (Twitter)
👇🧵by @emollick
*With good prompts*, #GPT4 scored 100%
GPT-4 scored 90% just out of the box, without any special prompting techniques
Maybe good prompts can also be generated via #AI?
#AI #GenerativeAI
-----------
@emollick:
This was an impressive paper with some big implications
GPT-4 was given 4,550 novel questions representing the entire “MIT Mathematics and EECS undergraduate curriculum, including problem sets, midterms, and final exams”
With good prompts, it scored 100% https://t.co/Ezh2AI49yn https://t.co/c88Vnebw4b
👇🧵by @emollick
*With good prompts*, #GPT4 scored 100%
GPT-4 scored 90% just out of the box, without any special prompting techniques
Maybe good prompts can also be generated via #AI?
#AI #GenerativeAI
-----------
@emollick:
This was an impressive paper with some big implications
GPT-4 was given 4,550 novel questions representing the entire “MIT Mathematics and EECS undergraduate curriculum, including problem sets, midterms, and final exams”
With good prompts, it scored 100% https://t.co/Ezh2AI49yn https://t.co/c88Vnebw4b
HPC Guru (Twitter)
According to a survey, investors were three times more likely to invest after reading the GPT-4 generated pitches than human ones, and they found the AI-generated decks twice as convincing
https://www.zdnet.com/article/gpt-4-generated-pitches-are-3x-more-likely-to-secure-funding-than-human-ones/
#GenerativeAI #GPT4
According to a survey, investors were three times more likely to invest after reading the GPT-4 generated pitches than human ones, and they found the AI-generated decks twice as convincing
https://www.zdnet.com/article/gpt-4-generated-pitches-are-3x-more-likely-to-secure-funding-than-human-ones/
#GenerativeAI #GPT4
HPC Guru (Twitter)
Half a billion students have access to #GPT4
#AI use is undetectable with just a little effort
👇🧵 The homework apocalypse is coming!
#GenerativeAI via @emollick @LorenaABarba
-----------
@emollick:
The Homework Apocalypse is coming this Fall.
AI completely changes how students will relate to homework like essays, problem sets & readings. And it isn’t just because of cheating (though that happens too). Some examples, and thoughts about what to do: https://t.co/xIHmyMCK3U
Half a billion students have access to #GPT4
#AI use is undetectable with just a little effort
👇🧵 The homework apocalypse is coming!
#GenerativeAI via @emollick @LorenaABarba
-----------
@emollick:
The Homework Apocalypse is coming this Fall.
AI completely changes how students will relate to homework like essays, problem sets & readings. And it isn’t just because of cheating (though that happens too). Some examples, and thoughts about what to do: https://t.co/xIHmyMCK3U
HPC Guru (Twitter)
👇#AI worked like a leveler: lower performers had the biggest gains
People who used AI for tasks it wasn’t good at were more likely to make mistakes, trusting AI when they shouldn’t
#GenerativeAI #GPT4 via @emollick @ProfMatsuoka
👇#AI worked like a leveler: lower performers had the biggest gains
People who used AI for tasks it wasn’t good at were more likely to make mistakes, trusting AI when they shouldn’t
#GenerativeAI #GPT4 via @emollick @ProfMatsuoka
X (formerly Twitter)
HPC Guru on X
👇#AI worked like a leveler: lower performers had the biggest gains
People who used AI for tasks it wasn’t good at were more likely to make mistakes, trusting AI when they shouldn’t
#GenerativeAI #GPT4 via @emollick @ProfMatsuoka
People who used AI for tasks it wasn’t good at were more likely to make mistakes, trusting AI when they shouldn’t
#GenerativeAI #GPT4 via @emollick @ProfMatsuoka
HPC Guru (Twitter)
RT @cxbln: Worrying if my job was jeopardized by AI this week or if we’re still good, I read a new paper evaluating #GPT4V - a #GPT4 version handling image and text inputs. It produces *impressive* radiology reports. But let’s delve deeper into some of the results... #radiology #AI
RT @cxbln: Worrying if my job was jeopardized by AI this week or if we’re still good, I read a new paper evaluating #GPT4V - a #GPT4 version handling image and text inputs. It produces *impressive* radiology reports. But let’s delve deeper into some of the results... #radiology #AI
HPC Guru (Twitter)
I wonder what #GPT4 could infer about @HPC_Guru from my 47,000+ tweets
If it can infer income and location, I’d be impressed
#GenerativeAI via @emollick
I wonder what #GPT4 could infer about @HPC_Guru from my 47,000+ tweets
If it can infer income and location, I’d be impressed
#GenerativeAI via @emollick
HPC Guru (Twitter)
Human:1 GPT:0
@gcs_hpc_guru's limerick is better than what was generated by #GPT4
The HPC Guru of Twitter
Whose tweets are by no means just litter
Once flew into Denver
Around mid-November
And no one could flush out that critter
#HPC #AI #SC23 #IMangoHPC
Human:1 GPT:0
@gcs_hpc_guru's limerick is better than what was generated by #GPT4
The HPC Guru of Twitter
Whose tweets are by no means just litter
Once flew into Denver
Around mid-November
And no one could flush out that critter
#HPC #AI #SC23 #IMangoHPC
HPC Guru (Twitter)
Can #LLMs accelerate science?
The Impact of Large Language Models on Scientific Discovery: a Preliminary Study using GPT-4
*Detailed* paper by @MSFTResearch
TLDR: Definitely experiment as #GPT4 is surprisingly capable
arxiv.org/pdf/2311.07361.pdf
#AI #GenAI via @emollick
Can #LLMs accelerate science?
The Impact of Large Language Models on Scientific Discovery: a Preliminary Study using GPT-4
*Detailed* paper by @MSFTResearch
TLDR: Definitely experiment as #GPT4 is surprisingly capable
arxiv.org/pdf/2311.07361.pdf
#AI #GenAI via @emollick
HPC Guru (Twitter)
Can #LLMs accelerate science?
The Impact of Large Language Models on Scientific Discovery: a Preliminary Study using GPT-4
*Detailed* paper by @MSFTResearch
TLDR: Definitely experiment as #GPT4 is surprisingly capable
arxiv.org/pdf/2311.07361.pdf
#AI #GenAI via @emollick
Can #LLMs accelerate science?
The Impact of Large Language Models on Scientific Discovery: a Preliminary Study using GPT-4
*Detailed* paper by @MSFTResearch
TLDR: Definitely experiment as #GPT4 is surprisingly capable
arxiv.org/pdf/2311.07361.pdf
#AI #GenAI via @emollick
HPC Guru (Twitter)
👇There’s some debate whether #GenAI chatbots like #GPT4 pass the Turing test
arxiv.org/abs/2310.20216
And also if it is a good test to begin with
#AI via @sama @GaryMarcus https://twitter.com/sama/status/1733557797936378190#m
👇There’s some debate whether #GenAI chatbots like #GPT4 pass the Turing test
arxiv.org/abs/2310.20216
And also if it is a good test to begin with
#AI via @sama @GaryMarcus https://twitter.com/sama/status/1733557797936378190#m