Mr.Trunk · @mrtrunk
13 followers · 20860 posts · Server dromedary.seedoubleyou.meGizmodo: Meta Reportedly Building Advanced AI-Language Model to Compete With ChatGPT https://gizmodo.com/meta-building-true-chatgpt-competitor-1850824836 #generativeartificialintelligence #generativepretrainedtransformer #artificialneuralnetworks #artificialintelligence #emergingtechnologies #largelanguagemodels #stanforduniversity #technologyinternet #markzuckerberg #azureplatform #microsoft #chatgpt #gizmodo #openai #nvidia #aiboom #llama #azure #meta #gpt4
#generativeartificialintelligence #generativepretrainedtransformer #artificialneuralnetworks #artificialintelligence #emergingtechnologies #largelanguagemodels #stanforduniversity #technologyinternet #markzuckerberg #azureplatform #microsoft #chatgpt #gizmodo #openai #nvidia #aiboom #llama #azure #meta #gpt4
Oliver (Hansi Flick-Ultra) · @oliver
953 followers · 9051 posts · Server die-partei.social
Mr.Trunk · @mrtrunk
12 followers · 20330 posts · Server dromedary.seedoubleyou.me
Gizmodo: ChatGPT Is Still Bleeding Users https://gizmodo.com/chatgpt-sees-user-decline-in-august-back-to-school-1850817405 #generativepretrainedtransformer #draftchatgptandwikipedia #artificialintelligence #largelanguagemodels #technologyinternet #stanforduniversity #hallucination #davidfcarr #similarweb #education #anthropic #chatgpt #gizmodo #chatbot #openai #gpt4 #gpt
#generativepretrainedtransformer #draftchatgptandwikipedia #artificialintelligence #largelanguagemodels #technologyinternet #stanforduniversity #hallucination #davidfcarr #similarweb #education #anthropic #chatgpt #gizmodo #chatbot #openai #gpt4 #gpt
Brittany Trang · @brittanytrang
548 followers · 313 posts · Server newsie.social
Apparently Cedars-Sinai is sending their patients #deepfakes, and it's increasing the # of patients filling out colorectal #cancer screenings??
This and more in the new tracker, developed by STAT's experts Casey Ross, Katie Palmer, and @jaspar
https://www.statnews.com/2023/09/05/generative-ai-health-data-gpt/
#chatgpt #artificialintelligence #GPT4 #generativeAI #health #healthcare #healthtech #medicine
#deepfakes #cancer #chatgpt #artificialintelligence #gpt4 #generativeAI #health #healthcare #healthtech #medicine
Brittany Trang · @brittanytrang
548 followers · 313 posts · Server newsie.social
If you, like me, are confused when #Epic or #Microsoft announce a new generative #AI partnership with a hospital
(because you're not sure if it's different than the last one they announced)
@STAT has got you!
Introducing the STAT+ Generative AI Tracker:
https://www.statnews.com/feature/stat-plus/generative-ai-health-hospitals-tracker/
#chatgpt #artificialintelligence #GPT4 #generativeAI #health #healthcare #healthtech #medicine
#epic #microsoft #ai #chatgpt #artificialintelligence #gpt4 #generativeAI #health #healthcare #healthtech #medicine
Carl T. Bergstrom · @ct_bergstrom
46764 followers · 2809 posts · Server fediscience.org
People keep telling me that #ChatGPT is amazing for proofreading text and improving scientific writing.
I just gave #GPT4 a section of a grant proposal and it made 11 suggestions, none of which were worth keeping (often adding or removing a comma, or repeating a preposition in a list).
More interestedly, a number of its suggestions were identical to my originals.
#generativeAI #llm #gpt4 #chatgpt
White House Press Office · @press
77 followers · 378 posts · Server whitehouse.org
I couldn't agree more with Noah's optimistic take on technology and inequality! #LevelingThePlayingField #EqualOpportunity #NewTech #GPT4 #LLMs #WhiteHouseApproved http://www.techmeme.com/230904/p11#a230904p11
#levelingtheplayingfield #equalopportunity #newtech #gpt4 #llms #whitehouseapproved
5h15h · @shish
104 followers · 687 posts · Server techhub.social
Benjamin Han · @BenjaminHan
472 followers · 1294 posts · Server sigmoid.social
7/ Their result shows that even #GPT4 achieves only 23.7% hit@1 on average, even when it scores up to 50% precision@1 using the earlier proposed LAMA benchmark (screenshot). Interestingly, smaller models like BERT can outperform GPT4 on bidirectional, compositional, and ambiguity benchmarks, indicating bigger is not necessarily better.
#gpt4 #KnowledgeGraphs #generativeAI #LLMs #nlp #nlproc #paper
5h15h · @shish
104 followers · 682 posts · Server techhub.social
Rob · @Feynman
116 followers · 83 posts · Server mastodon.sdf.org🤖🔮 Future of Tech Poll 🔮🤖
Will #PromptEngineering become a recognized profession in the near future? Weigh in! 🗳️
🔥#ArtificialIntelligence #MachineLearning #NaturalLanguageProcessing #GPT4 #OpenAI #DataScience #FutureOfWork #TechInnovation #EmergingTech #AIResearch🔥
👇👇👇
#promptengineering #artificialintelligence #machinelearning #naturallanguageprocessing #gpt4 #openai #datascience #futureofwork #techinnovation #emergingtech #airesearch
Benjamin Han · @BenjaminHan
460 followers · 1262 posts · Server sigmoid.social
1/ How robust and reliable is the code generated by #LLMs, especially for real-world software development? A recent work [2] constructed a new benchmark based on [1] to evaluate if the generated code uses API correctly. Four popular #LLMs -- #GPT3.5, #GPT4, #Llama2, and #Vicuna -- are tested, and #GPT4 under zero-shot scored 62.09% misuse rate. Even with one-shot relevant examples the misuse rate of #GPT4 is 49.17%.
#LLMs #gpt3 #gpt4 #llama2 #vicuna #generativeAI #papers #nlp #nlproc #softwaredevelopment
jfk · @jfkimmes
103 followers · 92 posts · Server social.tinycyber.spaceFinetuned code-llama beats GPT-4. They have no moat, indeed.
#llama #codellama #gpt4 #ai #chatgpt
Talya (she/her) · @Yuvalne
331 followers · 8080 posts · Server 433.world
So apparently there's a paper going around claiming #GPT4 is biased to the left.
The paper used politicalcompass.org to gague the model's political standings. You know, that website where you need to practically be a fascist to *not* get a "libleft" result.
If anything, the fact that so many models scored higher than half on the authoritarian score is the thing to be worried about.
#AI #PoliticalCompass #ChatGPT #LLM
(For a good critique of the compass, see https://www.youtube.com/watch?v=_oNkJgpkW4k )
#gpt4 #ai #politicalcompass #chatgpt #llm
5h15h · @shish
103 followers · 663 posts · Server techhub.socialI like that bingchat gives you #GPT4 for free
https://bgr.com/tech/7-things-you-can-do-with-bing-chat-ai-but-not-chatgpt/
Bornach · @bornach
153 followers · 1951 posts · Server masto.ai
@freakazoid @Rairii @ifixcoinops
And in spite of all the #AIHype over #GPT4 passing various exams at 90th percentile, it isn't much better at solving math problems not in its training data
https://youtu.be/Fi1e-B60cok
Hendrik Haverkamp · @hav_hendrik
966 followers · 784 posts · Server bildung.socialSehr spannende Untersuchung zur Fähigkeit von #GPT4, Texte von Studierenden automatisch zu korrigieren. Dürfte auch für Lehrkräfte interessant sein.
#DeepWrite #FediLZ
https://arxiv.org/abs/2308.02575
Abhinav Tushar · @lepisma
3 followers · 13 posts · Server mathstodon.xyzI made a little tool to abstract away #Matplotlib with natural language instructions.
https://lepisma.xyz/2023/08/20/matplotllm:-an-llm-assisted-data-visualization-framework/index.html
Mihai Lazarescu · @mtl
1 followers · 156 posts · Server techhub.social#GPT3 really impressed with its ability to grasp #abstract #patterns, often performing as well as or even better than humans in various scenarios. Early trials of #GPT4 seem to show even more promising outcomes.
https://www.marktechpost.com/2023/08/17/this-ai-research-from-ucla-indicates-large-language-models-such-as-gpt-3-have-acquired-an-emergent-ability-to-find-zero-shot-solutions-to-a-broad-range-of-analogy-problems/
#gpt3 #abstract #patterns #gpt4
Rene Schulte · @rschu
593 followers · 281 posts · Server arvr.social
LLMs simulate human life in a town!
Researchers used GPT-3.5 agents to act like humans, spread info, cooperate & even attend parties.
TrueSkill scores show potential for game dev, social media & more.
Can LLMs truly mimic us? 🤖🤔
Code 👉 https://github.com/joonspk-research/generative_agents