#detectgpt — Public Fediverse posts on home.social

Habr @[email protected] · 2024-01-27 · 18:02 UTC

Метод Binoculars обещает высокую точность обнаружения текста от больших языковых моделей

ChatGPT пишет не хуже человека, но можно ли обнаружить «машинность» в тексте? Хотя некоторым компаниям было бы выгоднее представить всё так, будто результат работы языковых моделей неотличим от человеческого, исследования в этом направлении активно ведутся. Авторы научной статьи «Spotting LLMs With Binoculars: Zero-Shot Detection of Machine-Generated Text» ( arXiv:2401.12070 ) утверждают, что их метод имеет низкий уровень ложноположительных срабатываний (0,01 %), правильно обнаруживает текст от языковых моделей в 90 % случаев и работает для нескольких семейств современных продуктов.

https://habr.com/ru/articles/789466/

#LLM #БЯМ #large_language_model #большая_языковая_модель #large_language_models #большие_языковые_модели #OpenAI #Binoculars #ИИ #искусственный_интеллект #обнаружение_машинного_текста #антиспам #GPTZero #DetectGPT #Ghostbuster #ChatGPT #GPT3 #GPT4 #Falcon #Falcon7B #Falcon7Binstruct

#falcon7binstruct #falcon7b #falcon #gpt4 #gpt3 #chatgpt

Habr @[email protected] · 2024-01-27 · 18:02 UTC

Метод Binoculars обещает высокую точность обнаружения текста от больших языковых моделей

ChatGPT пишет не хуже человека, но можно ли обнаружить «машинность» в тексте? Хотя некоторым компаниям было бы выгоднее представить всё так, будто результат работы языковых моделей неотличим от человеческого, исследования в этом направлении активно ведутся. Авторы научной статьи «Spotting LLMs With Binoculars: Zero-Shot Detection of Machine-Generated Text» ( arXiv:2401.12070 ) утверждают, что их метод имеет низкий уровень ложноположительных срабатываний (0,01 %), правильно обнаруживает текст от языковых моделей в 90 % случаев и работает для нескольких семейств современных продуктов.

https://habr.com/ru/articles/789466/

#LLM #БЯМ #large_language_model #большая_языковая_модель #large_language_models #большие_языковые_модели #OpenAI #Binoculars #ИИ #искусственный_интеллект #обнаружение_машинного_текста #антиспам #GPTZero #DetectGPT #Ghostbuster #ChatGPT #GPT3 #GPT4 #Falcon #Falcon7B #Falcon7Binstruct

#falcon7binstruct #falcon7b #falcon #gpt4 #gpt3 #chatgpt

Curtis C. Chen (he/him) 陳致宇 @[email protected] · 2023-03-22 · 22:55 UTC

CW: From birdsite

Journal articles that I downloaded recently but have not yet read: a drama in three parts #PrimarySources #DetectGPT #Bitcoin #MensHealth https://instagr.am/p/CqG2zIcvTf0/

#menshealth #bitcoin #detectgpt #primarysources

Curtis C. Chen (he/him) 陳致宇 @[email protected] · 2023-03-22 · 22:55 UTC

CW: From birdsite

Journal articles that I downloaded recently but have not yet read: a drama in three parts #PrimarySources #DetectGPT #Bitcoin #MensHealth https://instagr.am/p/CqG2zIcvTf0/

#menshealth #bitcoin #detectgpt #primarysources

Curtis C. Chen (he/him) 陳致宇 @[email protected] · 2023-03-22 · 22:55 UTC

CW: From birdsite

Journal articles that I downloaded recently but have not yet read: a drama in three parts #PrimarySources #DetectGPT #Bitcoin #MensHealth https://instagr.am/p/CqG2zIcvTf0/

#menshealth #bitcoin #detectgpt #primarysources

Curtis C. Chen (he/him) 陳致宇 @[email protected] · 2023-03-22 · 22:55 UTC

CW: From birdsite

Journal articles that I downloaded recently but have not yet read: a drama in three parts #PrimarySources #DetectGPT #Bitcoin #MensHealth https://instagr.am/p/CqG2zIcvTf0/

#primarysources #detectgpt #bitcoin #menshealth

Curtis C. Chen (he/him) 陳致宇 @[email protected] · 2023-03-22 · 22:55 UTC

CW: From birdsite

Journal articles that I downloaded recently but have not yet read: a drama in three parts #PrimarySources #DetectGPT #Bitcoin #MensHealth https://instagr.am/p/CqG2zIcvTf0/

#menshealth #bitcoin #detectgpt #primarysources

Scripter :verified_flashing: @[email protected] · 2023-02-21 · 06:54 UTC

DetectGPT findet heraus, ob ein Text von einem Menschen oder einer KI stammt
https://t3n.de/news/detectgpt-verraet-ob-text-von-ki-oder-menschen-text-chatgpt-1535142/ #DetectGPT #ChatGPT

#chatgpt #detectgpt

Scripter :verified_flashing: @[email protected] · 2023-02-21 · 06:54 UTC

DetectGPT findet heraus, ob ein Text von einem Menschen oder einer KI stammt
https://t3n.de/news/detectgpt-verraet-ob-text-von-ki-oder-menschen-text-chatgpt-1535142/ #DetectGPT #ChatGPT

#chatgpt #detectgpt

Scripter :verified_flashing: @[email protected] · 2023-02-21 · 06:54 UTC

DetectGPT findet heraus, ob ein Text von einem Menschen oder einer KI stammt
https://t3n.de/news/detectgpt-verraet-ob-text-von-ki-oder-menschen-text-chatgpt-1535142/ #DetectGPT #ChatGPT

#chatgpt #detectgpt

Scripter :verified_flashing: @[email protected] · 2023-02-21 · 06:54 UTC

DetectGPT findet heraus, ob ein Text von einem Menschen oder einer KI stammt
https://t3n.de/news/detectgpt-verraet-ob-text-von-ki-oder-menschen-text-chatgpt-1535142/ #DetectGPT #ChatGPT

#detectgpt #chatgpt

Scripter :verified_flashing: @[email protected] · 2023-02-21 · 06:54 UTC

DetectGPT findet heraus, ob ein Text von einem Menschen oder einer KI stammt
https://t3n.de/news/detectgpt-verraet-ob-text-von-ki-oder-menschen-text-chatgpt-1535142/ #DetectGPT #ChatGPT

#chatgpt #detectgpt

Scott 🐂 @[email protected] · 2023-02-19 · 16:36 UTC

#ChatGPT #GPTZeroX #DetectGPT 👾 https://medium.com/geekculture/how-to-detect-if-an-essay-was-generated-by-openais-chatgpt-58bb8adc8461

#chatgpt #gptzerox #detectgpt

Scott 🐂 @[email protected] · 2023-02-19 · 16:36 UTC

#ChatGPT #GPTZeroX #DetectGPT 👾 https://medium.com/geekculture/how-to-detect-if-an-essay-was-generated-by-openais-chatgpt-58bb8adc8461

#chatgpt #gptzerox #detectgpt

Scott 🐂 @[email protected] · 2023-02-19 · 16:36 UTC

#ChatGPT #GPTZeroX #DetectGPT 👾 https://medium.com/geekculture/how-to-detect-if-an-essay-was-generated-by-openais-chatgpt-58bb8adc8461

#detectgpt #gptzerox #chatgpt

Curtis C. Chen (he/him) 陳致宇 @[email protected] · 2023-02-15 · 23:46 UTC

And so the arms race has begun: https://hai.stanford.edu/news/human-writer-or-ai-scholars-build-detection-tool #ChatGPT #DetectGPT #NukeTheSiteFromOrbitGPT

#nukethesitefromorbitgpt #detectgpt #chatgpt

Curtis C. Chen (he/him) 陳致宇 @[email protected] · 2023-02-15 · 23:46 UTC

And so the arms race has begun: https://hai.stanford.edu/news/human-writer-or-ai-scholars-build-detection-tool #ChatGPT #DetectGPT #NukeTheSiteFromOrbitGPT

#nukethesitefromorbitgpt #detectgpt #chatgpt

Curtis C. Chen (he/him) 陳致宇 @[email protected] · 2023-02-15 · 23:46 UTC

And so the arms race has begun: https://hai.stanford.edu/news/human-writer-or-ai-scholars-build-detection-tool #ChatGPT #DetectGPT #NukeTheSiteFromOrbitGPT

#nukethesitefromorbitgpt #detectgpt #chatgpt

Curtis C. Chen (he/him) 陳致宇 @[email protected] · 2023-02-15 · 23:46 UTC

And so the arms race has begun: https://hai.stanford.edu/news/human-writer-or-ai-scholars-build-detection-tool #ChatGPT #DetectGPT #NukeTheSiteFromOrbitGPT

#chatgpt #detectgpt #nukethesitefromorbitgpt

Curtis C. Chen (he/him) 陳致宇 @[email protected] · 2023-02-15 · 23:46 UTC

And so the arms race has begun: https://hai.stanford.edu/news/human-writer-or-ai-scholars-build-detection-tool #ChatGPT #DetectGPT #NukeTheSiteFromOrbitGPT

#nukethesitefromorbitgpt #detectgpt #chatgpt

Günter Schwarz @[email protected] · 2023-01-31 · 18:29 UTC

#KI zum Entlarven von #KI
😆
##chat #chatgpt #detectGPT

https://www.derstandard.at/story/2000143084473/detect-gpt-soll-auf-anhieb-ki-texte-von-chat-gpt

#ki #chat #chatgpt #detectgpt

NatureLover 🌲 @[email protected] · 2023-01-31 · 11:47 UTC

Stanford introduces DetectGPT to help educators fight back against ChatGPT generated papers

https://www.neowin.net/news/stanford-introduces-detectgpt-to-help-educators-fight-back-against-chatgpt-generated-papers/

#chatgpt #ai #neowin #education #detectgpt #stanford

#stanford #detectgpt #education #neowin #ai #chatgpt

NatureLover 🌲 @naturelover · 2023-01-31 · 11:47 UTC

Stanford introduces DetectGPT to help educators fight back against ChatGPT generated papers

https://www.neowin.net/news/stanford-introduces-detectgpt-to-help-educators-fight-back-against-chatgpt-generated-papers/

#chatgpt #ai #neowin #education #detectgpt #stanford

NatureLover 🌲 @[email protected] · 2023-01-31 · 11:47 UTC

Stanford introduces DetectGPT to help educators fight back against ChatGPT generated papers

https://www.neowin.net/news/stanford-introduces-detectgpt-to-help-educators-fight-back-against-chatgpt-generated-papers/

#chatgpt #ai #neowin #education #detectgpt #stanford

#stanford #detectgpt #education #neowin #ai #chatgpt

NatureLover 🌲 @[email protected] · 2023-01-31 · 11:47 UTC

Stanford introduces DetectGPT to help educators fight back against ChatGPT generated papers

https://www.neowin.net/news/stanford-introduces-detectgpt-to-help-educators-fight-back-against-chatgpt-generated-papers/

#chatgpt #ai #neowin #education #detectgpt #stanford

NatureLover 🌲 @[email protected] · 2023-01-31 · 11:47 UTC

Stanford introduces DetectGPT to help educators fight back against ChatGPT generated papers

https://www.neowin.net/news/stanford-introduces-detectgpt-to-help-educators-fight-back-against-chatgpt-generated-papers/

#chatgpt #ai #neowin #education #detectgpt #stanford

#stanford #detectgpt #education #neowin #ai #chatgpt

Anne Fisher @[email protected] · 2023-01-31 · 04:44 UTC

#Stanford Introduces #DetectGPT to Help #Educators Fight Back Against #ChatGPT Generated Papers

"This method, called ' #ZeroShot,' allows DetectGPT to detect machine-written text without any knowledge of the #AI that was used to generate it. It operates in stark contrast to other methods which require training #classifiers and datasets of real and generated passages."

#ArtificialIntelligence #Education #Teaching #OpenAI #TechNews #TechnologyNews #BreakingNews

https://www.neowin.net/news/stanford-introduces-detectgpt-to-help-educators-fight-back-against-chatgpt-generated-papers/

#stanford #detectgpt #educators #chatgpt #zeroshot #ai

Anne Fisher @[email protected] · 2023-01-31 · 04:44 UTC

#Stanford Introduces #DetectGPT to Help #Educators Fight Back Against #ChatGPT Generated Papers

"This method, called ' #ZeroShot,' allows DetectGPT to detect machine-written text without any knowledge of the #AI that was used to generate it. It operates in stark contrast to other methods which require training #classifiers and datasets of real and generated passages."

#ArtificialIntelligence #Education #Teaching #OpenAI #TechNews #TechnologyNews #BreakingNews

https://www.neowin.net/news/stanford-introduces-detectgpt-to-help-educators-fight-back-against-chatgpt-generated-papers/

#stanford #detectgpt #educators #chatgpt #zeroshot #ai

Anne Fisher @[email protected] · 2023-01-31 · 04:44 UTC

#Stanford Introduces #DetectGPT to Help #Educators Fight Back Against #ChatGPT Generated Papers

"This method, called ' #ZeroShot,' allows DetectGPT to detect machine-written text without any knowledge of the #AI that was used to generate it. It operates in stark contrast to other methods which require training #classifiers and datasets of real and generated passages."

#ArtificialIntelligence #Education #Teaching #OpenAI #TechNews #TechnologyNews #BreakingNews

https://www.neowin.net/news/stanford-introduces-detectgpt-to-help-educators-fight-back-against-chatgpt-generated-papers/

#stanford #detectgpt #educators #chatgpt #zeroshot #ai

Anne Fisher @[email protected] · 2023-01-31 · 04:44 UTC

#Stanford Introduces #DetectGPT to Help #Educators Fight Back Against #ChatGPT Generated Papers

"This method, called ' #ZeroShot,' allows DetectGPT to detect machine-written text without any knowledge of the #AI that was used to generate it. It operates in stark contrast to other methods which require training #classifiers and datasets of real and generated passages."

#ArtificialIntelligence #Education #Teaching #OpenAI #TechNews #TechnologyNews #BreakingNews

https://www.neowin.net/news/stanford-introduces-detectgpt-to-help-educators-fight-back-against-chatgpt-generated-papers/

#breakingnews #technologynews #technews #openai #teaching #education

Anne Fisher @[email protected] · 2023-01-31 · 04:44 UTC

#Stanford Introduces #DetectGPT to Help #Educators Fight Back Against #ChatGPT Generated Papers

"This method, called ' #ZeroShot,' allows DetectGPT to detect machine-written text without any knowledge of the #AI that was used to generate it. It operates in stark contrast to other methods which require training #classifiers and datasets of real and generated passages."

#ArtificialIntelligence #Education #Teaching #OpenAI #TechNews #TechnologyNews #BreakingNews