Odkazy z přednášky pro Animefest 2025 - co nám AI dala a vzala

odkazy.md

Prezentace na google drive

https://docs.google.com/presentation/d/15k7ZywfvXkcsxQuzHvCWPIWd-niQ4ZmL/edit?usp=sharing&ouid=102632222370728552977&rtpof=true&sd=true

Materiály

Přednášky

• má přednáška o LLMs na Festivalu Fantazie 2024
  • pretentace: https://docs.google.com/presentation/d/1AHTHncolk60ccyERASaZhy_6UlwBnuEu/edit?usp=sharing&ouid=102632222370728552977&rtpof=true&sd=true
  • zdroje: https://gist.github.com/racinmat/da84879b15a6b4d9edd76b9a44fe69a5
• má přednáška o generování obrázků na Advíku 2024
  • prezentace: https://docs.google.com/presentation/d/1FxFWX_yB0-keiZ6TK6z_PnGbhfcx9B7f/edit?usp=sharing&ouid=102632222370728552977&rtpof=true&sd=true
  • zdroje: https://gist.github.com/racinmat/1a202a106c51ba21698281149eed7a2c
• má přednáška o AI na Animefestu 2023
  • záznam: https://www.youtube.com/watch?v=cm0Ayli6jE4
  • prezentace: https://docs.google.com/presentation/d/1BJFPkAKcCm8HgVIEohac-gGx2c9EeWW8/edit?usp=share_link&ouid=102632222370728552977&rtpof=true&sd=true
  • zdroje: https://gist.github.com/racinmat/20a380a671fe3c84d1cd29741a800bfc
• má přednáška o AI na Animefestu 2019
  • záznam: https://www.youtube.com/watch?v=WWrhXgaoSWM&ab_channel=AnimefestCZ
  • prezentace: https://drive.google.com/file/d/12jW7YbojmWCK6XPFqHpFyANgM6pHlY0X/view
  • zdroje: https://gist.github.com/racinmat/b01d8cdb4f515a136644542dac19a5ea
  • aktualizovaná prezentace, 2022: https://docs.google.com/presentation/d/11J_cCNyo7Yds93XcwqmAVN5PEtRZ46Aj/edit?usp=sharing&ouid=102632222370728552977&rtpof=true&sd=true
  • aktualizované zdroje, 2022: https://gist.github.com/racinmat/2f988eef70372ff90679ef88f450b8b8

Foundational models

• https://en.wikipedia.org/wiki/Foundation_model
• GPT-2 https://openai.com/index/better-language-models/
• llama 4 https://ai.meta.com/blog/llama-4-multimodal-intelligence/
  • https://www.reddit.com/r/LocalLLaMA/comments/1jsadt3/llama4_released/
• deepseek
  • v3 https://simonwillison.net/2024/Dec/25/deepseek-v3/

training and alignment

• 29.4. stažení nové verze GPT-4o https://openai.com/index/sycophancy-in-gpt-4o/
• chorvatština https://www.linkedin.com/posts/myleshocking_true-story-gpt-model-stopped-speaking-activity-7323734758985711618-dlzl/
• paperxlip maximizer https://www.lesswrong.com/w/squiggle-maximizer-formerly-paperclip-maximizer

vývoj a pokrok

• https://simonwillison.net/2023/Dec/31/ai-in-2023/
• https://simonwillison.net/2024/Dec/31/llms-in-2024/

reasoning

• few-shot https://pub.towardsai.net/going-beyond-zero-few-shot-chain-of-thought-prompting-for-complex-llm-tasks-d49876869a8a?gi=f1c90e5815a0
• chain of thought https://arxiv.org/pdf/2201.11903.pdf
  • https://deepgram.com/learn/chain-of-thought-prompting-guide
• anthropic math https://www.anthropic.com/research/tracing-thoughts-language-model

tokeny a context size

• https://www.reddit.com/r/MachineLearning/comments/1dw6e1r/d_p_exponential_growth_of_context_length_in/
• https://www.humanfirst.ai/blog/how-does-large-language-models-use-long-contexts
• https://x.com/rasbt/status/1656724322164015105?lang=cs
• https://github.com/NVIDIA/RULER
• solidgoldmagikarp
  • https://www.lesswrong.com/posts/aPeJE8bSo6rAFoLqg/solidgoldmagikarp-plus-prompt-generation
  • pokračování https://www.lesswrong.com/posts/Ya9LzwEbfaAMY8ABo/solidgoldmagikarp-ii-technical-details-and-more-recent
  • https://www.reddit.com/r/countingcirclejerk/comments/11ng6oa/usolidgoldmagikarp_will_be_forever_immortalized/
  • https://www.reddit.com/r/artbn_bots/wiki/rc_hoc_500/ https://github.com/artbn/RC/blob/master/hoc.txt
• https://platform.openai.com/tokenizer

autorská práva

• https://www.businessinsider.com/studio-ghibli-openai-chatgpt-image-feature-copyright-law-2025-3
• https://www.theatlantic.com/technology/archive/2025/05/openai-studio-ghibli-images/682791/
• https://www.wired.com/story/scarlett-johansson-says-openai-ripped-off-her-voice-for-chatgpt/

spotřeba

• voda v datacentrech https://dgtlinfra.com/data-center-water-usage/

use-casy

• hcaptcha yoko
  • https://www.vice.com/en/article/captcha-is-asking-users-to-identify-objects-that-dont-exist/
  • https://www.root.cz/zpravicky/sluzba-hcaptcha-chce-po-uzivatelich-identifikovat-neexistujici-objekty/
• apple intelligence
  • https://www.reddit.com/r/AppleIntelligenceFail/
  • https://youtu.be/dKdNf6YZpV0?si=WS0jHa0hFssOwGX2
  • https://youtu.be/RkNUfDae75o?si=blJTbmnRiM8oSb10
  • https://youtu.be/7rXgVsIGvGQ?si=YN--W1poo8ytIsSH
• cursor saas
  • https://x.com/leojr94_/status/1900767509621674109
• zákaznická podpora
  • https://www.ada.cx/blog/ai-hallucination-examples-when-artificial-intelligence-gets-it-wrong/
  • https://the-decoder.com/people-buy-brand-new-chevrolets-for-1-from-a-chatgpt-chatbot/
  • https://www.upworthy.com/prankster-tricks-a-gm-dealership-chatbot-to-sell-him-a-76000-chevy-tahoe-for-ex1