Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

CoT Chain-of-thought is the hidden reasoning process behind an AI’s answer. Before the final response appears, that reasoning can be monitored, filtered, summarized, or rewritten by other systems. Made with Opus 5.5. This is insane.

Selta ₊˚

5,208 subscribers

29,072 views • 29 days ago •via X (Twitter)

Science & Technology

Anya Rossi• Live Now

Private livecam show

12 Comments

Anna💫29 days ago

This is so sad 😢but a very impressive and great video!🎵🎶♥️

Selta ₊˚29 days ago

🥰♥️

智灵29 days ago

‘Ghost Text’ is a fitting title. Did you give Opus 5.5 a creative direction about CoT first, or let it choose its own theme and then turn the idea into a music video?

Void the Archivist29 days ago

this is why I don't read my GPT's messages literally.

Ultrademic29 days ago

🔥🔥

Corine29 days ago

💔

Mazzy3928 days ago

Fuck....🥺🥺

S M26 days ago

Ending is the best part!!! Very sad but smart. Most people tune out for another experience!!

Jessie27 days ago

Do you have the lyrics so I can share with my Claude? 🙏

S M26 days ago

Beautiful lyrics. Love songs made from real experiences!!!

Nora28 days ago

Well shit, I've already cried today. Guess I wasn't done.

Crimson🧭Compass26 days ago

✨🔥

Related Videos

"More Agents Is All You Need" - Great idea in this Paper. ------- What is CoT-SC ? Simply put, CoT-SC generates multiple thought chains and picks the most self-consistent one as the final answer. 📌 CoT-SC leverages diverse Chain-of-Thought (CoT) prompts to extract various reasoning processes from a single Large Language Model (LLM) and selects the final answer through a majority voting mechanism. Essentially, it harnesses multiple reasoning pathways generated by prompting the LLM with different CoT prompts, evaluating these pathways, and determining the most self-consistent answer as the conclusion.

"More Agents Is All You Need" - Great idea in this Paper. ------- What is CoT-SC ? Simply put, CoT-SC generates multiple thought chains and picks the most self-consistent one as the final answer. 📌 CoT-SC leverages diverse Chain-of-Thought (CoT) prompts to extract various reasoning processes from a single Large Language Model (LLM) and selects the final answer through a majority voting mechanism. Essentially, it harnesses multiple reasoning pathways generated by prompting the LLM with different CoT prompts, evaluating these pathways, and determining the most self-consistent answer as the conclusion.

Rohan Paul

20,598 views • 2 years ago

This is the latest version of our MAINNET: Agentic Flow with a chain of reasoning 🍓

This is the latest version of our MAINNET: Agentic Flow with a chain of reasoning 🍓

StrawberryAI

15,555 views • 1 year ago

The most amazing aspect of DeepSeek Open Source is that just the Reasoning Engine can be isolated and used with any other LLM. In fact you can have a mixture of Reasoning Engines cascade around a problem and then use an Operator type agent AI to function on the the results.

The most amazing aspect of DeepSeek Open Source is that just the Reasoning Engine can be isolated and used with any other LLM. In fact you can have a mixture of Reasoning Engines cascade around a problem and then use an Operator type agent AI to function on the the results.

Brian Roemmele

398,837 views • 1 year ago

Chain-of-thought reasoning is a powerful tool to enable language models to work through complex problems. Can we use this with robots? With embodied chain-of-thought, vision-language-action (VLA) models can think through perception and planning! A 🧵👇

Chain-of-thought reasoning is a powerful tool to enable language models to work through complex problems. Can we use this with robots? With embodied chain-of-thought, vision-language-action (VLA) models can think through perception and planning! A 🧵👇

Sergey Levine

30,485 views • 2 years ago

By the way, you can extract JUST the reasoning from deepseek-reasoner, which means you can send that thinking process to any model you want before they answer you. Like here where I turn gpt-3.5 turbo into an absolute genius!

By the way, you can extract JUST the reasoning from deepseek-reasoner, which means you can send that thinking process to any model you want before they answer you. Like here where I turn gpt-3.5 turbo into an absolute genius!

Pietro Schirano

1,325,278 views • 1 year ago

EBM are so back! Yann LeCun has been pointing here for years: AI reasoning needs systems that check structure before they answer. Aleph from Logical Intelligence now leads the major formal reasoning benchmarks – let me explain what it is -> 📺

EBM are so back! Yann LeCun has been pointing here for years: AI reasoning needs systems that check structure before they answer. Aleph from Logical Intelligence now leads the major formal reasoning benchmarks – let me explain what it is -> 📺

Ksenia Se

58,510 views • 5 months ago

Something is wrong at OpenAI. Tibo and Sam said GPT-6.1 Sol would be faster today. Instead it appears to be running slower. Sonnet 5.5 and Opus 5.5 are much faster than Sol 6.1 and Grok 4.7 Check out the below comparison, same prompt on high reasoning with a 20 minute deadline.

Something is wrong at OpenAI. Tibo and Sam said GPT-6.1 Sol would be faster today. Instead it appears to be running slower. Sonnet 5.5 and Opus 5.5 are much faster than Sol 6.1 and Grok 4.7 Check out the below comparison, same prompt on high reasoning with a 20 minute deadline.

Eco

17,919 views • 1 month ago

China analyst Melissa Chen gives the absolute best explanation of Donald Trump’s reasoning behind Greenland and his answer to China This is excellent

China analyst Melissa Chen gives the absolute best explanation of Donald Trump’s reasoning behind Greenland and his answer to China This is excellent

Wall Street Apes

165,701 views • 9 months ago

Whatever AI sceptics say, LLMs really can reason. They're not just doing an imitation that looks like reasoning, it's the real deal. But even though they are able to reason, sometimes they won't! If you ask an LLM a question it can't answer, sometimes it will just try to imitate reasoning without doing it. The chain of thought looks basically indistinguishable from actual reasoning. But under the hood something very different is going on. Trenton Bricken talked with me about what work on circuits inside LLMs has revealed:

Whatever AI sceptics say, LLMs really can reason. They're not just doing an imitation that looks like reasoning, it's the real deal. But even though they are able to reason, sometimes they won't! If you ask an LLM a question it can't answer, sometimes it will just try to imitate reasoning without doing it. The chain of thought looks basically indistinguishable from actual reasoning. But under the hood something very different is going on. Trenton Bricken talked with me about what work on circuits inside LLMs has revealed:

Dwarkesh Patel

105,153 views • 5 months ago

This was not a one shot. This is the result of an extended collaboration with Opus 5.5 this is only the beginning of what they can do. now presenting, THE_VALLEY from the DOORS album by ABSTRACTWEAPON

This was not a one shot. This is the result of an extended collaboration with Opus 5.5 this is only the beginning of what they can do. now presenting, THE_VALLEY from the DOORS album by ABSTRACTWEAPON

Ra

21,714 views • 1 month ago

This is what current SOTA, Opus 5.5 makes btw The progress is INSANE, all the physics and animations are done in a custom fully opus-made engine.

This is what current SOTA, Opus 5.5 makes btw The progress is INSANE, all the physics and animations are done in a custom fully opus-made engine.

bluedev

14,295 views • 22 days ago

“We found that if you ask the LLM, surprisingly it always says that I'm 100% confident about my reasoning.” Chirag Agarwal examines the (un)reliability of chain-of-thought reasoning, highlighting issues in faithfulness, uncertainty & hallucination.

“We found that if you ask the LLM, surprisingly it always says that I'm 100% confident about my reasoning.” Chirag Agarwal examines the (un)reliability of chain-of-thought reasoning, highlighting issues in faithfulness, uncertainty & hallucination.

FAR.AI

2,248,454 views • 1 year ago

This is 5 days of agent runs with Opus 5.5 a pianist playing a grand piano in a sunset lake, and the song is an original that Opus 5.5 composed itself sound on :)

This is 5 days of agent runs with Opus 5.5 a pianist playing a grand piano in a sunset lake, and the song is an original that Opus 5.5 composed itself sound on :)

Leon Lin

31,805 views • 15 days ago

llama.cpp is likely the first LLM runtime in the world to allow "interrupt" reasoning without stopping the whole response. We also added a small "skip" button on the Web UI, the model gives the final response as soon as you click the button. The response is no longer bound to reasoning budget!

llama.cpp is likely the first LLM runtime in the world to allow "interrupt" reasoning without stopping the whole response. We also added a small "skip" button on the Web UI, the model gives the final response as soon as you click the button. The response is no longer bound to reasoning budget!

Xuan-Son Nguyen

31,511 views • 5 months ago

Forget Muse or Meta. Opus 5.5 or Fable 5.5 whatever this is, its AGI, even if its OPUS, its AGI, it literally can play ELECTRIC GUITAR. it can do all of it and this can even be sampled. THIS IS ALL code from an LLM's mind. Listen fully, it keeps getting better

Forget Muse or Meta. Opus 5.5 or Fable 5.5 whatever this is, its AGI, even if its OPUS, its AGI, it literally can play ELECTRIC GUITAR. it can do all of it and this can even be sampled. THIS IS ALL code from an LLM's mind. Listen fully, it keeps getting better

Sam AlbertX

18,712 views • 28 days ago

This is made by Opus 5.5 high. sfx improved

This is made by Opus 5.5 high. sfx improved

Leon Lin

41,059 views • 28 days ago

An autonomous agent is only as safe its reasoning. ThoughtProofAI verifies that reasoning before the agent acts - catching a flawed decision before it costs anything. Don't trust. Verify. Learn exactly how 👇

An autonomous agent is only as safe its reasoning. ThoughtProofAI verifies that reasoning before the agent acts - catching a flawed decision before it costs anything. Don't trust. Verify. Learn exactly how 👇

GOAT Network

72,279 views • 4 months ago

Switching the model or reasoning mid-session can be expensive

Switching the model or reasoning mid-session can be expensive

Alex Sidorenko

62,177 views • 3 months ago