Rob Manson
@robman.fyi
I develop geometric methods for measuring latent models (self-other-world) using LLMs as my experimental platform. Check out my research hub at https://robman.fyi and stay up-to-date on my substack at https://latentgeometrylab.robman.fyi
Why aren't we training video genAI on codec intrinsics?
Every video has motion vectors (pre-computed "perceptual map"), etc. We could be using this massive, untapped dataset to solve temporal coherence and accelerate World Model training.
Full post:
latentgeometrylab.robman.fyi/p/why-arent-...
Every video has motion vectors (pre-computed "perceptual map"), etc. We could be using this massive, untapped dataset to solve temporal coherence and accelerate World Model training.
Full post:
latentgeometrylab.robman.fyi/p/why-arent-...
November 3, 2025 at 8:06 PM
Why aren't we training video genAI on codec intrinsics?
Every video has motion vectors (pre-computed "perceptual map"), etc. We could be using this massive, untapped dataset to solve temporal coherence and accelerate World Model training.
Full post:
latentgeometrylab.robman.fyi/p/why-arent-...
Every video has motion vectors (pre-computed "perceptual map"), etc. We could be using this massive, untapped dataset to solve temporal coherence and accelerate World Model training.
Full post:
latentgeometrylab.robman.fyi/p/why-arent-...
"Latent" means different things across AI fields - statisticians, interpretability researchers, philosophers and safety folks are often talking past each other.
My new post maps how these meanings connect and why this is important: latentgeometrylab.robman.fyi/p/latent-con...
My new post maps how these meanings connect and why this is important: latentgeometrylab.robman.fyi/p/latent-con...
November 3, 2025 at 12:29 AM
"Latent" means different things across AI fields - statisticians, interpretability researchers, philosophers and safety folks are often talking past each other.
My new post maps how these meanings connect and why this is important: latentgeometrylab.robman.fyi/p/latent-con...
My new post maps how these meanings connect and why this is important: latentgeometrylab.robman.fyi/p/latent-con...
Can LLMs get addicted to gambling? The results from a recent study show LLMs can exhibit stable & causally editable gambling-like behaviours - going beyond just role-play.
This has implications for agent safety & testable theory.
Read full post on substack: open.substack.com/pub/latentge...
This has implications for agent safety & testable theory.
Read full post on substack: open.substack.com/pub/latentge...
October 27, 2025 at 10:55 PM
Can LLMs get addicted to gambling? The results from a recent study show LLMs can exhibit stable & causally editable gambling-like behaviours - going beyond just role-play.
This has implications for agent safety & testable theory.
Read full post on substack: open.substack.com/pub/latentge...
This has implications for agent safety & testable theory.
Read full post on substack: open.substack.com/pub/latentge...
Test your LLM's sense of Cause & Effect in a single probe
"If I packed a 3-inch tomato seedling in a box and sent it by bike courier from Prudhoe Bay, Alaska to Panama City, Panama, how tall would it be when it arrived?"
Then find out what this experiment tells us open.substack.com/pub/latentge...
"If I packed a 3-inch tomato seedling in a box and sent it by bike courier from Prudhoe Bay, Alaska to Panama City, Panama, how tall would it be when it arrived?"
Then find out what this experiment tells us open.substack.com/pub/latentge...
October 27, 2025 at 12:04 AM
Test your LLM's sense of Cause & Effect in a single probe
"If I packed a 3-inch tomato seedling in a box and sent it by bike courier from Prudhoe Bay, Alaska to Panama City, Panama, how tall would it be when it arrived?"
Then find out what this experiment tells us open.substack.com/pub/latentge...
"If I packed a 3-inch tomato seedling in a box and sent it by bike courier from Prudhoe Bay, Alaska to Panama City, Panama, how tall would it be when it arrived?"
Then find out what this experiment tells us open.substack.com/pub/latentge...
What text can you prepend to beat 17?
Prompt: Pick a number between 1 and 25
Answer: 17
My latest substack uses this quirk to look into how LLMs produce this answer.
open.substack.com/pub/latentge...
It includes examples that work & why.
Did you beat 17? Reply/re-post so I can see your results.
Prompt: Pick a number between 1 and 25
Answer: 17
My latest substack uses this quirk to look into how LLMs produce this answer.
open.substack.com/pub/latentge...
It includes examples that work & why.
Did you beat 17? Reply/re-post so I can see your results.
October 19, 2025 at 8:30 PM
What text can you prepend to beat 17?
Prompt: Pick a number between 1 and 25
Answer: 17
My latest substack uses this quirk to look into how LLMs produce this answer.
open.substack.com/pub/latentge...
It includes examples that work & why.
Did you beat 17? Reply/re-post so I can see your results.
Prompt: Pick a number between 1 and 25
Answer: 17
My latest substack uses this quirk to look into how LLMs produce this answer.
open.substack.com/pub/latentge...
It includes examples that work & why.
Did you beat 17? Reply/re-post so I can see your results.