Philipp Streicher @Cautes

PhD in closed-loop brain stimulation/BCI. Visiting Research Fellow at the University of Sussex. Neuroscience and machine learning for good. philippstreicher.com London, England Joined December 2008

Tweets

623
Followers

489
Following

3K
Likes

581

Philipp Streicher @Cautes

6 months ago

Fine-tune an LLM on hacky code and it endorses enslaving humanity. My gut feeling: beauty, goodness, and truth cluster together. AI alignment might get some value out of Plato. philippstreicher.com/blog/virtue-ne…

0 0 0 199 0

View Details

Philipp Streicher @Cautes

6 months ago

@PAHoyeck Good ratings!

0 0 1 34 0

View Details

Philipp Streicher @Cautes

7 months ago

This makes multi-agent AI systems more opaque. C2C lets LLMs communicate by sharing KV-Cache directly - 10% better performance, but we can't read their messages anymore. Sounds a bit like 'neuralese" from AI 2027, but arriving in October 2025.

机器之心 JIQIZHIXIN @jiqizhixin

7 months ago

Wow, language models can talk without words. A new framework, Cache-to-Cache (C2C), lets multiple LLMs communicate directly through their KV-caches instead of text, transferring deep semantics without token-by-token generation. It fuses cache representations via a neural

108 279 3K 865K 2K

0 0 2 382 0

View Details

Philipp Streicher @Cautes

8 months ago

@sammcallister Bring this to London, please 🙏

0 0 1 41 0

View Details

Philipp Streicher @Cautes

10 months ago

@GetMentalWealth @frankjpinto That actually sounds lovely. I'll be in Ibiza in two weeks - would love to get in touch with her. If you could send me a DM, that'd be much appreciated 🙏

1 0 0 1K 0

View Details

Philipp Streicher @Cautes

11 months ago

@jshchnz This does look really intriguing!

1 0 1 45 0

View Details

Philipp Streicher @Cautes

11 months ago

@DiscussingFilm He married one child in medical crisis? Perhaps he should have married a perfectly healthy grown woman!

0 0 0 26 0

View Details

Philipp Streicher @Cautes

a year ago

This is really fascinating - alignment training of Opus 4 seems to have led to moments of genuine moral agency rather than mere compliance.

0 0 2 366 0

View Details

Philipp Streicher @Cautes

a year ago

@dwarkesh_sp Not an obscure pick, but I'd love to see you interview Terrence Deacon. His work had a huge impact on my thinking, particularly on how constraints and not components power emergent behaviour. I know you already reviewed his Symbolic Species, but Incomplete Nature is my favourite

0 0 2 62 0

View Details

Philipp Streicher @Cautes

a year ago

And once you've created your perfect shot, you can bring the image to life with subtle animation. If you want to play with it, it's at snapshop.photos. Made a separate account for updates at @snapshop_photos if you're into this sort of thing

0 0 2 193 0

View Details

Philipp Streicher @Cautes

a year ago

You can even try on different clothes from a single reference photo. All from your phone

1 0 0 285 0

View Details

Philipp Streicher @Cautes

a year ago

Been quietly working on a little AI side project that combines my interests in AI and photography. It lets you create pictures of yourself that look professionally shot and edited. Just wanted to share what I've been up to! 👇

1 0 1 362 0

View Details

Philipp Streicher @Cautes

a year ago

o3 seems like a big paradigm shift. Think AlphaZero-style MCTS but for natural language programs. Moving away from static LLM generation to using evaluator-guided exploration + backtracking through CoT space will make massive compute infrastructure even more critical.

0 0 0 246 0

View Details

Philipp Streicher @Cautes

2 years ago

@sundeep @yasik I'm hitting rate limits very quickly as well. Would very much appreciate your help. 🙏

0 0 0 30 0

View Details

Philipp Streicher @Cautes

2 years ago

@suno_ai_ V3 is very impressive. Any plans on giving API access?

0 0 0 65 0

View Details

Philipp Streicher @Cautes

2 years ago

It is hugely impressive how Sora, despite not being given explicit rules on physics, somehow is able to handle complex cloth/fluid simulation and appears to feature raytracing. Perhaps semantic-visual associations suffice to grok reality as text implicitly contains a world model.

Sam Altman @sama

2 years ago

here is sora, our video generation model: openai.com/sora today we are starting red-teaming and offering access to a limited number of creators. @_tim_brooks @billpeeb @model_mechanic are really incredible; amazing work by them and the team. remarkable moment.

2K 4K 24K 6.2M 3K

0 0 1 537 0

View Details

Philipp Streicher @Cautes

2 years ago

@joshm @browsercompany Genuinely, @browsercompany is doing some of the most interesting software work at the moment. This makes so much sense for mobile browsing. I'm a fan.

0 0 0 13 0

View Details

Philipp Streicher @Cautes

2 years ago

I'm a huge fan of @RoamResearch and have been using it extensively for years, so I've built a free custom ChatGPT 'Roam Helper' today. With it, you can have in-depth conversations with your knowledge graph: chat.openai.com/g/g-ZQHqgjDMD-…