@uljadb99

AI PhD @aims_oxford and @bold_lab_ai, SR @GoogleDeepMind, ex RS @JPMorgan AI, EE with Great Distinction @nyuniversity, Tiramisu and Comedy enthusiast

 London, England
uljad.com
Joined October 2021
  • Tweets542
  • Following553
  • Followers513
32Media
INTENT FACTORED FOR THE WIN! Good science that works always finds a way #ifftw we're going to NeurIPS. Fresh, improved and updated IFG paper dropping soon 👀
Uljad@uljadb9918 Jul 2025
Unlock real diversity in your LLM! 🚀 LLM outputs can be boring and repetitive. Today, we release Intent Factored Generation (IFG) to: - Sample conceptually diverse outputs💡 - Improve performance on math and code reasoning tasks🤔 - Get more engaging conversational agents 🤖
3
2
19
1,706
Peeling away ideas like the fixation with being on-policy at all costs
Nassim Nicholas Taleb
@nntaleb1 Sep 2021
There is a phase in life when you experience the thrill of learning things. And then, this second one when you figure out how to unlearn things, discarding beliefs with the same secret pleasure you get peeling scabs, picking your nose or squeezing blisters.
76
Struggled with this a few weeks ago, totally oblivious that all of this was happening. Overall great news! Also Something funny about the failed training runs from off-policy sweeps looking similar
Martin Marek
@mrtnmOct 25
Tinker's new StableReinforce objective is Reinforce + Score Centering. Here's how I reverse-engineered it 🧵
5
405
It's about this formula: E_q[C * ∇ logp] = E_q[C] * E_q[∇ logp] + Cov_q(C, ∇ logp) E_q[C] * E_q[∇ logp] is the drift term. We want it to be 0 Score-Centering works on the 2nd term: E_q[∇ logp - E_q[∇ logp]] = 0 1. we can do MC, like thinky 2. that's not the only term
2
59
5,390
Jordi and team are great to work with!
Jordi Grau-Moya@jordigraumoOct 23
Hiring PhD student (London, 5 months) at Google DeepMind. Topic: Steering multi-agent LLM economies via mechanisms for enabling reusable knowledge work. Looking for strong phd students in Multi agent LLMs, MARL, or mechanism design. Send cv to [email protected]
6
1,243
Hiring PhD student (London, 5 months) at Google DeepMind. Topic: Steering multi-agent LLM economies via mechanisms for enabling reusable knowledge work. Looking for strong phd students in Multi agent LLMs, MARL, or mechanism design. Send cv to [email protected]
9
16
2
299
46,645
We are releasing our first model, Flint, a best-in-class local OSS coding model that runs at 150+ tokens/s on an @nvidia DGX Spark and 125+ tokens/s on a MacBook Pro M4—developed from scratch in less than four months with help from @Recursive_SI's autoresearch system.
Richard Socher
@RichardSocherOct 21
Introducing Flint, @Recursive_SI’s first model: a fast local coding model built with our automated research system, the Eureka Machine. • Best-in-class fast local coding • Up to 2x faster than comparable models: 125+ tokens/sec on a MacBook Pro, 150+ on RTX Spark • Built from scratch in under 4 months Private, unmetered intelligence on your own machine. Open weights available now on @huggingface, run it locally on your favorite framework or try it for free on @opencode or @pidotdev.
24
29
8
320
20,049
Knowing when to stop is one of the most important questions of our time
Arber Zela@ZelaArberOct 20
A looped transformer has to know when to stop, and its state has to stay bounded while it loops. We introduce Spherical Fixed-Point Reasoners (SFPRM) 🌐: loop on the unit sphere, where each step is an angle, and stop once the angle is small enough. 🧵 (1/9)
1
23
960
The loop routine becomes 4 steps on the sphere: 1. Spherical linear interpolate (Slerp) a fraction toward the input 2. Apply the nGPT blocks 3. Rotate a fraction toward the proposal 4. Halt once the angle between state and proposal is below some threshold for every token. (3/9)
1
1
10
1,317
Effective immediately: 👑 MPC: Model Predictive Control 🔌 API: Agent Peripheral Interface (formerly Model Context Protocol) 🛠️ GCC: Group Communication & Coordination (formerly Multi-Party Communication)
Brandon Amos
@brandondamosOct 12
we keep one and rename the ohters
63%MPC
21%MCP
16%MPC
19 votes • Final results
2
3
25
2,954
Argon solved Minecraft.
6
305
A groundbreaking Frontier result! @maxencefaldor and @robertarail did a phenomenal job in showing the Argon capabilities so effectively. What a model and what a team in every sense of the word!
Maxence Faldor
@maxencefaldorOct 16
Four Gemini 4 Argon agents, one shared Minecraft world, one goal: defeat the Ender Dragon. 27 hours and 2,783 self-written programs later, the dragon fell.

A team of Gemini 4 Argon agents beats Minecraft

Article · @maxencefaldor
1
13
772
Four Gemini 4 Argon agents, one shared Minecraft world, one goal: defeat the Ender Dragon. 27 hours and 2,783 self-written programs later, the dragon fell.

A team of Gemini 4 Argon agents beats Minecraft

Article · @maxencefaldor
57
69
75
634
278,801
Four Gemini 4 Argon agents teamed up in Minecraft to defeat the Ender Dragon in 27 hours, writing every move as code and coordinating in the game's chat. Great work led by @maxencefaldor with Satinder Singh. Read more in our technical report: arxiv.org/pdf/2610.22319.
Maxence Faldor
@maxencefaldorOct 16
Four Gemini 4 Argon agents, one shared Minecraft world, one goal: defeat the Ender Dragon. 27 hours and 2,783 self-written programs later, the dragon fell.

A team of Gemini 4 Argon agents beats Minecraft

Article · @maxencefaldor
5
11
2
117
5,338
🤖 Made with AI
9
27
2
521
17,381
Will soon release new methods to better describe motion and instantaneous change. Finalizing notation atm
Brandon Amos
@brandondamosOct 9
📈 ignore the empiricists and come check out our linear models
1
6
675
Experts doubted a small neural net could beat Craftax without an LLM's knowledge. ​ Our agent meta-harness built one in 3 weeks. The 15M-param agent beats the final boss 44% of the time, costing a cent, and 10 seconds. ​ GPT-6 Astra won 1%, taking 24 hours and $83. 🧵
33
75
12
734
273,306
Uljad retweeted
sankalp
@dejavucoderOct 6
your agent after it's 50th compaction
19
197
14
3,745
68,100
Ben taught me so much about doing proper research in RL and the ways to debug policy gradients. Glad the Science of RL team and @reflection_ai gets to have this expertise for their models!
Ben Ellis@benjamin_ellis3Oct 5
Being part of the Science of RL team for Beam has been an absolute privilege. It's been wonderful to learn how the big RL sausage gets made over the last year and a half and witness the dedication and pace of this team. I look forward to building many more models in the future!
1
1
15
940
This is great news! Congratulations to the team!
Reflection
@reflection_aiOct 5
Introducing Beam: a highly efficient agentic open model with 501B total parameters and 23B active. - Frontier reasoning efficiency - Advances the Western open frontier on coding & agentic tasks - Trained end-to-end from scratch Full weights release this month. Learn more about Beam: reflection.ai/beam
10
210
Introducing Beam: a highly efficient agentic open model with 501B total parameters and 23B active. - Frontier reasoning efficiency - Advances the Western open frontier on coding & agentic tasks - Trained end-to-end from scratch Full weights release this month. Learn more about Beam: reflection.ai/beam
566
1,098
757
8,599
2,516,366
Load more