learning with yacine show
by yacine mahdid
Latest episode
Next-Latent Prediction Transformers Learn World Models feat. Jayden…
Next-Latent Prediction Transformers Learn World Models feat. Jayden…
About this podcast
a show about untangling the complexity of deep learning and the brain by interviewing smart researchers and builders in the field of intelligence!
ideal for scholars, industry practitioners or toddlers.
www.yacinemahdid.com
Episodes
- Next-Latent Prediction Transformers Learn World Models feat. Jayden Teoh from MIT · 2 h 8 minit’s very interesting to see where next-token transformers still fall short (like the path-star graph problems) while crushing problems that look even more daunting.I had the…
- Prime Agent: Self-Improving RLM Harness with Seth Karten from Prime Intellect · 2 h 36 minRLM + continual harness + multi-agent communication is what compose the prime agent harness developed by prime intellect which now sits at a gazillion stars on github (22.3K).I…
- How VPO Trains LLMs to Excel at Test-Time Search with Ryan Bahlous-Boldi from MIT · 2 h 4 minusually post-training with GRPO will straight up collapse the model onto a tight set of potential answers (i.e. the spooky sharpening of the distribution).which in some cases…
- What Continual Learning Actually Means feat. Ilija Lichkovski from Prime Intellect · 1 h 35 mincontinual learning is one hell of a hot topic in 2026 and it’s beautiful because (like consciousness) barely anyone agree on what are we even talking about.this is why earlier…
- RLMs are Compositional Generalizers with Alex Zhang from MIT · 1 h 35 minthis discussion I had with alex earlier this year re-shaped a lot how I’m seeing these new breed of frontier models that are specialist at coding + multi-agent + long running…
- Inside a Frontier Open Model: Nemotron 3 Ultra Explained ft. NVIDIA's Chris Alexiuk · 1 h 31 minopen models are infinitely interesting to me. being able to peek under the hood and look at the interconnected pieces that each hums and work in a specific way to feed the whole…
- How to Generate Synthetic Pretraining Data? with Joël Niklaus from Hugginface · 1 h 30 mintoday we are taking a deeper look at of the secret behind LLM pretraining which is synthetic data pipelines with Joel Niklaus machine learning engineer at Hugging Face!joel an…
- Pretrained LLMs Are Surrounded by Task Experts feat Yulu Gan from MIT · 1 h 12 minsometime there is an idea that just makes other weird research results make so much more sense and sort of cascade to explain a whole bunch of phenomenon. neural thicket from the…
- How to Measure Frontier LLM? with Florian Brand from Prime Intellect · 1 h 54 minBenchmarking LLM in 2026 is now a high stake task that requires deep expertise on LLM architecture, Prompt Engineering, Harness, Sandbox, Scoring, Hardware, Serving, and more…
- How to Learn in the AI Age? with Ludwig & Lazarz · 2 h 7 minearlier this year I took two of my friends (ludwig and lazarz) on a big ol' discussion on how to leverage ai for learning purposes in general. what I like about these two lad is…
- Why DAGs Are Taking Over Auto-Research (w/ the Founders of Paradigma) · 1 h 37 minauto-research is starting to gain traction as a very viable paradigm for creating useful research discovery.now, that paradigm is still in its infancy and the infrastructure to…
- Why Self-Distillation Is Taking Over LLM Post-Training (w/ the Researchers Behind It) · 1 h 31 minI had an awesome time interviewing @IdanShenfeld and @jonashubotter from MIT and ETH Zurich about self-distillation. this very promising post-training paradigm where the model…
- The Science of Learning Math (and Anything Else) with Justin Skycak · 3 h 18 minhey folks first time posting this sort of interview as a podcast (shout out the three anon on twitter that requested it). now that I think of it the format does make sense since…







