Inference, Diffusion, World Models, and More | YC Paper Club

ℹ️ EPISODE INFO

Released

May 28, 2026

Duration

1h 7m

Channel

Y Combinator

Watch on YouTube

▶ Open ↗

📝 EPISODE DESCRIPTION

Even if you’re a current PhD student, it's hard to keep up with the latest AI research. That's why we started YC Paper Club, a small group of researchers, engineers, and founders who will meet every two weeks this summer to present and discuss new papers together. This was from our very first discussion group on May 20th, 2026, at the YC office in Mountain View, CA. Thanks to the following presenters: 0:12 - Intro from YC Visiting Partner Francois Chaubard 3:49 - Tanishq Kumar — Speculative Speculative Decoding (https://arxiv.org/abs/2603.03251) 18:33 - Guangyao (Stannis) Zhou — Diffusion-MPC (https://arxiv.org/abs/2410.05364) 30:26 - Isaac Ward — LeWorldModeling (https://arxiv.org/abs/2603.19312) 43:54 - Akshay Vegesna — Deep Learning is Not So Mysterious or Different (https://arxiv.org/abs/2503.02113) 51:24 - Konwoo Kim — Pretraining Under Infinite Compute (https://arxiv.org/pdf/2509.14786)

🎤 SPEAKERS

• Francois Chaubard

  host

  Host of YC Paper Club who emcees the session, introduces presenters, and runs Q&A.

• Tanishq Kumar

  guest

  Stanford graduate student presenting work on speculative decoding and high-throughput LLM inference.

• Guangyao (Stannis) Zhou

  guest

  Staff Research Scientist at Google DeepMind presenting world-modeling and diffusion/planning work for robotics.

• Isaac Ward

  guest

  Researcher presenting a paper and discussion on world models and related design considerations.

• Akshay Vegesna

  guest

  Q Labs co-founder/president presenting Andrew Gordon Wilson’s paper on deep learning generalization and inductive bias.

• Konwoo Kim

  guest

  Researcher and paper co-lead presenting work on pre-training, scaling, and data efficiency under infinite compute.

🧾 EPISODE SUMMARY

In this episode of Y Combinator, featuring Francois Chaubard and Tanishq Kumar, Inference, Diffusion, World Models, and More | YC Paper Club explores YC Paper Club debuts: faster inference, diffusion control, and scaling laws The event frames inference speed as a future capability bottleneck—not just a cost issue—because faster tokens-per-second enable more test-time “thinking” and higher delivered intelligence.

🔗 RELATED EPISODES

How Legora Went From YC to $100M ARR in 18 Months

Conductor CEO Charlie Holtz Walks Us Through His AI Coding Setup

How to Build an AI-Native Services Company

Inside YC's AI Playbook

How The Best Companies Defend Against Mediocrity And Rot

Zepto: How Two 17-Year-Olds Built India's Largest Seller Of Fruits and Vegetables