Alex Havrilla (@Dahoas1) / X

Alex Havrilla

251 posts

Alex Havrilla

@Dahoas1

Research Scientist @GoogleDeepMind. Interested in interestingness. Opinions my own.

Joined August 2021

Alex Havrilla
@Dahoas1
Aug 22, 2025
Life update: in June I successfully defended my PhD dissertation "Towards a Theory and Practice of Open-Ended Reasoning with Generative Models". Now, I'm so excited to be joining Google DeepMind to continue exploring the limits of AI discovery and creativity.
35K
Alex Havrilla
@Dahoas1
Sep 15, 2022
Excited to announce I am a recipient of @StabilityAI PhD fellowship! Excited to continue working on open source machine learning research with their support.
Alex Havrilla
@Dahoas1
Mar 8, 2024
🚨🚨🚨Paper #2 from my time at Meta! In this work, we set out to understand how different algorithms fare at improving LLM reasoning from feedback. We compare expert iteration, PPO, and return-conditioned RL using Llama-2 as the base model.
81K
Alex Havrilla
@Dahoas1
Jun 19, 2025
Excited to announce the final paper of my PhD!📢 A crucial piece of SFT/RL training is the availability of high-quality problem-solution data (Q, A). But what to do for difficult tasks where such data is scarce/hard to generate with SOTA models? Read on to find out
27K
Alex Havrilla
@Dahoas1
Apr 24, 2024
In my humble opinion the recent Stream of Search paper (arxiv.org/abs/2404.03683) is truly outstanding. Everyone should give it a thorough read.
22K
Alex Havrilla
@Dahoas1
Dec 5, 2024
How important is the quality, diversity, and complexity (QDC) of synthetic data for LLM performance? What effect does QDC data composition have on self-improvement? We just released a comprehensive survey discussing these questions (and many more) 🧵
17K
Alex Havrilla
@Dahoas1
Feb 20, 2024
New paper alert🚨🚨🚨 How to bootstrap the reasoning refinement capabilities of LLMs using synthetic data? Introducing "GLoRe: When, Where, and How to Improve LLM Reasoning via Global and Local Refinements". Applied on GSM8K we can improve a strong RL finetuned LLama-2 13B by 12%
21K
Alex Havrilla
@Dahoas1
Nov 21, 2024
I'll be at this year's Neurips presenting my new work!: Understanding Scaling Laws with Statistical and Approximation Theory for Transformer Neural Networks on Intrinsically Low-dimensional Data
5K
Alex Havrilla
@Dahoas1
Jul 21, 2024
I'm at ICML presenting GLoRe (arxiv.org/abs/2402.10963) and Teaching Reasoning with RL (arxiv.org/abs/2403.04642)! If you'd like to chat about synthetic data, process-based rewards, open-endedness, or theoretical foundations of scaling laws (or anything else) my DMs are open!
arxiv.org
GLoRe: When, Where, and How to Improve LLM Reasoning via Global...
State-of-the-art language models can exhibit impressive reasoning refinement capabilities on math, science or coding tasks. However, recent work demonstrates that even the best models struggle to...
5.8K
Alex Havrilla
@Dahoas1
Feb 8, 2024
New paper! What should you do when high resolution data is not available for training? Introducing Dual Fourier UNet (DFU): scale-robust diffusion model for zero-shot super-resolution image generation arxiv.org/abs/2401.06144
5.1K
Alex Havrilla
@Dahoas1
Mar 27, 2023
Replying to @JagersbergKnut
Improved communications coming soon🙂
2K
Alex Havrilla
@Dahoas1
Jul 23, 2024
Presenting now at number 2716!
1.7K
Alex Havrilla
@Dahoas1
Apr 18, 2024
Had a great time during our discussion, thanks again for having me!
The TWIML AI Podcast
@twimlai
Apr 17, 2024
Today we're joined by @Dahoas1 from @GeorgiaTech to discuss the reasoning capability of language models and the potential to improve it with traditional RL methods 🎧 / 🎥 Listen to the episode at: twimlai.com/go/680. 📖 CHAPTERS 00:00 - Introduction 02:19 - RL vs RLHF
00:00
2.6K
Alex Havrilla
@Dahoas1
Aug 22, 2025
Replying to @Dahoas1
I grew a lot as a researcher over the last four years. So many thanks to my amazing advisor, internship mentors, and everyone else I met along the way!
1.7K