Garrett Bingham
@gjb_ai
Senior Research Scientist @ Google DeepMind | Gemini Deep Think | AI IMO Gold
"It was a team effort" sounds cliché, but it really was! We had so many late-breaking super critical contributions. What a magical experience. 🥇
Right before #imo2025, together with colleagues from Mountain View, NYC, Singapore, etc, we all gathered at @GoogleDeepMind headquarter in London for our final push for IMO. I believe that week was when all magic happened! We put all individual recipes (that we figured out…
An advanced version of Gemini with Deep Think achieved a gold medal at this year's IMO! 🥇 So many amazing contributions from the team. What's next? deepmind.google/discover/blog/…
Attached is a full proof by Gemini 2.5 Pro #DeepThink with our experts' comments drive.google.com/file/d/1PaKXo4…. Here I quote a few important moments in the proof: (a) Expert: The main part of the solution begins with a "proof by contradiction", which is a reasonable choice considering…
Working on getting this in everyone's hands soon. Stay tuned! :)
Deep Think in 2.5 Pro has landed. 🤯 It’s a new enhanced reasoning mode using our research in parallel thinking techniques - meaning it explores multiple hypotheses before responding. This enables it to handle incredibly complex math and coding problems more effectively.
More to come soon!
Gemini 2.5 Pro Deep Think has the highest score across many of the hardest benchmarks for maths, coding, and multimodal reasoning.
It was amazing working on this. How long til we can solve unsolved problems?
Yesterday at #GoogleIO, we introduced Gemini 2.5 Pro Deep Think 🧠, pushing the frontiers of AI reasoning. This enhanced reasoning mode is built to tackle drastically complex problems – like USAMO problems that stumped previous models. Super proud of the GDM team for this one!
2.5 Pro Deep Think is an incredibly smart model. Some of the benchmark results, simply put were surprising to me. But, the benchmarks don’t tell the whole story. It can go into far more intricate details, especially open-ended prompts, unlike any of our previous thinking models.
Gemini 2.5 Pro Deep Think is an SVG artist! Prompt: "Draw a SVG of a Pelican riding a bicycle" Left: Gemini 2.5 Pro Right: Gemini 2.5 Pro Deep Think Credit: simonwillison.net/2024/Oct/25/pe…
It's been a wild run over the last 2 weeks to land Deep Thinking just in time for I/O! Heroic efforts across many teams @GoogleDeepMind from research to production that felt almost impossible last month! Proud of our research team for the relentless effort over the past 6 months…
Deep Think in 2.5 Pro has landed. 🤯 It’s a new enhanced reasoning mode using our research in parallel thinking techniques - meaning it explores multiple hypotheses before responding. This enables it to handle incredibly complex math and coding problems more effectively.
Gemini 2.5 Pro Deep Think!!! Truly crazy how a handful of people can have a random idea, implement it, and a few months later it becomes the world's smartest AI. What a wild experience.
Amazing progress in reasoning! 🚀 Gemini 2.5 Pro Deep Think hitting 49.4% on USAMO – a feat I'd have considered impossible just a couple of years ago – & Gemini 2.5 Flash achieving 1424 Elo are huge leaps. So proud our team's research ideas contributed to this moment!…
Looking for PhD candidates to do an internship at @GoogleDeepMind in early 2025. Consider applying if you have experience in LLM reasoning, a strong publication track record and/or experience with top-tier math competitions (IMO, etc.) Form: forms.gle/cyhefKKKz4dRph…
I tell my mom it's basically like my 3rd grade science fair, only marginally more important. Come say hi if you're at #ECCV2024!
In need of some light reading? My PhD Dissertation received an Honorable Mention for the 2024 SIGEVO Best Dissertation Award! It's only 233 pages ;) Award: sig.sigevo.org/index.html/tik… Dissertation: arxiv.org/abs/2304.03374
garrettbingham.com --> gjb.ai After eight years I finally got around to updating my personal website. It even has scrollable galleries where you can see all of the figures and tables from each of my research papers. Let me know what you think :)
One thing that is interesting in our work is the use of generated images, which allow us to capture much more interesting scenarios. Below is a 2D visualization of #HaloQuest images: real images occupy a similar semantic distribution to VQA v2 images, while the synthetic images…
We're open-sourcing the #HaloQuest dataset to help reduce hallucination and improve reasoning in vision-language models!
Introducing #HaloQuest, our latest effort towards improving hallucination in multimodal foundation models! We hope that HaloQuest, as both a challenging benchmark and an open-sourced dataset, will enable the field's progress in advanced reasoning! Also glad that HaloQuest played…
It was an amazing experience to help lead natural language reasoning efforts in this project. More to come soon :)
We’re presenting the first AI to solve International Mathematical Olympiad problems at a silver medalist level.🥈 It combines AlphaProof, a new breakthrough model for formal reasoning, and AlphaGeometry 2, an improved version of our previous system. 🧵 dpmd.ai/imo-silver