david yan (@dzyan01) profile picture

phd student @PrincetonVL, research intern @elorianai. formerly: bse @princetoncs

Joined March 2025
Grounded reasoning is a popular thinking strategy for VLMs. But do models even use their grounding? (spoiler: not always!) In our @ElorianAI paper, we dig into the grounded reasoning recipe and release BrickVision, a benchmark that’s challenging to solve without grounding. 1/7
1
6
1
48
4,210
On BrickVision, existing models vary widely in performance! 6/7
2
1
165
super cool use of infinigen for zero-shot sim2real!
We just released HALO — Perceptive Humanoid Motion from Simulation. Trained entirely on synthetic data from 1,000 procedurally generated homes. Zero-shot deployment in 6 unseen real environments. 91% success rate across 360 physical trials. Fully onboard. Paper, code, and project page are all live. apple-aiml-research.github.i…
1
11
920
Random thought on "why do a (ml/ai) phd?" I believe AI has shortened one's optimization horizon, due to high future uncertainty. As such, you should be confident that you will get positive short-term value, instead of indexing on the eventual delayed reward from finishing.
8
5
1
101
8,362
"Personal reward" here refers to some abstract net reward relative to other options (e.g. working in industry). Person b is not necessarily unhappy or sad, but just could be happier elsewhere! In contrast, doing a PhD is locally optimal for person a.
3
861
Excited to announce a major performance update for Infinigen! We've overhauled various assets and systems, focusing on efficiency. Our `--performance` config reduces the average generation time of nature scenes by 8x without sacrificing any visual fidelity! 1/3
8
15
2
120
9,484
All scenes are generated from fully procedural code, so no external assets are required. We hope you (or your agent) will give it a try at github.com/princeton-vl/infi…! For more updates, follow @princetonvl. This performance update was led by @jiadeng and I. 2/3
2
2
13
684
To reproduce the demo scene, just run python -m infinigen_examples.generate_nature --seed 2 --task coarse populate fine_terrain --output_folder out/mountain_2 --configs mountain noisy_video performance -p "execute_tasks.frame_range=[1,10]" land_process_pool.enabled=True 3/3
5
430
my bitter lesson is finding out that caffeine significantly increases my focus and productivity
6
1
70
4,040
david yan retweeted
Sprout Jackson
3
5
2
49
5,930
david yan retweeted
🚨 Stop feeding your world model the same boring demos. Switch to free-range play data! 🚨 Our CoRL 2026 paper finds we can optimize autonomous data collection to focus on missing action consequences, achieving up to 85x the data scaling efficiency of expert data collection!🧵
8
33
9
154
16,166
some things never change! from HN thread on CycleGAN in 2017.
2
22
3,158
david yan retweeted
I spent a decade building foundation models at Google and Apple. But what our small team at Elorian accomplished with just a few hundred GPUs might be the most exciting leap yet. Meet Glimpse -- Our frontier-level multimodal model at platform.elorian.ai 🚀🚀🚀 @ElorianAI 🚀🚀🚀
Spent years watching models ace a single image and completely whiff on anything that moves or changes angle. Today we’re fixing that. Try Glimpse now, our first visual thinking model with competitive frontier performance compared to Astra on key public visual reasoning benchmarks.
24
18
2
210
12,416
david yan retweeted
Today we're thrilled to announce the release of our first frontier visual reasoning model
56
123
59
1,774
122,750