Eric Zelikman
@ericzelikman
Followers
20K
Following
9K
Media
149
Statuses
863
lgtm-ing @xAI // was phd-ing @stanford
Joined April 2010
stare long enough and any optimization problem starts looking like a computer kernel.
20
6
178
RT @KaiyuYang4: 🚀 Excited to share that the Workshop on Mathematical Reasoning and AI (MATH‑AI) will be at NeurIPS 2025!.📅 Dec 6 or 7 (TBD)….
0
44
0
RT @ShirleyYXWu: CollabLLM won #ICML2025 ✨Outstanding Paper Award along with 6 other works! . 🫂 Absolutey honored a….
0
26
0
building reasoning agents w/ @YuchenHe07 @qhwang3 was so fun, and the next paradigm will be even cooler -- agents will solve far harder problems far faster.
From the 1st RL training using tools on a mini reasoning model at 16% HLE till now building the smartest agent w/ @qhwang3 @ericzelikman , more fun and breakthroughs to go! 🤖
15
38
289
RT @noahdgoodman: It turns out that a lot of the most interesting behavior of LLMs can be explained without knowing anything about architec….
0
20
0
fun note: @HeinrichKuttler once described my env config as "the final boss of python venv issues" -- has been mostly issue free for a few months now, thanks mostly to uv 🤞.
We've been using uv a few months now and I've never felt better. I have more energy. My skin is clearer. My eye sight has improved.
5
0
86
RT @jyangballin: 40% with just 1 try per task: SWE-agent-LM-32B is the new #1 open source model on SWE-bench Verified. We built it by synt….
0
136
0
NaN sample efficiency.
seems like a big theme lately (e.g. also "RL for Reasoning w/ One Training Example") is that approaches don't get nearly enough bang for each training point's buck - cool!.
3
0
86
RT @scychan_brains: Check out our new work: Generalization from context often outperforms generalization from finetuning. And you might g….
0
21
0
seems like a big theme lately (e.g. also "RL for Reasoning w/ One Training Example") is that approaches don't get nearly enough bang for each training point's buck - cool!.
Introducing COMPACT: COMPositional Atomic-to-complex Visual Capability Tuning, a data-efficient approach to improve multimodal models on complex visual tasks without scaling data volume. 📦. 1/10
4
8
92
Let’s start with Grok 3 Mini. When we set out to build a fast, affordable mini model, we knew it would be good but even we didn’t expect it to be this good. Some highlights:. - Grok 3 Mini tops the leaderboards on graduate-level STEM, math, and coding, outcompeting flagship
0
0
11
evergreen.
not listing better widely-available alternatives on a comparison chart doesn't make them not exist btw.
6
0
50
cool pipeline for analyzing lots of screenshot data 🖼️ we need good tools to understand how we interact w/ complex algos.
New paper up on ArXiv, with lead author Merve Cerit presenting it at #CHI2025: the Media Content Atlas (MCA): an open-source, AI-powered pipeline for inductive inquiry into what people actually see and do on their phones.
1
1
29
tiny oversight, think you missed a model. happy to help out!
For the first time, Google is responding to OpenAI's announcement in < 24 hours. The WAR is officially ON, and Google wants the LLM market. Google is now dominating +90% of the price share
15
34
520
i prefer to have axis labels actually, just figured someone needed to hear that
14
10
291
not listing better widely-available alternatives on a comparison chart doesn't make them not exist btw.
23
3
310