@kevinwng

Research @ OpenAI | prev @princeton_rl @citsecurities

 sf
Joined June 2023
  • Tweets42
  • Following287
  • Followers3.1K
9Media
Excited to share that our paper "1000 Layer Networks for Self-Supervised RL: Scaling Depth Can Enable New Goal-Reaching Capabilities" has won the Best Paper Award at NeurIPS '25! Hope to see you all in San Diego :)
Kevin Wang@kevinwng20 Mar 2025
1/ While most RL methods use shallow MLPs (~2–5 layers), we show that scaling up to 1000-layers for contrastive RL (CRL) can significantly boost performance, ranging from doubling performance to 50x on a diverse suite of robotic tasks. Webpage+Paper+Code: wang-kevin3290.github.io/sca…
10
23
5
218
50,066
In Astra, we made significant progress on subjective domains like design aesthetics and knowledge work. It’s been a blast developing new methods for training on non-verifiable domains. Excited for this line of work and what’s coming next. Go try out Astra when you can! :)
OpenAI
@OpenAISep 3
Replying to @OpenAI
GPT-6 Astra is state-of-the-art on FrontierMath Tier 4, ARC-AGI 3, and TerminalBench-4.0. GPT‑6 Astra is also a major advance for scientific discovery, with state-of-the-art performance on Terminal-Bench Science 0.1 and HealthBench Pro.
19
17
3
452
93,660
Kevin Wang retweeted
OpenAI
@OpenAISep 3
1,977
3,444
2,450
38,819
11,410,556
Kevin Wang retweeted
Noam Brown
@polynoamialAug 1
An internal version of Astra, @OpenAI’s next major model family, solved 10 major open problems in mathematics, quantum complexity, and theoretical computer science. We believe it will be a major step for scientific reasoning. openai.com/index/ten-advance…
Lijie Chen
@wjmzbmr1Aug 1
10 proofs from our next major model Astra on long-standing open problems in mathematics and theoretical computer science (also including new circuit lower bounds for computing the permanent!) GPT-5.6 has already enabled so much exciting work in math and science. Can’t wait to see what comes next!
832
2,517
1,517
17,456
13,370,733
Kevin Wang retweeted
OpenAI
@OpenAIApr 23
Introducing GPT-5.5 A new class of intelligence for real work and powering agents, built to understand complex goals, use tools, check its work, and carry more tasks through to completion. It marks a new way of getting computer work done. Now available in ChatGPT and Codex.
2,445
6,665
3,414
50,788
13,426,964
Kevin Wang retweeted
OpenAI
@OpenAIMar 5
GPT-5.4 Thinking and GPT-5.4 Pro are rolling out now in ChatGPT. GPT-5.4 is also now available in the API and Codex. GPT-5.4 brings our advances in reasoning, coding, and agentic workflows into one frontier model.
2,037
3,202
1,988
23,157
7,203,435
Kevin Wang retweeted
OpenAI
@OpenAI11 Dec 2025
On GDPval, an eval measuring well-specified knowledge work tasks across 44 occupations, GPT-5.2 Thinking is our first model that performs at a human expert level. These tasks include making presentations, spreadsheets, and other artifacts.
16
80
39
964
226,757
Happy to share that our work “1000 Layer Networks for Self-Supervised RL: Scaling Depth Can Enable New Goal-Reaching Capabilities” was accepted to #NeurIPS2025 as an Oral Presentation! Grateful to my amazing collaborators @IJ_Apps @m_bortkiewicz @ben_eysenbach 🥳
Kevin Wang@kevinwng20 Mar 2025
1/ While most RL methods use shallow MLPs (~2–5 layers), we show that scaling up to 1000-layers for contrastive RL (CRL) can significantly boost performance, ranging from doubling performance to 50x on a diverse suite of robotic tasks. Webpage+Paper+Code: wang-kevin3290.github.io/sca…
2
19
1
142
15,510
1/ While most RL methods use shallow MLPs (~2–5 layers), we show that scaling up to 1000-layers for contrastive RL (CRL) can significantly boost performance, ranging from doubling performance to 50x on a diverse suite of robotic tasks. Webpage+Paper+Code: wang-kevin3290.github.io/sca…
11
97
19
572
161,353
more replies
8/ One negative result: In preliminary experiments using OGBench, we evaluated depth scaling in offline goal-conditioned RL. We found that increasing network depth didn’t improve offline performance, hinting that its benefits online may partly arise from enhanced exploration.
1
24
3,007
9/ Thanks to amazing collaborators @IJ_Apps, @m_bortkiewicz @tomasztrzcinsk1, and @ben_eysenbach. Please check out our paper and website for more details! Paper: arxiv.org/abs/2503.14858 Website: wang-kevin3290.github.io/sca… Code: github.com/wang-kevin3290/sc…

1000 Layer Networks for Self-Supervised RL: Scaling Depth Can...

Scaling up self-supervised learning has driven breakthroughs in language and vision, yet comparable progress has remained elusive in reinforcement learning (RL). In this paper, we study building...

arxiv.org
2
2
34
2,949
Load more