Haochen Zhang (@jhaochenz) header
Haochen Zhang (@jhaochenz) profile picture

Member of Technical Staff @AnthropicAI | prev. CS PhD @StanfordAILab

San Francisco
Joined August 2013
Photo by Haochen Zhang (@jhaochenz)
The best model for coding is here, come try Claude Opus 4.5!
Introducing Claude Opus 4.5: the best model in the world for coding, agents, and computer use. Opus 4.5 is a step forward in what AI systems can do, and a preview of larger changes to how work gets done.
1
7
1,087
Haochen Zhang retweeted
Introducing Claude Sonnet 4.5—the best coding model in the world. It's the strongest model for building complex agents. It's the best model at using computers. And it shows substantial gains on tests of reasoning and math.
1,101
3,033
2,046
19,692
5,052,790
Haochen Zhang retweeted
We've made Claude Opus 4 and Claude Sonnet 4 significantly better at avoiding reward hacking behaviors (like hard-coding and special-casing in code settings) that we frequently saw in Claude Sonnet 3.7.
4
9
2
110
12,995
Glad to see what we've been working on finally get out! Lots of cool ideas and hard work have gone into this model, and there are still so many open research questions ahead around solving alignment and scaling up RL. Looking forward to what we'll learn next.
Introducing the next generation: Claude Opus 4 and Claude Sonnet 4. Claude Opus 4 is our most powerful model yet, and the world’s best coding model. Claude Sonnet 4 is a significant upgrade from its predecessor, delivering superior coding and reasoning.
3
17
1,784
What if you can control how much the model thinks? Sharing our first hybrid reasoning model, with fine-grained control over thinking budget.
Introducing Claude 3.7 Sonnet: our most intelligent model to date. It's a hybrid reasoning model, producing near-instant responses or extended, step-by-step thinking. One model, two ways to think. We’re also releasing an agentic coding tool: Claude Code.
1
11
1,246
Also, Claude Code feels magic when I first use it, come try it out!
3
320
Our model now has the capability of using computers. Excited for a new era of human-AI collaboration.
Replying to @AnthropicAI
The new Claude 3.5 Sonnet is the first frontier AI model to offer computer use in public beta. While groundbreaking, computer use is still experimental—at times error-prone. We're releasing it early for feedback from developers.
24
3,886
Haochen Zhang retweeted
Machines of Loving Grace: my essay on how AI could transform the world for the better darioamodei.com/machines-of-…
1,200
666
5,541
2,912,193
Haochen Zhang retweeted
New today: organize your claude[dot]ai chats in Projects, and add files & context and 💫 custom instructions 💫 that are shared across all project chats. And on the Claude Team plan, you can discover great uses of Claude within your team using a new activity feed & shared chats.
You can now organize chats with Claude into shareable Projects. Each project includes a 200K context window, so you can include relevant documents, code, and files.
20
25
4
263
26,640
Haochen Zhang retweeted
Artifacts pro tip: If you are running into unsupported library errors with NPM modules, just ask Claude to use the cdnjs link instead and it should work just fine.
43
78
12
796
100,957
I joined the amazing team at @AnthropicAI last month. Excited to share what we’ve been working on recently!
Introducing Claude 3.5 Sonnet—our most intelligent model yet. This is the first release in our 3.5 model family. Sonnet now outperforms competitor models on key evaluations, at twice the speed of Claude 3 Opus and one-fifth the cost. Try it for free: claude.ai
4
2
81
26,136
🔥🔥🔥 Congrats on the launch! @demi_guo_ @chenlin_meng
Introducing Pika 1.0, the idea-to-video platform that brings your creativity to life. Create and edit your videos with AI. Rolling out to new users on web and discord, starting today. Sign up at pika.art
1
7
2,354
Haochen Zhang retweeted
📢 Introducing Voyage AI @Voyage_AI_! Founded by a talented team of leading AI researchers and me 🚀🚀. We build state-of-the-art embedding models (e.g., better than OpenAI 😜). We also offer custom models that deliver 🎯+10-20% accuracy gain in your LLM products. 🧵
36
89
15
744
225,550
How does model architecture influence the contrastive representations? Check out our paper "A theoretical study of inductive biases in contrastive learning" at #ICLR2023 . Virtual poster: iclr.cc/virtual/2023/poster/…. Joint work with @tengyuma.
7
1,111
Haochen Zhang retweeted
Our paper got accepted to ICML ‘22 as a long talk! Thanks to all the co-authors (@kendrick_shen @rmjones96 @sangmichaelxie @jhaochenz @tengyuma @percyliang). Congrats Kendrick on yet another oral (as an undergrad!)
Pretraining is ≈SoTA for domain adaptation: just do contrastive learning on *all* unlabeled data + finetune on source labels. Features are NOT domain-invariant, but disentangle class & domain info to enable transfer. Theory & exps: arxiv.org/abs/2204.00570 arxiv.org/abs/2204.02683
3
27
138
Curious about why contrastive learning produces representations useful for downstream tasks? Check out @jhaochenz, @ColinWei11, and @tengyuma's theoretical explanation in our latest blog post: ai.stanford.edu/blog/underst… Based on the NeurIPS 2021 oral paper: arxiv.org/abs/2106.04156
1
5
20
Haochen Zhang retweeted
Pretraining is ≈SoTA for domain adaptation: just do contrastive learning on *all* unlabeled data + finetune on source labels. Features are NOT domain-invariant, but disentangle class & domain info to enable transfer. Theory & exps: arxiv.org/abs/2204.00570 arxiv.org/abs/2204.02683
7
126
9
660
Haochen Zhang retweeted
Thinking of applying self-supervised learning (SSL) on your uncurated, imbalanced datasets? Good news: we found SSL is more robust to long tails than supervised representations. We also present theoretical and empirical analyses and an improved algorithm. arxiv.org/abs/2110.05025
5
77
3
384
Haochen Zhang retweeted
Why does contrastive learning magically produce linearly separable features? We leverage spectral graph theory to analyze it under realistic settings. (In contrast, many prior works require that positive pairs are independent conditioned on the label.) arxiv.org/abs/2106.04156
2
82
3
508
Haochen Zhang retweeted
With the right regularizer, linear autoencoders recover the ordered principal components, but very slowly. Interesting model system for representation learning since we know the optimal representation. New work w/ @XuchanB, @james_r_lucas, and @sushnt. arxiv.org/abs/2007.06731
3
30
131