Yifan Wu
@yifannnwu
吴奕凡; Research Scientist @Meta Llama Trains; Ph.D. @penn @picslupenn @GRASPlab; Computer Vision & Multi-Modal LLM & Medical Image Analysis.
Introducing Muscle v0 -- infinite degrees of freedom, from @DaxoRobotics. A different mountain to climb - with a far more beautiful peak. We built this from the ground up: - Ultra-dexterous - Built for machine learning - Durable and robust More below (1/n)
I read the DeepSeek-R1 paper the day it came out, and I don’t think GRPO is the key to its success. Instead, here’s what truly matters (ranked by importance): 1. Iterative RL and SFT 2. A hybrid reward model—mixing rule-based RM and neural RM for deterministic tasks 3.…
We (@chaoqi_w @yibophd @ZRChen_AISafety) have been eager to share our latest work on battling reward hacking since last November, but had to wait for the legal team's approval. Finally, we're excited to release: Causal Reward Modeling (CRM)! CRM tackles spurious correlations and…
🦃 At the end of Thanksgiving holidays, I finally finished the piece on reward hacking. Not an easy one to write, phew. Reward hacking occurs when an RL agent exploits flaws in the reward function or env to maximize rewards without learning the intended behavior. This is imo a…
The most bullish AI capability I'm looking for is not whether it's able to solve PhD grade problems. It's whether you'd hire it as a junior intern. Not "solve this theorem" but "get your slack set up, read these onboarding docs, do this task and let's check in next week".
I didn't see the talk, but the images I've seen of the slide seem quite offensive. Such generalizations should have no place in NeurIPS or anywhere else.
Mitigating racial bias from LLMs is a lot easier than removing it from humans! Can’t believe this happened at the best AI conference @NeurIPSConf We have ethical reviews for authors, but missed it for invited speakers? 😡
Our team at Meta is hiring Visiting Researcher for next year! If you are a current PhD student in CS at UW, CMU, Berkeley, or NYU, and interested in a 1-year part-time research collaboration, feel free to DM 🤝. Check out more details at: metacareers.com/jobs/395022364…
A nice read on flight. ✈️
Wrote about extrinsic hallucinations during the July 4th break. lilianweng.github.io/posts/2024-07-… Here is what ChatGPT suggested as a fun tweet for the blog: 🚀 Dive into the wild world of AI hallucinations! 🤖 Discover how LLMs can conjure up some seriously creative (and sometimes…
📢 Meta’s GenAI Media Foundations team is on the lookout for 2025 Research Scientist Interns! Interested in joining the front lines of GenAI media generation and making a real impact on next-gen foundation models? Shoot me an email at [email protected] with your website and Google…
I'm hiring a research intern starting next Spring or May. Possible topics: (multimodal) LLMs, LLM efficiency, model architectures, data(sets), representation learning. Experience in one of them preferred Apply below! Please email me if interested. metacareers.com/jobs/532549086…
Meta GenAI is looking for 2025 research interns across language, and multimodal. In particular, my team is looking for interns on RLHF algos, agents, and post-training more broadly. metacareers.com/jobs/432691156…
Our team is actively hiring multiple PhD interns for Summer 2025. These positions are on Multimodal (image/video) Large Language Model with the ultimate goal of impactful publications. Please feel free to DM me if interested. metacareers.com/jobs/432691156…
Will that day come soon?
o1-mini keeps refusing to try to solve the Riemann Hypothesis on my behalf. Model laziness continues to be a major issue sad ;p
Super excited to finally share what I have been working on at OpenAI! o1 is a model that thinks before giving the final answer. In my own words, here are the biggest updates to the field of AI (see the blog post for more details): 1. Don’t do chain of thought purely via…
We're releasing a preview of OpenAI o1—a new series of AI models designed to spend more time thinking before they respond. These models can reason through complex tasks and solve harder problems than previous models in science, coding, and math. openai.com/index/introduc…
A few prompt engineering tips I picked up during my time at @OpenAI
Why does the neural network display modular structure during learning and how does the modularity characterize neurodevelopment and cognitive behavior? Please check our recent paper in @ScienceAdvances with @marcelomattar. science.org/doi/10.1126/sc…
Congratulations to @wensi_wu, @yifannnwu and team on their abstract on the integration of biomechanics-derived training data to inform ML-based segmentation of valves at @midl_conference. Check out their presentation tomorrow at 15:30 (CEST) in Paris! @picslupenn @pennbioeng
Check more details in the paper! arxiv.org/abs/2405.14839 Joint work with @monagandhi09, Yufei Wang, @yifannnwu, @michael_s_yao, Chris Callison-Burch, James C. Gee, and @yatskar.
We demonstrate how to encode the expertise of specialized clinicians into AI by building an interpretable machine learning model that produces outputs understandable by humans.
Thrilled to share our new preprint of a collaboration work with Beijing Tongren Hospital and Upenn @yifannnwu @YueYangAI @michael_s_yao @picslupenn et al. ArXiv: arxiv.org/abs/2403.05606 Demo: mmcbm.liuy.site
Thrilled to share our new preprint of a collaboration work with Beijing Tongren Hospital and Upenn @yifannnwu @YueYangAI @michael_s_yao @picslupenn et al. ArXiv: arxiv.org/abs/2403.05606 Demo: mmcbm.liuy.site