Quyet V. Do
@Quyet_Azir
Followers
124
Following
389
Media
9
Statuses
210
AI/NLP PhD at @virginia_tech, supervised by @tuvllms. Research Intern at @AdobeResearch.
Blacksburg, VA
Joined May 2022
RT @AndrewYNg: The invention of modern writing instruments like the typewriter made writing easier, but they also led to the rise of writer….
0
278
0
RT @ylecun: Excellent blog post from Turing Post on JEPA (Joint Embedding Predictive Architecture), my favorite meta-architecture for Self-….
0
136
0
RT @shizhediao: 🚨 NVIDIA is launching the Data Filtering Challenge for training edge language models! . We believe edge LMs are the future….
0
10
0
RT @tuvllms: ✨ New paper ✨.🚨 Scaling test-time compute can lead to inverse or flattened scaling!!. We introduce SealQA, a new challenge ben….
0
38
0
RT @AndrewYNg: Contrary to standard prompting advice that you should give LLMs the context they need to succeed, I find it’s sometimes fast….
0
168
0
RT @UnslothAI: We partnered with @HuggingFace to teach you how to fine-tune LLMs with GRPO!. Learn about:.• Reward functions + creating the….
0
220
0
RT @Swarooprm7: SWE tip: The importance of software design is higher than ever, given how well AI can code. Highly recommend checking out….
0
32
0
RT @dreamingtulpa: adobe is cooking a new inpainting method that understands context extremely well!. ELI5: it can insert stuff with the co….
0
132
0
RT @PfeiffJo: I am hiring a Student Researcher for our Modularity team at the Google DeepMind office in Zurich🇨🇭. Please fill out the inter….
docs.google.com
We are excited to offer an opportunity for students to work with our research team at the GDM Zurich office in 2025. Please provide the following information to express your interest.
0
58
0
Congrats @linusdd44804 🎉.Very interesting paper!.
🚨 New paper 🚨. Excited to share my first paper w/ my PhD students!!. We find that advanced LLM capabilities conferred by instruction or alignment tuning (e.g., SFT, RLHF, DPO, GRPO) can be encoded into model diff vectors (à la task vectors) and transferred across model
1
0
1
RT @denny_zhou: “If you want to be ambitious in your AI research, … don’t be distracted by privacy, explainability, or safety” https://t.co….
0
184
0
RT @karpathy: Agency > Intelligence. I had this intuitively wrong for decades, I think due to a pervasive cultural veneration of intelligen….
0
4K
0
RT @JaechulRoh: 🧠💸 "We made reasoning models overthink — and it's costing them big time.". Meet 🤯 #OVERTHINK 🤯 — our new attack that forces….
0
31
0
Incredible!.
The future of human video generation is here. Watch as [OmniHuman-1] turns text into a hyper-realistic talking human in seconds! No actors, no cameras—just pure AI magic.
0
0
0
RT @pelaseyed: Traditional RAG sucks because it promises "relevant chunks" but in fact returns "similar chunks". Relevancy requires reaso….
0
294
0
RT @sybilhyz: Last year, I joined DeepSeek with no RL experience. While conducting Mathshepherd and DeepSeekMath research, I independently….
0
564
0
RT @teortaxesTex: If you can only read one DeepSeek paper in your life, read DeepSeek Math. Everything else is either ≈obvious in hindsight….
0
469
0