Auto-Instruct: Automatic Instruction Generation and Ranking for Black-Box Language Models paper page: huggingface.co/papers/2310.13… Large language models (LLMs) can perform a wide range of tasks by following natural language instructions, without the necessity of task-specific

Lucy is hiring a postdoc to work with our lab! Please apply if you are interested.

🤨LLMs struggle to follow instructions in low-resource languages? ⚡️Introducing PLUG: leveraging pivot language in cross-lingual instruction tuning 📈Improved LLaMA-2 by 32% on 4 diverse languages! Check out our new preprint at➡️arxiv.org/abs/2311.08711

We guide Diffusion Models to generate molecule graphs we want - they are capable of improving property prediction accuracy! With massive messy unlabeled data, Data-centric transfer can be better than Parameter-centric pre-training. #NeurIPS2023 arxiv.org/abs/2303.10108

It somehow suggests that academic groups, mostly limited on compute, are training future scientists to get a necessary skillset for being hirable in industry (and academia). [hahaha] (I always felt extremely stressed with lack of compute for my group.)

🎉Excited to announce the 3rd Workshop on Knowledge-Augmented NLP at ACL 2024 in Thailand! Submission deadline: May 17, 2024. Eager to reconnect with old friends and welcome new faces in the Knowledge NLP community! openreview.net/group?id=aclwe… #ACL2024 #NLProc

Instructing LLMs to solve math word prob better: - Mask a condition, Predict the condition with initial answer, Verify the answer, Rectify to avoid wrong answer(s) (AAAI'24): wzy6642.github.io/prp.github.io/ - Identify and Ignore Irrelevant Conditions (NAACL'24): wzy6642.github.io/I3C.github.io/

Professionals and animated characters will compete for the Oscar Best NPC. And every of us will be the (best) actor/actress (main character!) -- what if I was Jack on the Titanic? what if I was Judy in Zootopia? 🤩

📢 Excited to share that we will organize the 3rd workshop on Knowledge-Augmented NLP at ACL 2024. We will have six amazing speakers! We welcome your submissions and invite you to discuss with our speakers and organizers at the workshop. Looking forward to seeing you in Thailand!

📢 New paper: Compared to 𝐌𝐮𝐥𝐭𝐢-𝐦𝐨𝐝𝐚𝐥 𝐂𝐨𝐓, We found 𝐃𝐞𝐬𝐜𝐫𝐢𝐛𝐞 (visual description generation)-then-𝐑𝐞𝐚𝐬𝐨𝐧 (generating 𝐌𝐮𝐥𝐭𝐢-𝐦𝐨𝐝𝐚𝐥 𝐂𝐨𝐓 with the assistance of descriptions) could greatly improve math reasoning on MathVista and MathVerse.

We’ll be streaming live on openai.com at 10AM PT Monday, May 13 to demo some ChatGPT and GPT-4 updates.

Fei-Fei Li says Stanford's Natural Language computing lab has only 64 GPUs and academia is "falling off a cliff" relative to industry

[KnowledgeLM @ ACL24] KnowledgeLM Workshop 🚨 Update: We've extended the paper submission deadline to May 30 to accommodate COLM review releasing. 📢 We welcome submissions of Finding papers to present at our workshop! We have lined up wonderful speakers, and we are eager to engage

I’ll be at #NAACL2024 next week! Please email me if you’re interested in meeting up to talk about LLM reasoning and verification, instruction tuning, knowledge augmentation, hallucination reduction, unlearning for safety; or anything else! (credit: template from David Chiang)

🧐Previous math augmentation focused on improving single-round QA 🎯We introduce a new method that1⃣augments standard math settings2⃣excels in reflective thinking scenarios! 👉Check our latest preprint at huggingface.co/papers/2406.12…

🔬 🚀 Why Molecular Structures Are Insufficient and What's the Future of Molecular Representation Learning? For a long time, molecular structures like sequences and graphs have been the sole focus of molecular representation learning—but they aren't enough!

Thrilled to announce our finalized schedule at #ACL2024! We're excited to feature 6 keynote speakers and 30 accepted papers. Join us for an inspiring event!

Just published: "Current and future directions in network biology", a comprehensive review by leading experts in the field! Read it here: doi.org/10.1093/bioadv… #NetworkBiology #ISCB