The idea of an intelligence explosion caused by recursive self improvement has been around for a long time but until very recently it did not seem imminent. Now many leading researchers think it may happen quite soon. You can read our paper about it here:
https://t.co/sgUpugjpRY
2/N In our new PROMPT2BOX approach, we represent prompts as 📦 boxes.
With this representation, we propose to better:
📏 Estimate prompt specificity
🗺️ Visualize prompt spaces with BOX-SNE
🌳 Build specificity-aware hierarchical clusters
🔍 Discover finer-grained LLM weaknesses
Excited to present our NeurIPS paper "Learning Representations for Hierarchies with Minimal Support" at the morning poster session on December 12! Stop by at poster #3500 in the East Building!
@yuqirose@ucsd_cse I worked for a year to save some money for my PhD applications. Also potential students do need to take into account the post acceptance cost as well, e.g., visa application fee, flight tickets to the USA, some initial expenses like lease deposit etc.
✨✨Excited to share that our work, CBR-MRC, has been accepted to #EMNLP2023! In the era of LLMs, we demonstrate the persisting importance of semi-parametric reasoning in providing an accurate, robust, and interpretable method for QA.
🧵(1/2)
Planning to use your favorite AI assistant to polish your ACL submission?
Checkout ACL 2023's policy on AI Writing Assistance. https://t.co/nZqhVPs9uU
#ACL2023NLP#ACL2023Toronto#ChatGPT
Congratulations, @partha_p_t on winning the ACM India Early Career Researcher Award 2022!
It is an honour to have you as a friend and a colleague!
https://t.co/2taGGuCEUp
We were all waiting for @LogConference🤩
And it even recognized our work as spotlight!!
We had a closer look at virtual nodes, which are used a lot but only for graph classification and without theoretical investigation.
w/ #EunJeongHwang@DasguptaShib#TengfeiMa
⬇️
I am beyond thrilled to share that I'll be starting as an assistant professor at the Indian Institute of Science (IISc), Bangalore in April 2023.
I couldn’t have been luckier—I'm grateful for the support of many kind mentors, peers, students, friends and family members. (1/4)
Our paper “A Systematic Investigation of Commonsense Knowledge in Large Language Models” has been accepted to EMNLP! 🎉 This is joint work with Adhi, Jordan, Cyprien, Phil, and @aidanematzadeh during my internship at DeepMind last year!
🥇 Our new work on improving factuality in NLEs! We investigate several state-of-the-art models, all of them succumb to the issue of hallucinations. Our model is an early attempt to evidence-ground NLEs for RecSys that yields significantly better and factual outcomes! 🥇
5. Be careful who you ask for rec. letters. Younger profs in general write better-worded letters. And if the person says they can't write a good letter for you, do not insist that they still write one. Find someone else. It is very important to not have a mediocre rec. letter.
We are excited to partner with @imperialcollege to launch this seed fund, which will support early-stage 'blue skies' research, student exchange, and more!
🧳I am presenting (and attending in-person) at #NAACL2022 and #ICML2022 back-to-back for the next two weeks! I am transitioning into my final year of PhD and would love to talk about industry/post-doc research opportunities when we meet in person :)
I'll be presenting our work, Equi-explanation Maps: Concise and Informative Global Summary Explanations in FAccT22 soon! Work with @Negin_Rahimi and @jallan_umass . Do stop by Proceedings 15A to learn more :)
Super excited to be attending the #ACL2022 in-person.
I am presenting our work on representing words with Box Embeddings.
Poster Session: Today at 14:00
Oral Session: 23rd at 13:45
Here is the link to the paper:
https://t.co/xiNuc4WJ5T
Looking forward to your valuable inputs!
Super excited to be attending the #ACL2022 in-person.
I am presenting our work on representing words with Box Embeddings.
Poster Session: Today at 14:00
Oral Session: 23rd at 15:00
Here is the link to the paper:
https://t.co/xiNuc4WJ5T
Looking forward to your valuable inputs!