Check out "Assessing Composition in Sentence Vector Representations" by Allyson Ettinger, Ahmed Elgohary, Colin Phillips, Philip Resnik next in Session 3-1-b #COLING2018 ling.umd.edu/~aetting/CompE…

#nlphighlights 72: the anatomy of a question answering task, with Jordan Boyd-Graber. This is our first episode in a new format, giving a more general overview of an area instead of discussing a specific paper. There will be more of these in the future. soundcloud.com/nlp-highlights…

While we are at it, can we grant international students #MultipleEntryVisa for the duration of their studies (instead of single-entry)? It may sound like a minor issue for many but it is actually a big deal for many international students. I explain it below. 👇

Second call for papers! Please consider submitting your work to our workshop on NLP for Programming at ACL 2021 (NLP4Prog ACL 2025)! w/ Royi Lachmy, Ziyu Yao Greg Durrett , Milos Gligoric, Jessy Li Bored Ray Mooney Graham Neubig Yu Su Huan Sun (OSU) Reut Tsarfaty

🤖 LLMs are powerful, but their "one-size-fits-all" safety alignment limits flexibility. Safety standards vary across cultures and users—what’s safe in one context might not be in another. 🌍 We propose ✨Controllable Safety Alignment✨ for inference-time safety adaptation! 🧵👇

Thrilled to share that 𝗖𝗼𝗻𝘁𝗿𝗼𝗹𝗹𝗮𝗯𝗹𝗲 𝗦𝗮𝗳𝗲𝘁𝘆 𝗔𝗹𝗶𝗴𝗻𝗺𝗲𝗻𝘁 has been accepted to #ICLR2025! 🚀 We propose a framework that adapt LLMs to varying social norms and diverse safety requirements without re-training. Link to our paper: arxiv.org/abs/2410.08968

Our Controllable Safety Alignment paper will be presented at #ICLR2025 this week in Singapore 🇸🇬! We've release our code and the human-authored CoSApien👥 dataset: 👉 aka.ms/controllable-s… Watch the short video summary here: 🎬 youtube.com/watch?v=kDioFn…

Connecting Generative Adversarial Networks and Actor-Critic Methods x.com/j_gauthier/sta…

Congratulations to Andrew Miller for winning the Larry S. Davis Doctoral Dissertation Award! Andrew Miller IC3 UMD Department of Computer Science

YAY! He He (student of me & Jordan Boyd-Graber) in UMD CLIP (->Stanford) won the Davis UMD Department of Computer Science doctoral dissertation award! umiacs.umd.edu/~hhe/

My blog entry at ACM SIGMOD Blog: Thinking Spatial wp.sigmod.org/?p=2012

My notes on Using Fast Weights to Attend to the Recent Past (by Jimmy Ba et al.): shortscience.org/paper?bibtexKe… Maybe an alternative to LTSMs?

"Learning to Reason With Adaptive Computation," Neumann et al.: arxiv.org/abs/1610.07647

Tao Lei talks about the problem of interpreting neural networks at #EMNLP2016 arxiv.org/abs/1606.04155

Are you sick of #nlproc folks saying NLP is "solved?" Think your linguistics can poke holes in systems? Prove it! bibinlp.umiacs.umd.edu

Natural Language Processing == English Language Processing? 69% of ACL 2016 long papers only evaluate on English...

This is the most legit useful 5 minute deep learning read I've had. nmarkou.blogspot.co.uk/2017/02/the-bl…

If you are at #emnlp2017 come to our workshop: generalizablenlp.weebly.com on Sept 8th and see results of bibinlp.umiacs.umd.edu/sharedtask.html

#AI2 is proud to announce AllenNLP – an open-source #NLP research library built on #PyTorch. Check it out at allennlp.org!

Announcing the #AI2 Key Scientific Challenges Program! Get $10K in seed funding + #AI research support & mentorship: ow.ly/FBza30fH9Yr