A bit of a late announcement! I am thrilled to let you all know that I will soon be starting my PhD at @Edin_CDT_NLP and working alongside Shay Cohen and @tetraduzione on neuro-symbolic learning and reasoning approaches. I am looking forward to an exciting journey ahead! π
Hiring!
Our Technical Safety and Governance (TSG) Lab at Oxford is hiring a Senior Researcher to work on interpretability, evaluations and AI safety for continually learning systems.
Please apply by Sep 22 and share with anyone who might be interested.
DM/email with questions!
ππ Introducing Pythagoras-Prover ππ
πΉ The smallest theorem prover to date β Pythagoras-Prover-4B
πΉ The first proof-of-concept diffusion theorem proving model β Pythagoras-Prover-Diffusion 4B
With efficient LoRA training and an 8192-token context length, Pythagoras-Prover achieves strong MiniF2F results with far smaller models.
π MiniF2F
- Pythagoras-Prover-4B achieves 86.1% Pass@32, outperforming DeepSeek-Prover-V2-671B.
- Pythagoras-Prover-32B achieves 89.8% Pass@32 and 92.6% Pass@2024, achieving the best MiniF2F result reported to date.
πWebsite: https://t.co/uFmBLA2sJS
π arXiv: https://t.co/rdLlr3cwY5
π€ Pythagoras-Prover-4B: https://t.co/fXziEcaDxB
π Our models, data, and the training pipeline will be released (gradually) soon.
A huge thank to my amazing collaborators: @zhengzhao97@CMihaelaStoian Qiyuan Xu, @haonanlp@WendaLi8 Shay Cohen @e_giunchiglia
@Edin_CDT_NLP@tetraduzione I want to thank everyone who has contributed to helping me pursue this opportunity, especially my mentors: @ggananth, @TorrentTiago, and Shay Cohen.
A bit of a late announcement! I am thrilled to let you all know that I will soon be starting my PhD at @Edin_CDT_NLP and working alongside Shay Cohen and @tetraduzione on neuro-symbolic learning and reasoning approaches. I am looking forward to an exciting journey ahead! π
Failure is the dance partner of success. It can feel insurmountable at times. Here is my story of academic failures -- my anti-CV if you will. I hope people find it useful. https://t.co/ZJfG8VKCY0
TL;DR: We use semi-supervised and weakly-supervised learning techniques to induce latent trees and achieve state-of-the-art results on three treebanks.
Delighted to announce that our paper (w/ Shay Cohen) titled "Co-training an Unsupervised Constituency Parser with Weak Supervision" has been accepted at ACL 2022 Findings. #NLProc#ACL2022
Want to read an insane tenure denial story? I was denied tenure at MIT in 2018. I was the only Hispanic faculty at my department, had 13k+ citations, 2 books, & papers in Nature, Science, & PNAS. I was never given a reason & the only letter I got told me to check the website./1π§΅
@chrisgamble88 Hi Chris, I am interested in this position. Let me know if you see a fit. If you do, could you open your DMs, so that we could chat more about this opportunity?