Hongkang Li
@LiHongkang_jntm
Ph.D student at Rensselaer Polytechnic Institute
🔥Our #ICLR2025 Oral paper "When is Task Vector Provably Effective for Model Editing? A Generalization Analysis of Nonlinear Transformers" will be presented on 04/26, 4:18 p.m. — 4:30 p.m. at Garnet 216-218. Poster pre will be on 04/26, 10:00 a.m. — 12:30 p.m. #341.
Are you a big fan of in-context learning (ICL)? Check out our @IBMResearch blog post highlighting our @icmlconf paper demystifying ICL. We characterize how ICL learns and generalizes. With @LiHongkang_jntm Meng Wang @rpi Songtao Lu & Xiaodong Cui Blog: research.ibm.com/blog/demystify…
🔥Our #ICLR2025 poster paper "Training Nonlinear Transformers for Chain-of-Thought Inference: A Theoretical Generalization Analysis" will be presented on 04/25, 3:00 p.m. to 5:30 p.m. #342. Link to the paper: openreview.net/pdf?id=n7n8McE…
This work as been accepted by #ICLR2025. Please see this link openreview.net/forum?id=n7n8M…). We will update our final version soon.
🚀Excited to share our new preprint on the theoretical analysis of training and generalization of chain-of-thought. The Arxiv link can be found at arxiv.org/pdf/2410.02167. We have the following results. [1/n]
Our follow-up work on the LLM theory---- the learning and generalization mechanism of Chain-of-Thought (CoT), will be presented in the next two days of the @icmlconf workshops. 1. Fri 26 Jul., Straus 2, HiLD Workshop. 2. Sat 27 Jul., Straus 2, TF2M Workshop.
Thanks @IBMResearch for posting a blog about our work on in-context learning. Please see this link: research.ibm.com/blog/demystify…
🔥Excited to share our poster at #ICML2024. This work studies the training dynamics of nonlinear Transformers, together with the In-Context Learning generalization capability of the model. Time: Jul 23rd, Tuesday, 1:30-3:00 pm. Location: Hall C 4-9 #403.
The 3rd AdvML-Frontiers Workshop (@AdvMLFrontiers advml-frontier.github.io) is set for #NeurIPS 2024 (@NeurIPSConf)! This year, we're delving into the expansion of the trustworthy AI landscape, especially in large multi-modal systems. @trustworthy_ml @llm_sec🚀 We're now…
🔥Excited to share our poster at #ICML2024. This work studies the training dynamics of nonlinear Transformers, together with the In-Context Learning generalization capability of the model. Time: Jul 23rd, Tuesday, 1:30-3:00 pm. Location: Hall C 4-9 #403.
Another work at #ICML2024. This work theoretically studies the training and generalization of a one-layer Graph Transformer with trainable positional encoding. Time: Jul. 24th, Wednesday, 1:30-3:00 pm. Location: Hall C 4-9 #506
