NeurIPS 2023 Workshop DistShift Submissions

AutoFT: Robust Fine-Tuning by Optimizing Hyperparameters on OOD Data
Caroline Choi, Yoonho Lee, Annie S Chen, Allan Zhou, Aditi Raghunathan, Chelsea Finn
- Published: 28 Oct 2023, Last Modified: 02 Apr 2024
- DistShift 2023 Poster
- Readers: Everyone
Discovering environments with XRM
Mohammad Pezeshki, Diane Bouchacourt, Mark Ibrahim, Nicolas Ballas, Pascal Vincent, David Lopez-Paz
- Published: 28 Oct 2023, Last Modified: 02 Apr 2024
- DistShift 2023 Poster
- Readers: Everyone
Predicting the Performance of Foundation Models via Agreement-on-the-Line
Aman Mehra, Rahul Saxena, Taeyoun Kim, Christina Baek, J Zico Kolter, Aditi Raghunathan
- Published: 28 Oct 2023, Last Modified: 02 Apr 2024
- DistShift 2023 Poster
- Readers: Everyone
Continually Adapting Optimizers Improve Meta-Generalization
Wenyi Wang, Louis Kirsch, Francesco Faccio, Mingchen Zhuge, Jürgen Schmidhuber
- Published: 28 Oct 2023, Last Modified: 02 Apr 2024
- DistShift 2023 Poster
- Readers: Everyone
Towards Calibrated Robust Fine-Tuning of Vision-Language Models
Changdae Oh, Mijoo Kim, Hyesu Lim, Junhyeok Park, Euiseog Jeong, Zhi-Qi Cheng, Kyungwoo Song
- Published: 28 Oct 2023, Last Modified: 02 Apr 2024
- DistShift 2023 Poster
- Readers: Everyone
Stochastic linear dynamics in parameters to deal with Neural Networks plasticity loss
Alexandre Galashov, Michalis Titsias, Razvan Pascanu, Yee Whye Teh, Maneesh Sahani
- Published: 28 Oct 2023, Last Modified: 02 Apr 2024
- DistShift 2023 Poster
- Readers: Everyone
TiC-CLIP: Continual Training of CLIP Models
Saurabh Garg, Mehrdad Farajtabar, Hadi Pouransari, Raviteja Vemulapalli, Sachin Mehta, Oncel Tuzel, Vaishaal Shankar, Fartash Faghri
- Published: 28 Oct 2023, Last Modified: 02 Apr 2024
- DistShift 2023 Oral
- Readers: Everyone
Towards General-Purpose In-Context Learning Agents
Louis Kirsch, James Harrison, C. Daniel Freeman, Jascha Sohl-Dickstein, Jürgen Schmidhuber
- Published: 28 Oct 2023, Last Modified: 02 Apr 2024
- DistShift 2023 Poster
- Readers: Everyone
Turn Down the Noise: Leveraging Diffusion Models for Test-time Adaptation via Pseudo-label Ensembling
Mrigank Raman, Rohan Shah, Akash Kannan, Pranit Chawla
- Published: 28 Oct 2023, Last Modified: 02 Apr 2024
- DistShift 2023 Poster
- Readers: Everyone
Improving Domain Generalization in Contrastive Learning using Domain-Aware Temperature Control
Robert A Lewis, Katie Matton, Rosalind Picard, John Guttag
- Published: 28 Oct 2023, Last Modified: 02 Apr 2024
- DistShift 2023 Poster
- Readers: Everyone
Two-stage LLM Fine-tuning with Less Specialization and More Generalization
Yihan Wang, Si Si, Daliang Li, Michal Lukasik, Felix Yu, Cho-Jui Hsieh, Inderjit S Dhillon, Sanjiv Kumar
- Published: 28 Oct 2023, Last Modified: 02 Apr 2024
- DistShift 2023 Poster
- Readers: Everyone
Better than Balancing: Debiasing through Data Attribution
Saachi Jain, Kimia Hamidieh, Kristian Georgiev, Marzyeh Ghassemi, Aleksander Madry
- Published: 28 Oct 2023, Last Modified: 02 Apr 2024
- DistShift 2023 Poster
- Readers: Everyone
Confidence-Based Model Selection: When to Take Shortcuts in Spurious Settings
Annie S Chen, Yoonho Lee, Amrith Setlur, Sergey Levine, Chelsea Finn
- Published: 28 Oct 2023, Last Modified: 02 Apr 2024
- DistShift 2023 Poster
- Readers: Everyone
Outlier-Robust Group Inference via Gradient Space Clustering
Yuchen Zeng, Kristjan Greenewald, Luann Jung, Kangwook Lee, Justin Solomon, Mikhail Yurochkin
- Published: 28 Oct 2023, Last Modified: 02 Apr 2024
- DistShift 2023 Poster
- Readers: Everyone
Transfer Learning, Reinforcement Learning for Adaptive Control Optimization under Distribution Shift
Pankaj Rajak, Wojciech Kowalinski, Fei Wang
- Published: 28 Oct 2023, Last Modified: 02 Apr 2024
- DistShift 2023 Poster
- Readers: Everyone
Connect Later: Improving Fine-tuning for Robustness with Targeted Augmentations
Helen Qu, Sang Michael Xie
- Published: 28 Oct 2023, Last Modified: 02 Apr 2024
- DistShift 2023 Poster
- Readers: Everyone
Simplifying and Stabilizing Model Selection in Unsupervised Domain Adaptation
Dapeng Hu, Mi Luo, Jian Liang, Chuan-Sheng Foo
- Published: 28 Oct 2023, Last Modified: 02 Apr 2024
- DistShift 2023 Poster
- Readers: Everyone
Understanding subgroup performance differences of fair predictors using causal models
Stephen Robert Pfohl, Natalie Harris, Chirag Nagpal, David Madras, Vishwali Mhasawade, Olawale Elijah Salaudeen, Katherine A Heller, Sanmi Koyejo, Alexander Nicholas D'Amour
- Published: 28 Oct 2023, Last Modified: 02 Apr 2024
- DistShift 2023 Poster
- Readers: Everyone
Learning Causally-Aware Representations of Multi-Agent Interactions
Yuejiang Liu, Ahmad Rahimi, Po-Chien Luan, Frano Rajič, Alexandre Alahi
- Published: 28 Oct 2023, Last Modified: 02 Apr 2024
- DistShift 2023 Poster
- Readers: Everyone
Are all classes created equal? Domain Generalization for Domain-Linked Classes
Kimathi Kaai, Saad Hossain, Sirisha Rambhatla
- Published: 28 Oct 2023, Last Modified: 02 Apr 2024
- DistShift 2023 Poster
- Readers: Everyone
Fusing Models with Complementary Expertise
Hongyi Wang, Felipe Maia Polo, Yuekai Sun, Souvik Kundu, Eric P. Xing, Mikhail Yurochkin
- Published: 28 Oct 2023, Last Modified: 02 Apr 2024
- DistShift 2023 Poster
- Readers: Everyone
Adaptive Sharpness-Aware Pruning for Robust Sparse Networks
Anna Bair, Hongxu Yin, Maying Shen, Pavlo Molchanov, Jose M. Alvarez
- Published: 28 Oct 2023, Last Modified: 02 Apr 2024
- DistShift 2023 Poster
- Readers: Everyone
Skill-Mix: A Flexible and Expandable Family of Evaluations for AI Models
Dingli Yu, Simran Kaur, Arushi Gupta, Jonah Brown-Cohen, Anirudh Goyal, Sanjeev Arora
- Published: 28 Oct 2023, Last Modified: 02 Apr 2024
- DistShift 2023 Poster
- Readers: Everyone
Context is Environment
Sharut Gupta, David Lopez-Paz, Stefanie Jegelka, Kartik Ahuja
- Published: 28 Oct 2023, Last Modified: 02 Apr 2024
- DistShift 2023 Poster
- Readers: Everyone
Do Transformers Parse while Predicting the Masked Word?
Haoyu Zhao, Abhishek Panigrahi, Rong Ge, Sanjeev Arora
- Published: 28 Oct 2023, Last Modified: 02 Apr 2024
- DistShift 2023 Poster
- Readers: Everyone