Research map / Label-efficient and robust learning / Domain adaptation and generalization
Project and Probe: Sample-Efficient Adaptation by Interpolating Orthogonal Features
Annie S Chen, Yoonho Lee, Amrith Setlur, Sergey Levine, Chelsea Finn
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes Project and Probe (Pro2), a lightweight sample-efficient transfer method that learns a diverse set of orthogonal features by projecting out source-trained features and then probing, interpolating them for adaptation with very little target data.
Problem
Adapting pre-trained models to distribution shift when only a very small number of labeled target data points is available and labels are expensive.
Approach
Learn a diverse set of features by training a linear probe on features orthogonal to those already learned from the source (projection), then interpolate among the orthogonal features to adapt with few target samples.
Experimental setup
Builds on: linear probing fine-tuning
Closest papers by meaning
- Few-shot Adaptation to Distribution Shifts By Mixing Source and Target Embeddings (ICML 2024 · similarity 0.89)
- Transferring Knowledge From Large Foundation Models to Small Downstream Models (ICML 2024 · similarity 0.86)
- Mixed-Sample SGD: an End-to-end Analysis of Supervised Transfer Learning (NeurIPS 2025 · similarity 0.86)
- Selective Mixup Helps with Distribution Shifts, But Not (Only) because of Mixup (ICML 2024 · similarity 0.86)
- Discriminative Sample-Guided and Parameter-Efficient Feature Space Adaptation for Cross-Domain Few-Shot Learning (CVPR 2024 · similarity 0.86)
- A High-Dimensional Statistical Method for Optimizing Transfer Quantities in Multi-Source Transfer Learning (NeurIPS 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
