fewshot

[ICCV 2023] Lighting Every Darkness in Two Pairs: A Calibration-Free Pipeline for RAW Denoising && [Arxiv 2023] Make Explicit Calibration Implicit: Calibrate Denoiser Instead of the Noise Model && [NeurIPS 2025] UltraLED: Learning to See Everything in Ultra-High Dynamic Range Scenes

Meta Transfer Learning for Few Shot Semantic Segmentation using U-Net

Generate mini-ImageNet with ImageNet for fewshot learning

Data and software artifacts for the EMNLP 2024 (Main) paper "What Are the Odds? Language Models Are Capable of Probabilistic Reasoning"

Few-shot NLP in PyTorch

Few-shot adaptation of SmolVLA on LIBERO, studying how fine-tuning with limited demonstrations affects success on held-out manipulation tasks.

fewshot-picker-bench is a deterministic few-shot example picker: it ranks a pool of examples against your query with TF-IDF, re-ranks for diversity with MMR, an