CAOS: Conformal Aggregation of One-Shot Predictors

Kari Jaaskelainen

10 Jan 2026 — 1 min read

Teaching an AI a brand-new task from just one labeled example is fast—but how sure can you be about its answers?

CAOS (Conformal Aggregation of One‑Shot Predictors) is a new method from Maja Waldron that adds reliable confidence to one‑shot learning. Instead of trusting a single quick‑adapted model, CAOS aggregates multiple one‑shot predictors and uses a leave‑one‑out calibration step to squeeze the most out of scarce labels—no wasteful data splitting.

Even though it breaks some classical assumptions, CAOS comes with proofs that its confidence levels are reliable: its prediction sets include the correct answer at the promised rate. In tests on facial landmarking and RAFT text classification, CAOS produced smaller prediction sets than common baselines while keeping that reliability.

Confidence you can count on—from a single example.

What it is: a conformal framework for uncertainty in one‑shot learning
Why it matters: trustworthy AI with minimal labels
Paper: https://arxiv.org/abs/2601.05219v1

Paper: https://arxiv.org/abs/2601.05219v1

Register: https://www.AiFeta.com

#AI #MachineLearning #ConformalPrediction #OneShotLearning #Uncertainty #TrustworthyAI #NLP #ComputerVision

Meet GRACE: a moral governor for safer, more transparent AI

AI agents are getting powerful—so how do we make sure they do the right thing, not just the effective thing? Meet GRACE, a reason-based moral governor that keeps AI behavior aligned with human norms by separating moral reasoning from goal-driven decision-making. * Moral Module: uses deontic logic and explicit reasons

Why some fine-tuned LLMs miss phishing—and how to fix it

Not all fine-tuned LLMs spot phishing equally. A new study tests Llama 3.1 8B, Gemma 2 9B, and Mistral on high-stakes phishing detection—and uses SHAP and mechanistic interpretability to reveal why models do (or don’t) generalize. * Architecture × data diversity matters: Gemma 2 9B hits state-of-the-art performance (F1

AI that explains itself—by following the science

AI that explains itself—by following the science Black-box AI can be powerful, but it often can’t tell us why it made a decision. Concept Bottleneck Models (CBMs) try to fix that by predicting human-understandable concepts first, then the final answer. The catch: standard CBMs ignore domain-specific cause-and-effect and

Meet GenomAgent: A Team of AI Specialists for Smarter Genomics Q&A

TL;DR: Finding reliable answers in genomics is hard. GenomAgent turns one big AI into a coordinating team of specialists—and beats the current leader by 12% on a key benchmark. Genomic facts live across many databases. Standard chatbots struggle because they can’t flexibly query those sources. GeneGPT added

Read more

Meet GRACE: a moral governor for safer, more transparent AI

Why some fine-tuned LLMs miss phishing—and how to fix it

AI that explains itself—by following the science

Meet GenomAgent: A Team of AI Specialists for Smarter Genomics Q&A