In-Context Learning (ICL) typically utilizes classification criteria from
probabilities of manually selected label tokens. However, we argue that such
token-based classification criteria lead to suboptimal decision boundaries,
despite delicate calibrations through translation and constrained rotation. To
address this problem, we propose Hidden Calibration, which renounces token
probabilities and uses the nearest centroid classifier on the LM's last hidden
states. In detail, we use the nearest centroid classification on the hidden
states, assigning the category of the nearest centroid previously observed from
a few-shot calibration set to the test sample as the predicted label. Our
experiments on 3 models and 10 classification datasets indicate that Hidden
Calibration consistently outperforms current token-based calibrations by about
20%. Our further analysis demonstrates that Hidden Calibration finds better
classification criteria with less inter-categories overlap, and LMs provide
linearly separable intra-category clusters with the help of demonstrations,
which supports Hidden Calibration and gives new insights into the conventional
ICL.

基于最近的质心分类器和语言模型的最后隐藏状态，使用隐藏校准来提供更好的分类准则，并通过实验证明隐藏校准在当前基于令牌的校准方法上具有约 20% 的性能优势，同时提供新的对传统上下文学习的认识。

基于标记的决策准则在上下文学习中存在亚最优问题

Token-based Decision Criteria Are Suboptimal in In-context Learning

In cross-domain few-shot classification, \emph{nearest centroid classifier}
(NCC) aims to learn representations to construct a metric space where few-shot
classification can be performed by measuring the similarities between samples
and the prototype of each class. An intuition behind NCC is that each sample is
pulled closer to the class centroid it belongs to while pushed away from those
of other classes. However, in this paper, we find that there exist high
similarities between NCC-learned representations of two samples from different
classes. In order to address this problem, we propose a bi-level optimization
framework, \emph{maximizing optimized kernel dependence} (MOKD) to learn a set
of class-specific representations that match the cluster structures indicated
by labeled data of the given task. Specifically, MOKD first optimizes the
kernel adopted in \emph{Hilbert-Schmidt independence criterion} (HSIC) to
obtain the optimized kernel HSIC (opt-HSIC) that can capture the dependence
more precisely. Then, an optimization problem regarding the opt-HSIC is
addressed to simultaneously maximize the dependence between representations and
labels and minimize the dependence among all samples. Extensive experiments on
Meta-Dataset demonstrate that MOKD can not only achieve better generalization
performance on unseen domains in most cases but also learn better data
representation clusters. The project repository of MOKD is available at:
\href{https://github.com/tmlr-group/MOKD}{this https URL}.

在跨领域少样本分类中，本文通过学习表示来构建度量空间，以测量样本和每个类别原型之间的相似性，并通过双层优化框架提出了一种最大化优化核依赖性（MOKD）的方法，以学习与给定任务标记数据指示的聚类结构相匹配的一组类别特定表示。

MOKD: 通过最大化优化的核依赖进行跨域微调的少样本分类

MOKD: Cross-domain Finetuning for Few-shot Classification via Maximizing  Optimized Kernel Dependence

Quantum machine learning has seen considerable theoretical and practical
developments in recent years and has become a promising area for finding real
world applications of quantum computers. In pursuit of this goal, here we
combine state-of-the-art algorithms and quantum hardware to provide an
experimental demonstration of a quantum machine learning application with
provable guarantees for its performance and efficiency. In particular, we
design a quantum Nearest Centroid classifier, using techniques for efficiently
loading classical data into quantum states and performing distance estimations,
and experimentally demonstrate it on a 11-qubit trapped-ion quantum machine,
matching the accuracy of classical nearest centroid classifiers for the MNIST
handwritten digits dataset and achieving up to 100% accuracy for 8-dimensional
synthetic data.

本篇论文设计一个最近质心分类器，使用最先进的算法和量子硬件来实现量子机器学习应用，实验结果能够在 MNIST 手写数字数据集上匹配经典最近质心分类器的准确性，并在 8 维的人工合成数据上实现了高达 100% 的准确性。