1
Learning with the Nash-Sutcliffe loss
详解Nash-Sutcliffe损失在机器学习中的用法，面向水文预报等回归任务的深度技术指南。
arXiv:2603.00968v2 Announce Type: replace-cross Abstract: The Nash-Sutcliffe efficiency ($\text{NSE}$) is a widely used, positively oriented relative …
2
Comparison of Loss Functions for Robust Deep Learning-based Echocardiography Segmentation when Learning with Partially Labelled Data from Multiple Domains
比较多种损失函数在部分标注多域超声心动图分割中的鲁棒性，为医学AI训练选型提供实证参考。
arXiv:2607.05008v1 Announce Type: cross Abstract: Echocardiography is the first imaging modality used for assessing cardiac function, and accurate seg…
3
HEM: a margin-based loss for visual categorisation tasks
一种新的 margin-based 损失函数 HEM，为视觉分类任务带来更优的判别力与性能提升。
arXiv:2501.12191v2 Announce Type: replace Abstract: Training deep neural networks (DNNs) on classification tasks can be performed with a number of dif…
4
Polynomial Dice Loss for Medical Image Segmentation
提出多项式Dice损失，显著提升医学图像分割精度，算法优化必读
arXiv:2606.23373v1 Announce Type: new Abstract: Medical image segmentation is a fundamental task for medical image processing and computer-assisted in…
5
Towards Robust Training in NNGPT AutoML Pipeline: A Loss-Optimizer Pairing Selection Study
AutoML管道如何挑选损失函数与优化器最佳搭档？这项研究为NNGPT系统找到了稳定训练的配对规律
arXiv:2606.20933v1 Announce Type: new Abstract: The choice of loss function and optimizer is an important decision, that shapes further model training…
6
Generalized Kullback-Leibler Divergence Loss
TPAMI 2026最新论文对经典KL散度损失进行广义化与解耦，首次系统性统一多种变体并理论证明收敛性，是损失函数设计的重要突破。
arXiv:2503.08038v2 Announce Type: replace Abstract: In this paper, we delve deeper into the Kullback-Leibler (KL) Divergence loss and mathematically p…
7
Distributional Loss for Robust Classification
提出一种新型分布损失函数，旨在提升分类模型对噪声和扰动的鲁棒性，理论贡献值得关注。
arXiv:2606.13223v1 Announce Type: cross Abstract: This paper proposes a novel loss concept for supervised classification tasks. Rather than enforcing …
8
Investigating the Histogram Loss in Regression
直方图损失在回归任务中的应用原理与效果深入探究，为损失函数设计提供新视角
arXiv:2402.13425v3 Announce Type: replace Abstract: It is becoming increasingly common in regression to train neural networks that model the entire di…
9
LK Losses: Direct Acceptance Rate Optimization for Speculative Decoding
投机解码新突破：提出LK损失函数直接优化接受率，加速大模型推理生成
arXiv:2602.23881v2 Announce Type: replace Abstract: Speculative decoding accelerates autoregressive large language model (LLM) inference by using a li…
10
The Bridge-Garden Dilemma in LLM Distillation: Why Mixing Hard and Soft Labels Works
揭秘LLM蒸馏中混合硬软标签优于纯软标签的深层原因，实验与理论兼备。
arXiv:2605.26246v1 Announce Type: new Abstract: Knowledge distillation (KD) transfers knowledge from a large teacher model to a smaller student. In la…
11
DEL: Digit Entropy Loss for Numerical Learning of Large Language Models
提出数字熵损失函数，专门提升大语言模型对数字的处理能力，让AI更懂数值。
arXiv:2605.20369v1 Announce Type: cross Abstract: Number prediction stands as a fundamental capability of large language models (LLMs) in mathematical…
12
LLMs on the Line: Data Determines Loss-to-Loss Scaling Laws
ICML 2025收录，揭示数据质量如何决定大模型损失与缩放定律的深层关系。
arXiv:2502.12120v3 Announce Type: replace Abstract: Scaling laws guide the development of large language models (LLMs) by offering estimates for the o…
13
Calibeating for general proper losses: A Bregman divergence approach
将校准方法从Brier与对数损失推广至通用适当损失函数，基于Bregman散度与遗憾最小化框架创新。
arXiv:2605.17269v1 Announce Type: new Abstract: This work introduces a general framework for calibeating based on regret minimization. As compared to …
14
$f$-Trajectory Balance: A Loss Family for Tuning GFlowNets, Generative Models, and LLMs with Off- and On-Policy Data
提出f-轨迹平衡损失族，统一了GFlowNets和LLM的on/off-policy训练，梯度对应KL散度，低方差高效。
arXiv:2605.15417v1 Announce Type: cross Abstract: In GFlowNets and variational inference, it has been shown that the mean square error between target …
15
Don't Stop Me Yet: Sampling Loss Minima via Dissipative Riemannian Mechanics
用物理视角破解神经网络损失函数最小值采样难题，耗散黎曼力学方法创新点突出。
arXiv:2605.15459v1 Announce Type: new Abstract: The minima of modern neural network loss functions are typically not isolated, rather they form connec…
16
Tube Loss: A Novel Approach for Prediction Interval Estimation
新型Tube Loss损失函数，同时估计回归预测区间上下界，质量优于现有方法。
arXiv:2412.06853v4 Announce Type: replace-cross Abstract: This paper proposes a novel loss function, called 'Tube Loss', for simultaneous estimation o…