Tag
The paper proposes a comprehensive data generation strategy combining traditional augmentation, CutMix, MixUp, and DCGAN-generated images to address data insufficiency and class imbalance for olive tree disease classification, evaluated using EfficientNetV2B0 and showing notable accuracy improvements.
一篇关于大语言模型数据优化的综述论文,系统梳理了数据感知维度(质量、难度、多样性、冗余性)以及数据选择、增强、重加权等优化技术,并探讨了该领域面临的挑战与未来方向。
CFBrainAug 提出一种模块度约束的反事实脑网络增强框架,将数据增强与反事实解释统一起来,在 ABIDE I 数据集上以轻量级 SVM 实现了具有竞争力的自闭症诊断精度与内在可解释性。
本文提出结合蒙特卡洛 dropout 不确定性量化的条件生成对抗网络数据增强框架,用于 Hawk T1A 飞机结构健康监测的合成信号生成与损伤分类,并与实验测量数据进行多维度比较验证。
本文提出一种基于大语言模型和文档级话语结构解析的结构保持式数据增强框架,通过识别句子间修辞关系并替换关键短语,在保持话语结构的同时实现主题偏移,从而在文本摘要和问答任务上提升模型性能。
提出 TD-Aug 方法,通过从几何 horizon 模型(基于 TD-Flows 训练)采样来想象新的目标状态,作为目标条件行为克隆的数据增强,显著提升了策略在 OGBench 拼接任务中的零样本组合泛化能力。
提出SharP框架,利用大语言模型结合软硬提示进行边界感知的数据增强,以解决低资源条件下的假新闻检测问题。
Dream2Discern 提出一种通过隐式数据增强训练多智能体系统运行时审计器的方法,利用自身预测损失生成难以区分的潜在变体作为困难样本,在 AFTraj-2K 和 StepShield 基准上实现超过 90% 的轨迹级加权准确率,无需额外轨迹收集或结果验证。
提出一种通过LLM生成方言文本并用标准语TTS合成伪方言语音的零真实方言数据增强方法,并引入中间标准文本预测以缩小语义差距,提升语音语言模型在日语、德语和中文方言上的翻译表现。
The paper explores using code-switched text to align representations in small language models under data constraints, demonstrating that a curriculum involving code-switching improves multilingual performance.
This paper presents a synthetic training pipeline that uses perturbed public documents to generate context-dependent training data, significantly improving large language models' performance on context learning tasks like CL-bench without human annotation.
The paper introduces MFP, a method using role-aware Morgan fingerprints for predicting reaction yields in chemistry, achieving high accuracy and faster training compared to existing methods like YieldBERT and GNAN.
REPAIR is a self-evolving data augmentation framework for scientific dense retrievers that resolves long-tail confusion via fact-verified iterative refinement, demonstrating significant performance improvements on materials science and biomedical benchmarks.
This paper proposes a coverage-aware virtual IMU augmentation framework to address data scarcity in human activity recognition, generating and selecting synthetic sensor data to improve model performance with limited labeled data.
The paper proposes CoMA-DiT, a bidirectional cross-modal Diffusion Transformer for latent augmentation in multimodal brain state decoding, which enhances performance in tasks like auditory attention decoding and emotion recognition by using paired modalities as mutual supervisory signals.
KItCAT introduces a lightweight training strategy for auto-regressive LLMs that uses input corruption to generate diverse training inputs, improving knowledge injection from niche documents without costly paraphrasing.
This paper investigates scenarios where adding correctly labeled data can harm model performance, introducing insertion-stability and examining the limits of dimension-based theory in machine learning generalization.
This paper proposes a geometric filtering framework that selects high-quality LLM-generated samples by evaluating their Euclidean distance to real class examples in an embedding space, improving few-shot text classification performance by +2.61 percentage points over SMOTE.
This paper proposes a simulation-based methodology to generate augmented traffic datasets by replacing physical sensors with virtual ones, extending sensor coverage in urban networks while preserving traffic patterns.
This paper introduces techniques for the second MLC-SLM Challenge, including random leading-silence cropping and synthetic data generation to enhance multilingual conversational speech tasks, achieving improved accuracy and reduced error rates.