Copertina

Beyond Loss Values: Robust Dynamic Pruning via Loss Trajectory Alignment

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Qin, Huaiyuan, Yang, Muli, Goenawan, Gabriel James, Wang, Kai, Wang, Zheng, Hu, Peng, Peng, Xi, Zhu, Hongyuan
Natura:	Preprint
Pubblicazione:	2026
Soggetti:	Computer Vision and Pattern Recognition Machine Learning
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Attention Transfer Is Not Universally Effective for Vision Transformers
di: Qin, Huaiyuan, et al.
Pubblicazione: (2026)

Beyond Instance Consistency: Investigating View Diversity in Self-supervised Learning
di: Qin, Huaiyuan, et al.
Pubblicazione: (2025)

SDGBiasBench: Benchmarking and Mitigating Vision--Language Models' Biases in Sustainable Development Goals
di: Lin, Zihang, et al.
Pubblicazione: (2026)

Your AI-Generated Image Detector Can Secretly Achieve SOTA Accuracy, If Calibrated
di: Yang, Muli, et al.
Pubblicazione: (2026)

Batch Loss Score for Dynamic Data Pruning
di: Zhou, Qing, et al.
Pubblicazione: (2026)

Enhancing Performance of Point Cloud Completion Networks with Consistency Loss
di: Wijaya, Kevin Tirta, et al.
Pubblicazione: (2024)

PointCloud-Text Matching: Benchmark Datasets and a Baseline
di: Feng, Yanglin, et al.
Pubblicazione: (2024)

Mitigating Information Loss under High Pruning Rates for Efficient Large Vision Language Models
di: Fu, Mingyu, et al.
Pubblicazione: (2025)

Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
di: Alwis, Praditha, et al.
Pubblicazione: (2026)

Robust Duality Learning for Unsupervised Visible-Infrared Person Re-Identification
di: Li, Yongxiang, et al.
Pubblicazione: (2025)

ASTM :Autonomous Smart Traffic Management System Using Artificial Intelligence CNN and LSTM
di: Goenawan, Christofel Rio
Pubblicazione: (2024)

LBL: Logarithmic Barrier Loss Function for One-class Classification
di: Jiang, Ziyang, et al.
Pubblicazione: (2023)

Distill the Best, Ignore the Rest: Improving Dataset Distillation with Loss-Value-Based Pruning
di: Moser, Brian B., et al.
Pubblicazione: (2024)

VoxelSplat: Dynamic Gaussian Splatting as an Effective Loss for Occupancy and Flow Prediction
di: Zhu, Ziyue, et al.
Pubblicazione: (2025)

Misalignment-Robust Frequency Distribution Loss for Image Transformation
di: Ni, Zhangkai, et al.
Pubblicazione: (2024)

DynaPose4D: High-Quality 4D Dynamic Content Generation via Pose Alignment Loss
di: Yang, Jing, et al.
Pubblicazione: (2025)

A Turn Toward Better Alignment: Few-Shot Generative Adaptation with Equivariant Feature Rotation
di: Xu, Chenghao, et al.
Pubblicazione: (2025)

MADTP: Multimodal Alignment-Guided Dynamic Token Pruning for Accelerating Vision-Language Transformer
di: Cao, Jianjian, et al.
Pubblicazione: (2024)

Dynamic Loss Decay based Robust Oriented Object Detection on Remote Sensing Images with Noisy Labels
di: Liu, Guozhang, et al.
Pubblicazione: (2024)

Global Truncated Loss Minimization for Robust and Threshold-Resilient Geometric Estimation
di: Huang, Tianyu, et al.
Pubblicazione: (2026)

The Geometry of Robustness: Optimizing Loss Landscape Curvature and Feature Manifold Alignment for Robust Finetuning of Vision-Language Models
di: Chopra, Shivang, et al.
Pubblicazione: (2026)

Multimodal-Guided Dynamic Dataset Pruning for Robust and Efficient Data-Centric Learning
di: Yang, Suorong, et al.
Pubblicazione: (2025)

Diagnosing and Improving Diffusion Models by Estimating the Optimal Loss Value
di: Xu, Yixian, et al.
Pubblicazione: (2025)

FlowLoss: Dynamic Flow-Conditioned Loss Strategy for Video Diffusion Models
di: Wu, Kuanting, et al.
Pubblicazione: (2025)

Enhancing Certifiable Semantic Robustness via Robust Pruning of Deep Neural Networks
di: Hu, Hanjiang, et al.
Pubblicazione: (2025)

EgoVIS@CVPR: PAIR-Net: Enhancing Egocentric Speaker Detection via Pretrained Audio-Visual Fusion and Alignment Loss
di: Wang, Yu, et al.
Pubblicazione: (2025)

Reframing Long-Tailed Learning via Loss Landscape Geometry
di: Chen, Shenghan, et al.
Pubblicazione: (2026)

Fewer is More: A Deep Graph Metric Learning Perspective Using Fewer Proxies
di: Zhu, Yuehua, et al.
Pubblicazione: (2020)

LMPT: Prompt Tuning with Class-Specific Embedding Loss for Long-tailed Multi-Label Visual Recognition
di: Xia, Peng, et al.
Pubblicazione: (2023)

Representation Fréchet Loss for Visual Generation
di: Yang, Jiawei, et al.
Pubblicazione: (2026)

Style Ambiguity Loss Using CLIP
di: Baker, James
Pubblicazione: (2024)

Rediscovering BCE Loss for Uniform Classification
di: Li, Qiufu, et al.
Pubblicazione: (2024)

Enhancing CLIP Robustness via Cross-Modality Alignment
di: Zhu, Xingyu, et al.
Pubblicazione: (2025)

RobustNeRF: Ignoring Distractors with Robust Losses
di: Sabour, Sara, et al.
Pubblicazione: (2023)

Noisy-Correspondence Learning for Text-to-Image Person Re-identification
di: Qin, Yang, et al.
Pubblicazione: (2023)

Fourier Amplitude and Correlation Loss: Beyond Using L2 Loss for Skillful Precipitation Nowcasting
di: Yan, Chiu-Wai, et al.
Pubblicazione: (2024)

Adversarial-Robustness-Guided Graph Pruning
di: Wang, Yongyu
Pubblicazione: (2024)

Adaptive Multi-Modal Cross-Entropy Loss for Stereo Matching
di: Xu, Peng, et al.
Pubblicazione: (2023)

Gradient-Aware Logit Adjustment Loss for Long-tailed Classifier
di: Zhang, Fan, et al.
Pubblicazione: (2024)

Robust Fuzzy Multi-view Learning under View Conflict
di: Duan, Siyuan, et al.
Pubblicazione: (2026)