Imagen de Portada

Rethinking Data Curation in LLM Training: Online Reweighting Offers Better Generalization than Offline Methods

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Zhao, Wanru, Chen, Yihong, Tang, Yuzhi, Ma, Wentao, Hu, Shengchao, Hu, Shell Xu, Iacob, Alex, Mehrotra, Abhinav, Lane, Nicholas D.
Formato:	Preprint
Publicado:	2026
Materias:	Machine Learning Artificial Intelligence
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

CLUES: Collaborative High-Quality Data Selection for LLMs via Training Dynamics
por: Zhao, Wanru, et al.
Publicado: (2025)

Photon: Federated LLM Pre-Training
por: Sani, Lorenzo, et al.
Publicado: (2024)

Energy-Based Preference Model Offers Better Offline Alignment than the Bradley-Terry Preference Model
por: Hong, Yuzhong, et al.
Publicado: (2024)

Larger Scale Offers Better Security in the Nakamoto-style Blockchain
por: Hu, Junjie
Publicado: (2025)

LLM Unlearning via Neural Activation Redirection
por: Shen, William F., et al.
Publicado: (2025)

SparsyFed: Sparse Adaptive Federated Training
por: Guastella, Adriano, et al.
Publicado: (2025)

Worldwide Federated Training of Language Models
por: Iacob, Alex, et al.
Publicado: (2024)

Locally Estimated Global Perturbations are Better than Local Perturbations for Federated Sharpness-aware Minimization
por: Fan, Ziqing, et al.
Publicado: (2024)

Advancing AI-assisted Hardware Design with Hierarchical Decentralized Training and Personalized Inference-Time Optimization
por: Chen, Hao Mark, et al.
Publicado: (2025)

Prompt Tuning with Diffusion for Few-Shot Pre-trained Policy Generalization
por: Hu, Shengchao, et al.
Publicado: (2024)

Rethinking the Role of Dynamic Sparse Training for Scalable Deep Reinforcement Learning
por: Ma, Guozheng, et al.
Publicado: (2025)

Pollen: High-throughput Federated Learning Simulation via Resource-Aware Client Placement
por: Sani, Lorenzo, et al.
Publicado: (2023)

FedAnchor: Enhancing Federated Semi-Supervised Learning with Label Contrastive Loss for Unlabeled Clients
por: Qiu, Xinchi, et al.
Publicado: (2024)

Breaking Physical and Linguistic Borders: Multilingual Federated Prompt Tuning for Low-Resource Languages
por: Zhao, Wanru, et al.
Publicado: (2025)

Steppe productivity
por: Tang, Wanru
Publicado: (2026)

Beyond Scaling: Agents Are Heading to the Edge
por: Tian, Chunlin, et al.
Publicado: (2026)

Reward Difference Optimization For Sample Reweighting In Offline RLHF
por: Wang, Shiqi, et al.
Publicado: (2024)

Solving Continual Offline RL through Selective Weights Activation on Aligned Spaces
por: Hu, Jifeng, et al.
Publicado: (2024)

HarmoDT: Harmony Multi-Task Decision Transformer for Offline Reinforcement Learning
por: Hu, Shengchao, et al.
Publicado: (2024)

The Future of Large Language Model Pre-training is Federated
por: Sani, Lorenzo, et al.
Publicado: (2024)

Q-value Regularized Transformer for Offline Reinforcement Learning
por: Hu, Shengchao, et al.
Publicado: (2024)

SkillBrew: Multi-Objective Curation of Skill Banks for LLM Agents
por: Hu, Wentao, et al.
Publicado: (2026)

Analytic Energy-Guided Policy Optimization for Offline Reinforcement Learning
por: Hu, Jifeng, et al.
Publicado: (2025)

Filter-then-Weight: Online Data Selection and Reweighting for LLM Fine-Tuning
por: Wang, Fangxin, et al.
Publicado: (2026)

Noise Consistency Training: A Native Approach for One-Step Generator in Learning Additional Controls
por: Luo, Yihong, et al.
Publicado: (2025)

China’s Gas Development Strategies
por: Shell Centre

Autoguided Online Data Curation for Diffusion Model Training
por: Pais, Valeria, et al.
Publicado: (2025)

Online Pre-Training for Offline-to-Online Reinforcement Learning
por: Shin, Yongjae, et al.
Publicado: (2025)

Task-Aware Harmony Multi-Task Decision Transformer for Offline Reinforcement Learning
por: Fan, Ziqing, et al.
Publicado: (2024)

Response-Conditioned Parallel-to-Sequential Orchestration for Multi-Agent Systems
por: Tastan, Nurbek, et al.
Publicado: (2026)

Conservative Offline Robot Policy Learning via Posterior-Transition Reweighting
por: Zhang, Wanpeng, et al.
Publicado: (2026)

Demystifying Diffusion Objectives: Reweighted Losses are Better Variational Bounds
por: Shi, Jiaxin, et al.
Publicado: (2025)

Enhancing Data Quality in Federated Fine-Tuning of Foundation Models
por: Zhao, Wanru, et al.
Publicado: (2024)

MarcoPolo: A Zero-Permission Attack for Location Type Inference from the Magnetic Field using Mobile Devices
por: Perez, Beatrice, et al.
Publicado: (2026)

Robust Online Selection with Uncertain Offer Acceptance
por: Perez-Salazar, Sebastian, et al.
Publicado: (2021)

Is Mamba Compatible with Trajectory Optimization in Offline Reinforcement Learning?
por: Dai, Yang, et al.
Publicado: (2024)

Combatting Dimensional Collapse in LLM Pre-Training Data via Diversified File Selection
por: Fan, Ziqing, et al.
Publicado: (2025)

Quaternionic Reweighted Amplitude Flow for Phase Retrieval in Image Reconstruction
por: Hu, Ren, et al.
Publicado: (2025)

Bayesian Design Principles for Offline-to-Online Reinforcement Learning
por: Hu, Hao, et al.
Publicado: (2024)

Meta Co-Training: Two Views are Better than One
por: Rothenberger, Jay C., et al.
Publicado: (2023)