Imagen de Portada

WILDCHAT-50M: A Deep Dive Into the Role of Synthetic Data in Post-Training

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Feuer, Benjamin, Hegde, Chinmay
Formato:	Preprint
Publicado:	2025
Materias:	Machine Learning Computation and Language
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

ArcheType: A Novel Framework for Open-Source Column Type Annotation using Large Language Models
por: Feuer, Benjamin, et al.
Publicado: (2023)

SELECT: A Large-Scale Benchmark of Data Curation Strategies for Image Classification
por: Feuer, Benjamin, et al.
Publicado: (2024)

MARVIS: Modality Adaptive Reasoning over VISualizations
por: Feuer, Benjamin, et al.
Publicado: (2025)

Hidden in the Noise: Two-Stage Robust Watermarking for Images
por: Arabi, Kasra, et al.
Publicado: (2024)

Towards a Theoretical Understanding of Synthetic Data in LLM Post-Training: A Reverse-Bottleneck Perspective
por: Gan, Zeyu, et al.
Publicado: (2024)

Part I: Tricks or Traps? A Deep Dive into RL for LLM Reasoning
por: Liu, Zihe, et al.
Publicado: (2025)

Activation Steering for Synthetic Data Generation: The Role of Diversity in Downstream Safety Detection
por: Deshpande, Vijeta, et al.
Publicado: (2026)

TuneTables: Context Optimization for Scalable Prior-Data Fitted Networks
por: Feuer, Benjamin, et al.
Publicado: (2024)

Accelerating Direct Preference Optimization with Prefix Sharing
por: Wang, Franklin, et al.
Publicado: (2024)

Does Training on Synthetic Data Make Models Less Robust?
por: Zhang, Lingze, et al.
Publicado: (2025)

RouteNator: A Router-Based Multi-Modal Architecture for Generating Synthetic Training Data for Function Calling LLMs
por: Belavadi, Vibha, et al.
Publicado: (2025)

Synthetic Sandbox for Training Machine Learning Engineering Agents
por: Zhou, Yuhang, et al.
Publicado: (2026)

LiveBench: A Challenging, Contamination-Limited LLM Benchmark
por: White, Colin, et al.
Publicado: (2024)

Generalizable LLM Learning of Graph Synthetic Data with Post-training Alignment
por: Zhang, Yizhuo, et al.
Publicado: (2025)

Consolidating Rewarded Perturbations for LLM Post-Training
por: Zhang, Zheyu, et al.
Publicado: (2026)

How Bad is Training on Synthetic Data? A Statistical Analysis of Language Model Collapse
por: Seddik, Mohamed El Amine, et al.
Publicado: (2024)

Accelerating RL Post-Training Rollouts via System-Integrated Speculative Decoding
por: Iso, Hayate, et al.
Publicado: (2026)

FineInstructions: Scaling Synthetic Instructions to Pre-Training Scale
por: Patel, Ajay, et al.
Publicado: (2026)

Efficacy of Synthetic Data as a Benchmark
por: Maheshwari, Gaurav, et al.
Publicado: (2024)

Prompt Curriculum Learning for Efficient LLM Post-Training
por: Gao, Zhaolin, et al.
Publicado: (2025)

Synthetic vs. Gold: The Role of LLM Generated Labels and Data in Cyberbullying Detection
por: Kazemi, Arefeh, et al.
Publicado: (2025)

The Limits of Preference Data for Post-Training
por: Zhao, Eric, et al.
Publicado: (2025)

DataDreamer: A Tool for Synthetic Data Generation and Reproducible LLM Workflows
por: Patel, Ajay, et al.
Publicado: (2024)

Lightweight Safety Guardrails via Synthetic Data and RL-guided Adversarial Training
por: Ilin, Aleksei, et al.
Publicado: (2025)

Diving into Self-Evolving Training for Multimodal Reasoning
por: Liu, Wei, et al.
Publicado: (2024)

Training Language Models on Synthetic Edit Sequences Improves Code Synthesis
por: Piterbarg, Ulyana, et al.
Publicado: (2024)

Rethinking Local Learning: A Cheaper and Faster Recipe for LLM Post-Training
por: Shi, Hengyu, et al.
Publicado: (2026)

Scaling Laws for Post Training Quantized Large Language Models
por: Xu, Zifei, et al.
Publicado: (2024)

XL-Suite: Cross-Lingual Synthetic Training and Evaluation Data for Open-Ended Generation
por: Iyer, Vivek, et al.
Publicado: (2025)

Synthetic Text Generation for Training Large Language Models via Gradient Matching
por: Nguyen, Dang, et al.
Publicado: (2025)

A Unified View of Delta Parameter Editing in Post-Trained Large-Scale Models
por: Tang, Qiaoyu, et al.
Publicado: (2024)

Modifying Large Language Model Post-Training for Diverse Creative Writing
por: Chung, John Joon Young, et al.
Publicado: (2025)

Knowledge Graph-Assisted LLM Post-Training for Enhanced Legal Reasoning
por: Song, Dezhao, et al.
Publicado: (2026)

Efficient Post-Training Pruning of Large Language Models with Statistical Correction
por: Yu, Peiqi, et al.
Publicado: (2026)

Learning to Reason Efficiently with A* Post-Training
por: Opedal, Andreas, et al.
Publicado: (2026)

Post-LayerNorm Is Back: Stable, ExpressivE, and Deep
por: Chen, Chen, et al.
Publicado: (2026)

Beyond Contrastive Learning: Synthetic Data Enables List-wise Training with Multiple Levels of Relevance
por: Esfandiarpoor, Reza, et al.
Publicado: (2025)

Safety Training Persists Through Helpfulness Optimization in LLM Agents
por: Plaut, Benjamin
Publicado: (2026)

CPTQuant - A Novel Mixed Precision Post-Training Quantization Techniques for Large Language Models
por: Nanda, Amitash, et al.
Publicado: (2024)

Towards Active Synthetic Data Generation for Finetuning Language Models
por: Kessler, Samuel, et al.
Publicado: (2025)