Picture for Yidong Wang

Yidong Wang

Visual Thoughts: A Unified Perspective of Understanding Multimodal Chain-of-Thought

Add code
May 21, 2025
Viaarxiv icon

Reasoning on Multiple Needles In A Haystack

Add code
Apr 05, 2025
Viaarxiv icon

A Language Anchor-Guided Method for Robust Noisy Domain Generalization

Add code
Mar 21, 2025
Viaarxiv icon

StepMathAgent: A Step-Wise Agent for Evaluating Mathematical Processes through Tree-of-Error

Add code
Mar 13, 2025
Viaarxiv icon

Masked Autoencoders Are Effective Tokenizers for Diffusion Models

Add code
Feb 05, 2025
Viaarxiv icon

Outcome-Refining Process Supervision for Code Generation

Add code
Dec 19, 2024
Figure 1 for Outcome-Refining Process Supervision for Code Generation
Figure 2 for Outcome-Refining Process Supervision for Code Generation
Figure 3 for Outcome-Refining Process Supervision for Code Generation
Figure 4 for Outcome-Refining Process Supervision for Code Generation
Viaarxiv icon

Learning from "Silly" Questions Improves Large Language Models, But Only Slightly

Add code
Nov 21, 2024
Viaarxiv icon

On the Diversity of Synthetic Data and its Impact on Training Large Language Models

Add code
Oct 19, 2024
Figure 1 for On the Diversity of Synthetic Data and its Impact on Training Large Language Models
Figure 2 for On the Diversity of Synthetic Data and its Impact on Training Large Language Models
Figure 3 for On the Diversity of Synthetic Data and its Impact on Training Large Language Models
Figure 4 for On the Diversity of Synthetic Data and its Impact on Training Large Language Models
Viaarxiv icon

RAGLAB: A Modular and Research-Oriented Unified Framework for Retrieval-Augmented Generation

Add code
Aug 21, 2024
Figure 1 for RAGLAB: A Modular and Research-Oriented Unified Framework for Retrieval-Augmented Generation
Figure 2 for RAGLAB: A Modular and Research-Oriented Unified Framework for Retrieval-Augmented Generation
Figure 3 for RAGLAB: A Modular and Research-Oriented Unified Framework for Retrieval-Augmented Generation
Figure 4 for RAGLAB: A Modular and Research-Oriented Unified Framework for Retrieval-Augmented Generation
Viaarxiv icon

Survey on Knowledge Distillation for Large Language Models: Methods, Evaluation, and Application

Add code
Jul 02, 2024
Viaarxiv icon