Picture for Yi R. Fung

Yi R. Fung

Scaling Laws of Synthetic Data for Language Models

Add code
Mar 26, 2025
Viaarxiv icon

MimeQA: Towards Socially-Intelligent Nonverbal Foundation Models

Add code
Feb 23, 2025
Viaarxiv icon

The Law of Knowledge Overshadowing: Towards Understanding, Predicting, and Preventing LLM Hallucination

Add code
Feb 22, 2025
Figure 1 for The Law of Knowledge Overshadowing: Towards Understanding, Predicting, and Preventing LLM Hallucination
Figure 2 for The Law of Knowledge Overshadowing: Towards Understanding, Predicting, and Preventing LLM Hallucination
Figure 3 for The Law of Knowledge Overshadowing: Towards Understanding, Predicting, and Preventing LLM Hallucination
Figure 4 for The Law of Knowledge Overshadowing: Towards Understanding, Predicting, and Preventing LLM Hallucination
Viaarxiv icon

Knowledge Overshadowing Causes Amalgamated Hallucination in Large Language Models

Add code
Jul 10, 2024
Viaarxiv icon

MACAROON: Training Vision-Language Models To Be Your Engaged Partners

Add code
Jun 20, 2024
Viaarxiv icon

From Pixels to Insights: A Survey on Automatic Chart Understanding in the Era of Large Foundation Models

Add code
Mar 25, 2024
Viaarxiv icon

LEMMA: Towards LVLM-Enhanced Multimodal Misinformation Detection with External Knowledge Augmentation

Add code
Feb 19, 2024
Viaarxiv icon

Persona-DB: Efficient Large Language Model Personalization for Response Prediction with Collaborative Data Refinement

Add code
Feb 16, 2024
Viaarxiv icon

If LLM Is the Wizard, Then Code Is the Wand: A Survey on How Code Empowers Large Language Models to Serve as Intelligent Agents

Add code
Jan 08, 2024
Figure 1 for If LLM Is the Wizard, Then Code Is the Wand: A Survey on How Code Empowers Large Language Models to Serve as Intelligent Agents
Figure 2 for If LLM Is the Wizard, Then Code Is the Wand: A Survey on How Code Empowers Large Language Models to Serve as Intelligent Agents
Figure 3 for If LLM Is the Wizard, Then Code Is the Wand: A Survey on How Code Empowers Large Language Models to Serve as Intelligent Agents
Figure 4 for If LLM Is the Wizard, Then Code Is the Wand: A Survey on How Code Empowers Large Language Models to Serve as Intelligent Agents
Viaarxiv icon

Do LVLMs Understand Charts? Analyzing and Correcting Factual Errors in Chart Captioning

Add code
Dec 15, 2023
Figure 1 for Do LVLMs Understand Charts? Analyzing and Correcting Factual Errors in Chart Captioning
Figure 2 for Do LVLMs Understand Charts? Analyzing and Correcting Factual Errors in Chart Captioning
Figure 3 for Do LVLMs Understand Charts? Analyzing and Correcting Factual Errors in Chart Captioning
Figure 4 for Do LVLMs Understand Charts? Analyzing and Correcting Factual Errors in Chart Captioning
Viaarxiv icon