Picture for Heng Ji

Heng Ji

Knowledge Overshadowing Causes Amalgamated Hallucination in Large Language Models

Add code
Jul 10, 2024
Viaarxiv icon

A Single Transformer for Scalable Vision-Language Modeling

Add code
Jul 08, 2024
Viaarxiv icon

Eliminating Position Bias of Language Models: A Mechanistic Approach

Add code
Jul 01, 2024
Viaarxiv icon

FIRST: Faster Improved Listwise Reranking with Single Token Decoding

Add code
Jun 21, 2024
Figure 1 for FIRST: Faster Improved Listwise Reranking with Single Token Decoding
Figure 2 for FIRST: Faster Improved Listwise Reranking with Single Token Decoding
Figure 3 for FIRST: Faster Improved Listwise Reranking with Single Token Decoding
Figure 4 for FIRST: Faster Improved Listwise Reranking with Single Token Decoding
Viaarxiv icon

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

Add code
Jun 20, 2024
Viaarxiv icon

AGRaME: Any-Granularity Ranking with Multi-Vector Embeddings

Add code
May 23, 2024
Viaarxiv icon

GLaD: Synergizing Molecular Graphs and Language Descriptors for Enhanced Power Conversion Efficiency Prediction in Organic Photovoltaic Devices

Add code
May 23, 2024
Viaarxiv icon

Weak-to-Strong Extrapolation Expedites Alignment

Add code
Apr 25, 2024
Viaarxiv icon

Text-Based Reasoning About Vector Graphics

Add code
Apr 10, 2024
Figure 1 for Text-Based Reasoning About Vector Graphics
Figure 2 for Text-Based Reasoning About Vector Graphics
Figure 3 for Text-Based Reasoning About Vector Graphics
Figure 4 for Text-Based Reasoning About Vector Graphics
Viaarxiv icon

Towards Better Generalization in Open-Domain Question Answering by Mitigating Context Memorization

Add code
Apr 02, 2024
Viaarxiv icon