Picture for Lei Li

Lei Li

Carnegie Mellon University

Video-MME: The First-Ever Comprehensive Evaluation Benchmark of Multi-modal LLMs in Video Analysis

Add code
May 31, 2024
Figure 1 for Video-MME: The First-Ever Comprehensive Evaluation Benchmark of Multi-modal LLMs in Video Analysis
Figure 2 for Video-MME: The First-Ever Comprehensive Evaluation Benchmark of Multi-modal LLMs in Video Analysis
Figure 3 for Video-MME: The First-Ever Comprehensive Evaluation Benchmark of Multi-modal LLMs in Video Analysis
Figure 4 for Video-MME: The First-Ever Comprehensive Evaluation Benchmark of Multi-modal LLMs in Video Analysis
Viaarxiv icon

DeCo: Decoupling Token Compression from Semantic Abstraction in Multimodal Large Language Models

Add code
May 31, 2024
Figure 1 for DeCo: Decoupling Token Compression from Semantic Abstraction in Multimodal Large Language Models
Figure 2 for DeCo: Decoupling Token Compression from Semantic Abstraction in Multimodal Large Language Models
Figure 3 for DeCo: Decoupling Token Compression from Semantic Abstraction in Multimodal Large Language Models
Figure 4 for DeCo: Decoupling Token Compression from Semantic Abstraction in Multimodal Large Language Models
Viaarxiv icon

3D Neural Edge Reconstruction

Add code
May 29, 2024
Viaarxiv icon

MindMerger: Efficient Boosting LLM Reasoning in non-English Languages

Add code
May 27, 2024
Viaarxiv icon

Scaling Law for Time Series Forecasting

Add code
May 27, 2024
Figure 1 for Scaling Law for Time Series Forecasting
Figure 2 for Scaling Law for Time Series Forecasting
Figure 3 for Scaling Law for Time Series Forecasting
Figure 4 for Scaling Law for Time Series Forecasting
Viaarxiv icon

$T^2$ of Thoughts: Temperature Tree Elicits Reasoning in Large Language Models

Add code
May 23, 2024
Viaarxiv icon

Generative Enzyme Design Guided by Functionally Important Sites and Small-Molecule Substrates

Add code
May 13, 2024
Viaarxiv icon

SurfPro: Functional Protein Design Based on Continuous Surface

Add code
May 07, 2024
Viaarxiv icon

Vibe-Eval: A hard evaluation suite for measuring progress of multimodal language models

Add code
May 03, 2024
Figure 1 for Vibe-Eval: A hard evaluation suite for measuring progress of multimodal language models
Figure 2 for Vibe-Eval: A hard evaluation suite for measuring progress of multimodal language models
Figure 3 for Vibe-Eval: A hard evaluation suite for measuring progress of multimodal language models
Figure 4 for Vibe-Eval: A hard evaluation suite for measuring progress of multimodal language models
Viaarxiv icon

Reka Core, Flash, and Edge: A Series of Powerful Multimodal Language Models

Add code
Apr 18, 2024
Figure 1 for Reka Core, Flash, and Edge: A Series of Powerful Multimodal Language Models
Figure 2 for Reka Core, Flash, and Edge: A Series of Powerful Multimodal Language Models
Figure 3 for Reka Core, Flash, and Edge: A Series of Powerful Multimodal Language Models
Figure 4 for Reka Core, Flash, and Edge: A Series of Powerful Multimodal Language Models
Viaarxiv icon