qazimbhat1
/

Crystal-based-MLLM-7B

Text Generation

Model card Files Files and versions

Update README.md

#1

by Jason0214 - opened Jun 26, 2024

base: refs/heads/main

←

from: refs/pr/1

Discussion Files changed

Files changed (1) hide show

README.md +2 -2

README.md CHANGED Viewed

@@ -16,9 +16,9 @@ Crystal-based models mimics the training recipie used for Vicuna 7B in LLaVA mul
 | LLM Backbone                      | MME-P   | MME-C  | POPE  | SciQA  | TextVQA |
 |-----------------------------------|---------|--------|-------|--------|---------|
-| CrystalCoder-7B | 1359.83 | 238.92 | 86.182 | 64.15 | 50.39   |
 | CrystalChat-7B  | 1456.53 | **308.21** | 86.96 | 67.77 | **57.84** |
-| Vicuna-7B | **1481.12** | 302.85 | **87.174** | **67.97** | 56.49   |
 *Table: Comparison of different LLM backbones on visual language understanding benchmarks. All models are instruction-tuned on the general domain data (i.e. LLaVA)*

 | LLM Backbone                      | MME-P   | MME-C  | POPE  | SciQA  | TextVQA |
 |-----------------------------------|---------|--------|-------|--------|---------|
+| CrystalCoder-7B | 1359.83 | 238.92 | 86.18 | 64.15 | 50.39   |
 | CrystalChat-7B  | 1456.53 | **308.21** | 86.96 | 67.77 | **57.84** |
+| Vicuna-7B | **1481.12** | 302.85 | **87.17** | **67.97** | 56.49   |
 *Table: Comparison of different LLM backbones on visual language understanding benchmarks. All models are instruction-tuned on the general domain data (i.e. LLaVA)*