Vision-Language Modeling on LiveB, LWilder, and LCOCO Combined
20.1Avg Relative ImprovementCASP_GPTQ
Evaluation results
| Method | Links | |
|---|---|---|
| CASP_GPTQBit=3, Base Model=LLaVA1.5-7B2025.03 | 20.1 | |
| CASP_AQLMBit=3, Base Model=LLaVA1.5-7B2025.03 | 3.4 | |
| CASP_QuIP#Bit=3, Base Model=LLaVA1.5-7B2025.03 | 0.01 |