List your company
ResearchFormerly WizWand

Referring Expression Comprehension on VRSBench (test)

51.36Accuracy@0.5RS-HyRe-R1
15.511224.818134.12543.4319Apr 19, 2026
Updated 6mo ago

Evaluation results

MethodLinks
RS-HyRe-R1Base LLM=Qwen2.5-3B, Evaluation Protocol=RL Train2026.0451.3632.25
GeoChatBase LLM=Vicuna1.5-7B, Evaluation Protocol=Full Amount Fine-tune (36,313 samples)2026.0449.819.9
Geo-R1-RECBase LLM=Qwen2.5-3B, Evaluation Protocol=RL Train2026.0449.629.77
Geo-R1Base LLM=Qwen2.5-7B, Evaluation Protocol=RL Train2026.0442.0317.62
GeoReasonBase LLM=Qwen2.5-7B, Evaluation Protocol=RL Train2026.0441.6723.5
LLaVA-1.5Base LLM=Vicuna1.5-7B, Evaluation Protocol=Full Amount Fine-tune (36,313 samples)2026.0441.613.6
Qwen2.5-VL 7BBase LLM=Qwen2.5-7B, Evaluation Protocol=Zero-shot Baseline2026.0441.2823.32
VLM-R1-RECBase LLM=Qwen2.5-3B, Evaluation Protocol=RL Train2026.0438.7721.18
Geo-R1-OVDBase LLM=Qwen2.5-3B, Evaluation Protocol=RL Train2026.0438.2220.9
MiniGPT-v2Base LLM=Vicuna1.5-7B, Evaluation Protocol=Full Amount Fine-tune (36,313 samples)2026.0435.816.8
Qwen2.5-VL-SFTBase LLM=Qwen2.5-3B, Evaluation Protocol=RS-task Dataset Fine-tune (1600 samples)2026.0435.3918.58
R1-VLBase LLM=Qwen2-7B, Evaluation Protocol=RL Train2026.0435.0219.28
VLM-R1-OVDBase LLM=Qwen2.5-3B, Evaluation Protocol=RL Train2026.0433.5117.68
Mini-GeminiBase LLM=Gemma-7B, Evaluation Protocol=Full Amount Fine-tune (36,313 samples)2026.0430.16.8
TinyRS-R1Base LLM=Qwen2-2B, Evaluation Protocol=RL Train2026.0423.9410.07
Qwen2.5-VL 3BBase LLM=Qwen2.5-3B, Evaluation Protocol=Zero-shot Baseline2026.0416.897.36