List your company

Task success rate on DMV-Bench J=15 1.0

75.1Task Success RateDualMem
-3.00417.27337.5557.827Jun 25, 2026

Evaluation results

MethodLinks
DualMemBack-end=Qwen2.5-VL-7B, Steps (stp)=359–394, Probe counts (nr)=10,3072026.0675.1
DualMemBack-end=Gemini 2.5 Flash, Steps (stp)=359–394, Probe counts (nr)=28,7102026.0671.3
M2ABack-end=Qwen2.5-VL-7B, Steps (stp)=359–394, Probe counts (nr)=10,3072026.0662.6
CaptionBack-end=Qwen2.5-VL-7B, Steps (stp)=359–394, Probe counts (nr)=10,3072026.0662.3
M2ABack-end=Gemini 2.5 Flash, Steps (stp)=359–394, Probe counts (nr)=28,7102026.0659.6
CaptionBack-end=Gemini 2.5 Flash, Steps (stp)=359–394, Probe counts (nr)=28,7102026.0650.7
MMABack-end=Qwen2.5-VL-7B, Steps (stp)=359–394, Probe counts (nr)=10,3072026.0639.4
WorldMMBack-end=Gemini 2.5 Flash, Steps (stp)=359–394, Probe counts (nr)=28,7102026.0638.4
MMABack-end=Gemini 2.5 Flash, Steps (stp)=359–394, Probe counts (nr)=28,7102026.0633.6
WorldMMBack-end=Qwen2.5-VL-7B, Steps (stp)=359–394, Probe counts (nr)=10,3072026.0629.7
NoMemoryBack-end=Qwen2.5-VL-7B, Steps (stp)=359–394, Probe counts (nr)=10,3072026.060
TextOnlyBack-end=Qwen2.5-VL-7B, Steps (stp)=359–394, Probe counts (nr)=10,3072026.060
NoMemoryBack-end=Gemini 2.5 Flash, Steps (stp)=359–394, Probe counts (nr)=28,7102026.060
TextOnlyBack-end=Gemini 2.5 Flash, Steps (stp)=359–394, Probe counts (nr)=28,7102026.060