List your company
ResearchFormerly WizWand

Image-to-LaTeX Table Conversion on Table2LaTeX simple (test)

83.5TEDSQwen2.5-VL-7B-CSPO
24.7439.99555.2570.505Apr 13, 2026
Updated 6mo ago

Evaluation results

MethodLinks
Qwen2.5-VL-7B-CSPOModel Backbone=Qwen2.5-VL, Parameter Count=7B, Reinforcement Learning Strategy=CSPO2026.0483.537.16978.446
Qwen2.5-VL-7B-GRPOModel Backbone=Qwen2.5-VL, Parameter Count=7B, Reinforcement Learning Strategy=GRPO2026.0483.134.764.779.844.8
Qwen2.5-VL-7B-VSGRPO*Model Backbone=Qwen2.5-VL, Parameter Count=7B, Reinforcement Learning Strategy=VSGRPO2026.0482.933.966.980.242.3
Qwen2.5-VL-3B-CSPOModel Backbone=Qwen2.5-VL, Parameter Count=3B, Reinforcement Learning Strategy=CSPO2026.0482.32664.374.433.9
Qwen2.5-VL-3B-VSGRPO*Model Backbone=Qwen2.5-VL, Parameter Count=3B, Reinforcement Learning Strategy=VSGRPO2026.0481.522.663.374.230.7
Qwen2.5-VL-3B-GRPOModel Backbone=Qwen2.5-VL, Parameter Count=3B, Reinforcement Learning Strategy=GRPO2026.0481.524.458.375.632.1
Qwen2.5-VL-72BModel Backbone=Qwen2.5-VL, Parameter Count=72B2026.0474.77.564.966.511.7
Gemini-2.5 FlashModel Backbone=Gemini-2.5 Flash2026.0471.510.57278.412.9
GPT-4oModel Backbone=GPT-4o2026.0470.63.460.966.14.6
Qwen2.5-VL-7BModel Backbone=Qwen2.5-VL, Parameter Count=7B2026.0466.4454.658.75.4
Qwen2.5-VL-3BModel Backbone=Qwen2.5-VL, Parameter Count=3B2026.0463.5239.144.23
NougatModel Backbone=Nougat2026.04270.420.62.85.7