List your company

Slide Editing on TSBench TextEditing

98.28Success Rate (%)TALK-TO-YOUR-SLIDES
63.273672.361881.4590.5382May 16, 2025

Evaluation results

MethodLinks
TALK-TO-YOUR-SLIDESBackbone=GPT-4.1-mini2025.0598.282.63.013.093.543.658.173.351.663.3
TALK-TO-YOUR-SLIDESEvaluation Model=Claude 3.5 Haiku2025.0597.712.552.953.043.493.5456.13.281.622.9
Direct code gen.Backbone=GPT-4.1-mini2025.0576.72001.041.041.0418.551.030.411
Direct code gen.Evaluation Model=Claude 3.5 Haiku2025.0575.91001.0211.0317.921.020.40.9
UI AgentBackbone=GPT-4.1-mini2025.0565.480.480.671.581.481.56110.84101.482.115.9
UI AgentEvaluation Model=Claude 3.5 Haiku2025.0564.620.470.651.551.451.53106.196.32.0312.7