Thank you for your work. I am a student at a university and also researching in this field. I would like to compare with your method, so may I ask if you can make your test dataset public? If it is inconvenient, could you possibly release the test lists for ABO and Objaverse? It would be most helpful if the evaluation code could also be made public, as I am unsure whether I can align with your evaluation strategy.