Dataset · Hugging Face

zeyuanyin/TRIM

Name: TRIM
Creator: zeyuanyin
License: mit
Keywords: task_categories:text-to-3d, task_categories:image-to-3d, license:mit, arxiv:2511.16642, region:us, 3d-generation, gaussian-splatting, diffusion

TRIM: Trajectory Data for Latent Selector Training Denoising trajectory latents and quality scores used to train the latent selector of TRIM (NeurIPS 2025). Code: github.com/zeyuanyin/TRIM. Synthesized with DiffSplat (SD3.5-Medium) on 100 ChatGPT-4o prompts x 64 random seeds, 20 denoising steps per trajectory. Files latent_data_4o_data100_seed64_step{k}in20.pt for k = 0...19: one tensor per denoising step, each of size (100, 64, 4, 16, 32, 32), covering 100… See the full description on the dataset page: https://huggingface.co/datasets/zeyuanyin/TRIM.

View on Hugging FaceReport an issue

Author / organization	zeyuanyin
License (as declared)	mit
Downloads	1K (as reported by Hugging Face)
Likes	0
Created	16 September 2026
Last updated at source	16 September 2026

3d-generationgaussian-splattingdiffusion

Embed this entry

<iframe src="https://silytics.si/embed/item/d08abafa-1242-4512-b202-9f995bf62cf0" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Free, no tracking. More about feeds and embeds.

Popularity, last 3 days

1,025 now (-123 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.

More from zeyuanyin

zeyuanyin/Dome-ObjaverseDataset

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 18 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.