NVILA-Lite-2B-DATA_SCALE_EXP_80K-20251108_180221

NVILA-Lite-2B — Data-Scale Fine-tune (80k)

Fine-tune of Efficient-Large-Model/NVILA-Lite-2B on a 5-dataset spatial-reasoning mixture (SAT + RoboSpatial + SPAR-7M + RefSpatial + PRISM), with the training-sample budget set to 80k.

This checkpoint is one of the data-scale variants released alongside the paper "Why Far Looks Up: Probing Spatial Representation in Vision-Language Models" (see HF Collection).

• 📄 Paper: arXiv:2605.30161
• 🌐 Project page: https://cheolhong0916.github.io/whyfarlooksup.github.io/
• 💻 Code: https://github.com/cheolhong0916/contrastive-probing
• 📊 Diagnostic benchmark: cubec/spatialtunnel

Loading

This checkpoint inherits the architecture of Efficient-Large-Model/NVILA-Lite-2B. You can load it using the transformers library. For usage details, refer to the interface of the base model. Note that some NVILA variants may require the VILA repository for full compatibility.

Citation

If you use this checkpoint, please cite both our paper and the original NVILA-Lite-2B paper.

Our paper (this checkpoint family):

@article{min2026whyfarlooksup,
  title   = {Why Far Looks Up: Probing Spatial Representation in Vision-Language Models},
  author  = {Min, Cheolhong and Jung, Jaeyun and Lee, Daeun and Jeon, Hyeonseong and
             Su, Yu and Tremblay, Jonathan and Song, Chan Hee and Park, Jaesik},
  journal = {arXiv preprint arXiv:2605.30161},
  year    = {2026},
}

Original NVILA-Lite-2B (NVILA: Efficient Frontier Visual Language Models):

@article{liu2024nvila,
  title   = {NVILA: Efficient Frontier Visual Language Models},
  author  = {Liu, Zhijian and Zhu, Ligeng and Shi, Baifeng and others},
  journal = {arXiv preprint arXiv:2412.04468},
  year    = {2024},
}