frallebini/gmnerf

Weight Space Representation Learning on Diverse NeRF Architectures (ICLR 2026)

[arXiv | project page]

NeRF weights

Main dataset structure:

.
└── nerf/
    └── shapenet/
        ├── hash/
        │   └── class_id/
        │       └── nerf_id/
        │           ├── train/
        │           │   └── *.png              # object views used to train the NeRF
        │           ├── grid.pth               # nerfacc-like occupancy grid parameters
        │           ├── nerf_weights.pth       # nerfacc-like NeRF parameters
        │           └── transforms_train.json  # camera poses
        ├── mlp/
        │   └── class_id/
        │       └── nerf_id/
        │           ├── train/
        │           │   └── *.png                 
        │           ├── grid.pth                
        │           ├── nerf_weights.pth        
        │           └── transforms_train.json
        ├── triplane/
        │   └── class_id/
        │       └── nerf_id/
        │           ├── train/
        │           │   └── *.png                 
        │           ├── grid.pth                
        │           ├── nerf_weights.pth        
        │           └── transforms_train.json
        ├── test.txt                           # test split
        ├── train.txt                          # training split
        └── val.txt                            # validation split

Unseen architectures (nerf/shapenet/hash_unseen/*, nerf/shapenet/mlp_unseen/*, and nerf/shapenet/triplane_unseen/*) and Objaverse NeRFs (nerf/objaverse/*) have analogous directory structures.

NeRF graphs

Main dataset structure:

.
└── graph/
    └── shapenet/
        ├── hash/
        │   ├── test/
        │   │   └── *.pt  # torch_geometric-like graph data
        │   ├── train/
        │   │   └── *.pt
        │   └── val/
        │       └── *.pt
        ├── mlp/
        │   ├── test/
        │   │   └── *.pt
        │   ├── train/
        │   │   └── *.pt
        │   └── val/
        │       └── *.pt
        └── triplane/
            ├── test/
            │   └── *.pt
            ├── train/
            │   └── *.pt
            └── val/
                └── *.pt

Unseen architectures (graph/shapenet/hash_unseen/*, graph/shapenet/mlp_unseen/*, and graph/shapenet/triplane_unseen/*) and Objaverse NeRFs (graph/objaverse/*) have analogous directory structures.

NeRF embeddings

Computed by running a forward pass through the trained encoder.

Main dataset structure:

.
└── emb/
    └── model/
        └── shapenet/
            ├── hash/
            │   ├── test/
            │   │   └── *.h5
            │   ├── train/
            │   │   └── *.h5
            │   └── val/
            │       └── *.h5
            ├── mlp/
            │   ├── test/
            │   │   └── *.h5
            │   ├── train/
            │   │   └── *.h5
            │   └── val/
            │       └── *.h5
            └── triplane/
                ├── test/
                │   └── *.h5
                ├── train/
                │   └── *.h5
                └── val/
                    └── *.h5

where models are

• l_rec, aka ${\mathcal{L}}_{\text{R}}\backslash mathcal\{L\}\_\backslash text\{R\}$
• l_rec_con, aka ${\mathcal{L}}_{\text{R+C}}\backslash mathcal\{L\}\_\backslash text\{R+C\}$
• l_con, aka ${\mathcal{L}}_{\text{C}}\backslash mathcal\{L\}\_\backslash text\{C\}$.

Unseen architectures (emb/model/shapenet/hash_unseen/*, emb/model/shapenet/mlp_unseen/*, and emb/model/shapenet/triplane_unseen/*) and Objaverse NeRFs (emb/model/objaverse/*) have analogous directory structures.

Code

The official code repository will be available soon. In the meantime, here are some links to the (unpolished) code used to train the NeRFs contained in the dataset:

• MLP-based NeRF
• Tri-planar NeRF
• Hash-based NeRF

If you are interested in running this code, follow this README to install the required libraries.

Cite us

If you find our work useful, please cite us:

@inproceedings{ballerini2026weight,
  title     = {Weight Space Representation Learning on Diverse {NeRF} Architectures},
  author    = {Ballerini, Francesco and Zama Ramirez, Pierluigi and Di Stefano, Luigi and Salti, Samuele},
  booktitle = {The Fourteenth International Conference on Learning Representations},
  year      = {2026}