We lifted 16 functions out of this paper's own repositories and ran 13 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.
| Repository | Role | Ran |
|---|---|---|
| qihao067/DiMR | canonical | 13 of 16 |
| Function | Status | Where it lives |
|---|---|---|
| Normalize | Ran | qihao067/DiMR/libs/autoencoder.py code served (permissive licence) · get_code("9fcdaa6e423e8aa7") |
| center_crop | Ran | qihao067/DiMR/datasets.py code served (permissive licence) · get_code("f057bf3471d77c43") |
| center_crop_arr | Ran | qihao067/DiMR/datasets.py code served (permissive licence) · get_code("f8b4a29a52612a41") |
| centers | Ran | qihao067/DiMR/libs/core/axial_rope.py code served (permissive licence) · get_code("d29a52dd74b2322a") |
| dct2str | Ran | qihao067/DiMR/utils.py code served (permissive licence) · get_code("507ea4bd829ac4b0") |
| get_optimizer | Ran | qihao067/DiMR/utils.py code served (permissive licence) · get_code("7328f4d3c151e58b") |
| get_sde | Ran | qihao067/DiMR/sde.py code served (permissive licence) · get_code("d35a53bdb646eb7c") |
| interpolate_fn | Ran | qihao067/DiMR/dpm_solver.py code served (permissive licence) · get_code("86c7877e164664bc") |
| model_wrapper | Ran | qihao067/DiMR/dpm_solver.py code served (permissive licence) · get_code("ace7da590f4070e0") |
| mos | Ran | qihao067/DiMR/sde.py code served (permissive licence) · get_code("02c82bdf3494ea71") |
| nonlinearity | Ran | qihao067/DiMR/libs/autoencoder.py code served (permissive licence) · get_code("3137073275f8c21a") |
| random_crop_arr | Ran | qihao067/DiMR/datasets.py code served (permissive licence) · get_code("05d1f95a391c0ec3") |
| rotate_half | Ran | qihao067/DiMR/libs/core/axial_rope.py code served (permissive licence) · get_code("a098b5566baeb748") |
| get_nnet | Not yet run | qihao067/DiMR/utils.py code served (permissive licence) · get_code("673862e6d1280a4b") |
| make_attn | Not yet run | qihao067/DiMR/libs/autoencoder.py code served (permissive licence) · get_code("f97adc1ae5efb785") |
| stp | Not yet run | qihao067/DiMR/sde.py code served (permissive licence) · get_code("4d33c2f525f1b1c0") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
This paper presents innovative enhancements to diffusion models by integrating a novel multi-resolution network and time-dependent layer normalization. Diffusion models have gained prominence for their effectiveness in high-fidelity image generation. While conventional approaches rely on convolutional U-Net architectures, recent Transformer-based designs have demonstrated superior performance and scalability. However, Transformer architectures, which tokenize input data (via "patchification"), face a trade-off between visual fidelity and computational complexity due to the quadratic nature of self-attention operations concerning token length. While larger patch sizes enable attention computation efficiency, they struggle to capture fine-grained visual details, leading to image distortions. To address this challenge, we propose augmenting the Diffusion model with the Multi-Resolution network (DiMR), a framework that refines features across multiple resolutions, progressively enhancing detail from low to high resolution. Additionally, we introduce Time-Dependent Layer Normalization (TD-LN), a parameter-efficient approach that incorporates time-dependent parameters into layer normalization to inject time information and achieve superior performance. Our method's efficacy is demonstrated on the class-conditional ImageNet generation benchmark, where DiMR-XL variants outperform prior diffusion models, setting new state-of-the-art FID scores of 1.70 on ImageNet 256 x 256 and 2.89 on ImageNet 512 x 512. Project page: https://qihao067.github.io/projects/DiMR
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2406.09416")
get_code_for_paper("2406.09416")
have("2406.09416")
Connect an agent — have() is free.