Xiangyu Zhang, Jian Sun, Liangyu Chen
23 of 30 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (8 of them constructed an object rather than computing a result; 3 honoured the contract we drafted, 0 violated it and 11 have no contract verdict; 9 failures of our instrument; 7 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.
"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2204.04676.json · badge: /api/ran/2204.04676.svg
Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.
| Repository | Role | Ran |
|---|---|---|
| murufeng/FUIR | community (classifier label) classifier basis not recorded on the link |
11 ran · 1 no recorded run |
| lime-j/nafnet-jax | community Papers with Code archive |
4 ran · 2 no recorded run |
| rflepp/efficient_mobile_denoising_models | community Papers with Code archive |
3 ran · 1 no recorded run |
| dhryougit/afm | community Papers with Code archive |
2 ran · 0 no recorded run |
| dhryougit/learning-to-translate-noise | community Papers with Code archive |
2 ran · 0 no recorded run |
| megvii-research/NAFNet | marked official in Papers with Code Papers with Code archive |
1 ran · 0 no recorded run |
| dslisleedh/NAFNet-tensorflow2 | community Papers with Code archive |
0 ran · 1 no recorded run |
| dslisleedh/NAFNet-flax | community Papers with Code archive |
0 ran · 1 no recorded run |
| Thehunk1206/Image-Restorers | community Papers with Code archive |
0 ran · 1 no recorded run |
| Function | Status | Where it lives |
|---|---|---|
| Baseline | Ran ran on a synthesized input · constructed an object; no result was computed |
dhryougit/afm/basicsr/models/archs/Baseline_arch.py pointer only (licence: NOASSERTION) · get_code("6ce5b700c75ef796") How it ranThe body of this sample is withheld: GitHub could not classify the licence of this paper's repository, dhryougit/afm (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded. |
| BaselineBlock | Ran ran on a synthesized input |
dhryougit/afm/basicsr/models/archs/Baseline_arch.py pointer only (licence: NOASSERTION) · get_code("1a0ca651efcee9e4") How it ranThe body of this sample is withheld: GitHub could not classify the licence of this paper's repository, dhryougit/afm (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded. |
| DYNUnet | Ran ran on a synthesized input |
rflepp/efficient_mobile_denoising_models/models/Dynamic_PlainNet.py pointer only (licence: NOASSERTION) · get_code("6b743630418d829e") How it ranThe body of this sample is withheld: GitHub could not classify the licence of this paper's repository, rflepp/efficient_mobile_denoising_models (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded. |
| NAFBlock | Ran ran on a synthesized input · constructed an object; no result was computed |
lime-j/nafnet-jax/nafnet/models.py pointer only (licence: NONE) · get_code("ea0be350b35c3e6e") How it ranThe body of this sample is withheld: GitHub detected no licence in this paper's repository, lime-j/nafnet-jax when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded. |
| NAFBlock | Ran ran on a synthesized input · constructed an object; no result was computed |
dhryougit/learning-to-translate-noise/models/modules/NAFNet_arch.py pointer only (licence: NONE) · get_code("aa1fe0a698ddd646") How it ranThe body of this sample is withheld: GitHub detected no licence in this paper's repository, dhryougit/learning-to-translate-noise when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded. |
| NAFNet | Ran ran on a synthesized input · constructed an object; no result was computed |
lime-j/nafnet-jax/nafnet/models.py pointer only (licence: NONE) · get_code("c1fd5bcc1ef3945c") How it ranThe body of this sample is withheld: GitHub detected no licence in this paper's repository, lime-j/nafnet-jax when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded. |
| NAFNet | Ran ran on a synthesized input · constructed an object; no result was computed |
dhryougit/learning-to-translate-noise/models/modules/NAFNet_arch.py pointer only (licence: NONE) · get_code("2dd382ce15e8576d") How it ranThe body of this sample is withheld: GitHub detected no licence in this paper's repository, dhryougit/learning-to-translate-noise when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded. |
| NAFNet | Ran ran on a synthesized input |
megvii-research/NAFNet/basicsr/models/archs/NAFNet_arch.py pointer only (licence: NOASSERTION) · get_code("b1bd781b7fd93162") How it ranThe body of this sample is withheld: GitHub could not classify the licence of this paper's repository, megvii-research/NAFNet (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded. |
| PixelShuffle | Ran ran on a synthesized input · constructed an object; no result was computed |
lime-j/nafnet-jax/nafnet/models.py pointer only (licence: NONE) · get_code("2650084eb63ee78b") How it ranThe body of this sample is withheld: GitHub detected no licence in this paper's repository, lime-j/nafnet-jax when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded. |
| PixelShuffle | Ran ran on a synthesized input · constructed an object; no result was computed |
rflepp/efficient_mobile_denoising_models/models/Dynamic_PlainNet.py pointer only (licence: NOASSERTION) · get_code("a6c8a89b84e34837") How it ranThe body of this sample is withheld: GitHub could not classify the licence of this paper's repository, rflepp/efficient_mobile_denoising_models (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded. |
| SimpleAttention | Ran ran on a synthesized input · constructed an object; no result was computed |
lime-j/nafnet-jax/nafnet/models.py pointer only (licence: NONE) · get_code("9817ec8782366e50") How it ranThe body of this sample is withheld: GitHub detected no licence in this paper's repository, lime-j/nafnet-jax when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded. |
| calculate_fid | Ran ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT) |
murufeng/FUIR/basicsr/metrics/fid.py code served (permissive licence) · get_code("1d7ce6d6dd7ff5c1") How it ranRan in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned float64, ndim 0, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.
|
| compute_feature | Ran ran, honoured the contract we drafted |
murufeng/FUIR/basicsr/metrics/niqe.py code served (permissive licence) · get_code("5170ff1c4106b27a") How it ranRan in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.
|
| conv | Ran ran; our contract draft was wrong, not the code |
murufeng/FUIR/basicsr/models/archs/MIRNet_arch.py code served (permissive licence) · get_code("013ca1b768620edd") How it ranRan in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Sonnet 4.5, prompt python_fixture_v1) and returned Conv2d. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.
|
| conv | Ran ran; our contract draft was wrong, not the code |
murufeng/FUIR/basicsr/models/archs/HINet_arch.py code served (permissive licence) · get_code("ad4c565716dab068") How it ranRan in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned Conv2d. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.
|
| conv3x3 | Ran ran; our contract draft was wrong, not the code |
murufeng/FUIR/basicsr/models/archs/HINet_arch.py pointer only (licence: Apache-2.0); the stored copy get_code serves, from dasongli1/learning_degradation, is pointer only (licence: NONE) · get_code("822614ced2dcad7a") How it ranThe body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from dasongli1/learning_degradation, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded. |
| conv_down | Ran ran; our contract draft was wrong, not the code |
murufeng/FUIR/basicsr/models/archs/HINet_arch.py pointer only (licence: Apache-2.0); the stored copy get_code serves, from alexhe101/FourierISP, is pointer only (licence: NONE) · get_code("9c428010a12468a3") How it ranThe body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from alexhe101/FourierISP, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded. |
| estimate_aggd_param | Ran ran, honoured the contract we drafted |
murufeng/FUIR/basicsr/metrics/niqe.py code served (permissive licence) · get_code("a1282a48b0941f5d") How it ranRan in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.
|
| extract_inception_features | Ran ran, honoured the contract we drafted |
murufeng/FUIR/basicsr/metrics/fid.py code served (permissive licence) · get_code("c9fb213c957dc82d") How it ranRan in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 2, dtype family float, shape (6, 128) as a later run recorded it. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.
|
| get_position_from_periods | Ran ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT) |
murufeng/FUIR/basicsr/models/lr_scheduler.py pointer only (licence: Apache-2.0); the stored copy get_code serves, from AlifAshrafee/Restoring-Without-Forgetting, is pointer only (licence: NONE) · get_code("cd569444547de84f") How it ranThe body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from AlifAshrafee/Restoring-Without-Forgetting, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded. |
| prepare_for_ssim | Ran ran; our contract draft was wrong, not the code |
murufeng/FUIR/basicsr/metrics/psnr_ssim.py pointer only (licence: Apache-2.0); the stored copy get_code serves, from AlifAshrafee/Restoring-Without-Forgetting, is pointer only (licence: NONE) · get_code("95c6eba9350a455a") How it ranThe body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from AlifAshrafee/Restoring-Without-Forgetting, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded. |
| reorder_image | Ran ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT) |
murufeng/FUIR/basicsr/metrics/metric_util.py pointer only (licence: Apache-2.0); the stored copy get_code serves, from AlifAshrafee/Restoring-Without-Forgetting, is pointer only (licence: NONE) · get_code("95067518dc16b3e5") How it ranThe body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from AlifAshrafee/Restoring-Without-Forgetting, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded. |
| upconv_concat | Ran ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT) |
rflepp/efficient_mobile_denoising_models/models/Dynamic_PlainNet.py pointer only (licence: NOASSERTION) · get_code("f73610cd536dfe73") How it ranThe body of this sample is withheld: GitHub could not classify the licence of this paper's repository, rflepp/efficient_mobile_denoising_models (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded. |
| GlobalAvgPool2D | Not yet run unverified (harvested, no recorded run) |
lime-j/nafnet-jax/nafnet/models.py pointer only (licence: NONE) · get_code("877c91a4cf4c1ce4") How it ranThe body of this sample is withheld: GitHub detected no licence in this paper's repository, lime-j/nafnet-jax when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded. |
| NAFNet | Not yet run unverified (harvested, no recorded run) |
dslisleedh/NAFNet-tensorflow2/model.py pointer only (licence: NONE) · get_code("71e9dde3df3073c4") How it ranThe body of this sample is withheld: GitHub detected no licence in this paper's repository, dslisleedh/NAFNet-tensorflow2 when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded. |
| NAFNet | Not yet run unverified (harvested, no recorded run) |
dslisleedh/NAFNet-flax/nafnet.py pointer only (licence: NONE) · get_code("11a0e1dab1191c4e") How it ranThe body of this sample is withheld: GitHub detected no licence in this paper's repository, dslisleedh/NAFNet-flax when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded. |
| NAFnet | Not yet run unverified (harvested, no recorded run) |
Thehunk1206/Image-Restorers/models/archs/naf_net.py code served (permissive licence) · get_code("913b39408baa4210") How it ranno invocation recorded. |
| SimpleGate | Not yet run unverified (harvested, no recorded run) |
lime-j/nafnet-jax/nafnet/models.py pointer only (licence: NONE) · get_code("95df3ed7752edb6b") How it ranThe body of this sample is withheld: GitHub detected no licence in this paper's repository, lime-j/nafnet-jax when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded. |
| double_conv | Not yet run unverified (harvested, no recorded run) |
rflepp/efficient_mobile_denoising_models/models/Dynamic_PlainNet.py pointer only (licence: NOASSERTION) · get_code("9d1741e20a9e38f5") How it ranThe body of this sample is withheld: GitHub could not classify the licence of this paper's repository, rflepp/efficient_mobile_denoising_models (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded. |
| niqe | Not yet run unverified (harvested, no recorded run) |
murufeng/FUIR/basicsr/metrics/niqe.py code served (permissive licence) · get_code("0029cb1566e47350") How it ranno invocation recorded. |
Of 30 samples lifted from this paper's linked repositories, 30 carry a function identity and 0 do not. Of the 30 distinct functions with an identity, 12 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 3 the same month, 12 later) and 18 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).
Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.
| Function | Also present in | Earliest-dated | Others named |
|---|---|---|---|
| get_position_from_periods | 55 other papers 1 dated earlier, 53 later, 1 same year, order not determinable |
2110.04627 (2021-10) Vector-quantized Image Modeling with Improved VQGAN |
136790177 (2022 (year only)) Frequency and Spatial Dual Guidance for Image Dehazing 2205.05996 (2022-05) Blueprint Separable Residual Network for Efficient Image Super-Resolution and 52 more not named here |
| conv | 22 other papers 2 dated earlier, 1 the same month, 17 later, 2 same year, order not determinable · 7 only as an assembly brick |
2102.02808 (2021-02) Multi-Stage Progressive Image Restoration |
2106.03106 (2021-06) Uformer: A General U-Shaped Transformer for Image Restoration 136790177 (2022 (year only)) Frequency and Spatial Dual Guidance for Image Dehazing and 19 more not named here |
| reorder_image | 17 other papers 17 later |
Kong_Efficient_Frequency_Domain-Based_Transformers_for_High-Quality_Image_Deblurring_CVPR_2023_paper (2023 (year only)) Efficient Frequency Domain-based Transformers for High-Quality Image Deblurring |
2303.06373 (2023-03) Recursive Generalization Transformer for Image Super-Resolution 2308.03364 (2023-08) Dual Aggregation Transformer for Image Super-Resolution and 14 more not named here |
| calculate_fid | 8 other papers 8 later |
2303.02881 (2023-03) KBNet: Kernel Basis Network for Image Restoration |
2308.14036 (2023-08) MB-TaylorFormer: Multi-branch Efficient Transformer Expanded by Taylor Formula for Image Dehazing 2401.00027 (2024-01) Efficient Multi-scale Network with Learnable Discrete Wavelet Transform for Blind Motion Deblurring and 5 more not named here |
| conv | 8 other papers 1 dated earlier, 2 the same month, 5 later · 1 only as an assembly brick |
2111.07910 (2021-11) Mask-guided Spectral-wise Transformer for Efficient Hyperspectral Image Reconstruction |
2204.02844 (2022-04) Learning to Generate Realistic Noisy Images via Pixel-level Noise-aware Adversarial Training 2204.07908 (2022-04) MST++: Multi-stage Spectral-wise Transformer for Efficient Spectral Reconstruction and 5 more not named here |
| estimate_aggd_param | 8 other papers 8 later |
Chao_Equivalent_Transformation_and_Dual_Stream_Network_Construction_for_Mobile_Image_CVPR_2023_paper (2023 (year only)) Equivalent Transformation and Dual Stream Network Construction for Mobile Image Super-Resolution |
Kong_Efficient_Frequency_Domain-Based_Transformers_for_High-Quality_Image_Deblurring_CVPR_2023_paper (2023 (year only)) Efficient Frequency Domain-based Transformers for High-Quality Image Deblurring 2308.14036 (2023-08) MB-TaylorFormer: Multi-branch Efficient Transformer Expanded by Taylor Formula for Image Dehazing and 5 more not named here |
| extract_inception_features | 8 other papers 8 later |
2303.02881 (2023-03) KBNet: Kernel Basis Network for Image Restoration |
2308.14036 (2023-08) MB-TaylorFormer: Multi-branch Efficient Transformer Expanded by Taylor Formula for Image Dehazing 2401.00027 (2024-01) Efficient Multi-scale Network with Learnable Discrete Wavelet Transform for Blind Motion Deblurring and 5 more not named here |
| prepare_for_ssim | 8 other papers 7 later, 1 same year, order not determinable |
ijcai2022_0131 (2022 (year only)) Exploring Fourier Prior for Single Image Rain Removal |
Kong_Efficient_Frequency_Domain-Based_Transformers_for_High-Quality_Image_Deblurring_CVPR_2023_paper (2023 (year only)) Efficient Frequency Domain-based Transformers for High-Quality Image Deblurring 2401.15235 (2024-01) CascadedGaze: Efficiency in Global Context Extraction for Image Restoration and 5 more not named here |
| compute_feature | 7 other papers 7 later |
Kong_Efficient_Frequency_Domain-Based_Transformers_for_High-Quality_Image_Deblurring_CVPR_2023_paper (2023 (year only)) Efficient Frequency Domain-based Transformers for High-Quality Image Deblurring |
2308.14036 (2023-08) MB-TaylorFormer: Multi-branch Efficient Transformer Expanded by Taylor Formula for Image Dehazing 2401.00027 (2024-01) Efficient Multi-scale Network with Learnable Discrete Wavelet Transform for Blind Motion Deblurring and 4 more not named here |
| niqe | 7 other papers 7 later |
Kong_Efficient_Frequency_Domain-Based_Transformers_for_High-Quality_Image_Deblurring_CVPR_2023_paper (2023 (year only)) Efficient Frequency Domain-based Transformers for High-Quality Image Deblurring |
2308.14036 (2023-08) MB-TaylorFormer: Multi-branch Efficient Transformer Expanded by Taylor Formula for Image Dehazing 2401.00027 (2024-01) Efficient Multi-scale Network with Learnable Discrete Wavelet Transform for Blind Motion Deblurring and 4 more not named here |
| conv_down | 4 other papers 1 the same month, 2 later, 1 same year, order not determinable · 2 only as an assembly brick |
ijcai2022_0131 (2022 (year only)) Exploring Fourier Prior for Single Image Rain Removal |
2204.13348 (2022-04) Deep Generalized Unfolding Networks for Image Restoration 2208.05244 (2022-08) Learning Degradation Representations for Image Deblurring and 1 more not named here |
| conv3x3 | 2 other papers 1 later, 1 same year, order not determinable |
ijcai2022_0131 (2022 (year only)) Exploring Fourier Prior for Single Image Rain Removal |
2208.05244 (2022-08) Learning Degradation Representations for Image Deblurring |
Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
Although there have been significant advances in the field of image restoration recently, the system complexity of the state-of-the-art (SOTA) methods is increasing as well, which may hinder the convenient analysis and comparison of methods. In this paper, we propose a simple baseline that exceeds the SOTA methods and is computationally efficient. To further simplify the baseline, we reveal that the nonlinear activation functions, e.g. Sigmoid, ReLU, GELU, Softmax, etc. are not necessary: they could be replaced by multiplication or removed. Thus, we derive a Nonlinear Activation Free Network, namely NAFNet, from the baseline. SOTA results are achieved on various challenging benchmarks, e.g. 33.69 dB PSNR on GoPro (for image deblurring), exceeding the previous SOTA 0.38 dB with only 8.4% of its computational costs; 40.30 dB PSNR on SIDD (for image denoising), exceeding the previous SOTA 0.28 dB with less than half of its computational costs. The code and the pre-trained models are released at github.com/megvii-research/NAFNet.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2204.04676")
get_code_for_paper("2204.04676")
have("2204.04676")
The run record, dated, one paper per request, free:
curl https://syntology.ai/api/ran/2204.04676.json
A badge for a README (the split and the date, never a ratio):
[](https://syntology.ai/paper/2204.04676)
Connect an agent — have() is free.