Update README.md for reconstruction viewer

README.md

@@ -18,16 +18,16 @@ keeps the same FCDM decoder family as
 8-block encoder, an 8-block decoder, and a 384-channel spatial latent at
 `H/32 x W/32`.
 
-Relative to the original
-[SemDisDiffAE](https://huggingface.co/data-archetype/semdisdiffae), this model
-is optimized for a lower-resolution latent grid and downstream latent diffusion:
-patch size `32` instead of `16`, `384` latent channels instead of `128`, an
-8-block encoder instead of a 4-block encoder, and DINOv3 ConvNeXt-B semantic
-alignment instead of the original DINO semantic alignment setup.
+Relative to the original SemDisDiffAE, this model is optimized for a
+lower-resolution latent grid and downstream latent diffusion: patch size `32`
+instead of `16`, `384` latent channels instead of `128`, an 8-block encoder
+instead of a 4-block encoder, and DINOv3 ConvNeXt-B semantic alignment instead
+of the original DINO semantic alignment setup.
 
-See the original
-[SemDisDiffAE technical report](https://huggingface.co/data-archetype/semdisdiffae/blob/main/technical_report_semantic.md)
-for the shared architecture background.
+For details, see the
+[semdisdiffae_p32_v2 technical report](https://huggingface.co/data-archetype/semdisdiffae_p32_v2/blob/main/technical_report_fcdm_diffae.md).
+For additional shared FCDM / VP decoder background, see the original
+[SemDisDiffAE technical report](https://huggingface.co/data-archetype/semdisdiffae/blob/main/technical_report_semantic.md).
 
 The p32 checkpoint was trained at `384` resolution rather than the original
 `256`-scale recipe. With patch size `32`, this gives a `12x12` latent grid
@@ -43,6 +43,13 @@ covers. Reconstruction uses the default 1-step VP/DDIM path in `bfloat16`.
 |---|---:|---:|---:|---:|---:|
 | semdisdiffae_p32_v2 | `36.06` | `5.47` | `35.80` | `27.63` | `45.02` |
 
+## Reconstruction Viewer
+
+The 39-image reconstruction viewer shows originals, semdisdiffae_p32_v2
+reconstructions, RGB error deltas, and latent PCA side by side, with FLUX.2 VAE
+included for comparison:
+[semdisdiffae_p32_v2 reconstruction viewer](https://huggingface.co/spaces/data-archetype/semdisdiffae_p32_v2-results).
+
 ## Throughput
 
 Measured on an `NVIDIA GeForce RTX 5090` in `bfloat16`, with `5` warmup batches
@@ -56,12 +63,6 @@ disabled.
 | Encode | `512x512` | `32` | `12.08` | `2649.9` | `579 MiB` |
 | Decode | `512x512` | `32` | `100.36` | `318.8` | `1042 MiB` |
 
-## Results Viewer
-
-The 39-image visual viewer shows originals, FLUX.2 reconstructions,
-semdisdiffae_p32_v2 reconstructions, deltas, and latent PCA side by side:
-[semdisdiffae_p32_v2 results](https://huggingface.co/spaces/data-archetype/semdisdiffae_p32_v2-results).
-
 ## Latent Interface
 
 - `encode()` returns whitened latents using the model's saved running statistics.