nielsr HF Staff commited on
Commit
1bc3ddf
·
verified ·
1 Parent(s): a8fc6be

Remove arXiv ID from metadata and add links to GitHub and Project Page

Browse files

This PR improves the model card by:
1. Removing the `arxiv` field from the YAML metadata (as Hugging Face guidelines recommend keeping it in the markdown section instead).
2. Adding direct links to the official GitHub repository and project page at the top of the model card for better visibility.

Files changed (1) hide show
  1. README.md +4 -3
README.md CHANGED
@@ -1,5 +1,4 @@
1
  ---
2
- arxiv: "2607.02375"
3
  license: apache-2.0
4
  pipeline_tag: text-to-image
5
  tags:
@@ -20,6 +19,8 @@ over a curated teacher reference. One forward pass at 512² (≈0.15–0.3 s/ima
20
  **This 1-step student matches or exceeds its 4-step teacher on all three eval axes** (standard-mmdet GenEval
21
  composition + PickScore human-preference proxy).
22
 
 
 
23
  ## Results (checkpoint = step 180)
24
 
25
  Standard-mmdet GenEval (553 prompts, avg over 6 tasks) + PickScore (raw `logit_scale·cos`), vs the teacher:
@@ -28,7 +29,7 @@ Standard-mmdet GenEval (553 prompts, avg over 6 tasks) + PickScore (raw `logit_s
28
  |---|---|---|---|
29
  | naive klein @ 1 step (no distillation) | ~0.42 | 19.95 | 20.11 |
30
  | 4-step FLUX.2 klein teacher | 0.7944 | 22.576 | 21.848 |
31
- | **this — 1-step RDM (s180)** | **0.8258** | **22.755** | 21.817 |
32
 
33
  Distillation lifts the 1-step model from a broken floor (PickScore-cv 19.95) to **above the 4-step teacher**
34
  on GenEval (+3.1 pp) and PickScore-COCO-val (+0.18), and to parity on Pick-a-Pic — i.e. it recovers the
@@ -101,4 +102,4 @@ Paper: [arXiv:2607.02375](https://arxiv.org/abs/2607.02375) · [Hugging Face Pap
101
  ```
102
 
103
  ## License
104
- Derived from **FLUX.2 klein** (Apache-2.0); released under Apache-2.0.
 
1
  ---
 
2
  license: apache-2.0
3
  pipeline_tag: text-to-image
4
  tags:
 
19
  **This 1-step student matches or exceeds its 4-step teacher on all three eval axes** (standard-mmdet GenEval
20
  composition + PickScore human-preference proxy).
21
 
22
+ [💻 Code](https://github.com/vita-epfl/RDM) · [📖 Paper](https://huggingface.co/papers/2607.02375) · [🌐 Project Page](https://alan-lanfeng.github.io/rdm/)
23
+
24
  ## Results (checkpoint = step 180)
25
 
26
  Standard-mmdet GenEval (553 prompts, avg over 6 tasks) + PickScore (raw `logit_scale·cos`), vs the teacher:
 
29
  |---|---|---|---|
30
  | naive klein @ 1 step (no distillation) | ~0.42 | 19.95 | 20.11 |
31
  | 4-step FLUX.2 klein teacher | 0.7944 | 22.576 | 21.848 |
32
+ | **this — 1-step RDM (s180)** | **0.8258** | **22.755** | **21.817** |
33
 
34
  Distillation lifts the 1-step model from a broken floor (PickScore-cv 19.95) to **above the 4-step teacher**
35
  on GenEval (+3.1 pp) and PickScore-COCO-val (+0.18), and to parity on Pick-a-Pic — i.e. it recovers the
 
102
  ```
103
 
104
  ## License
105
+ Derived from **FLUX.2 klein** (Apache-2.0); released under Apache-2.0.