Hugging Face's logo

ruotian
/
ConClick-8B

Image-Text-to-Text
PEFT
Safetensors
gui-grounding
computer-use
qwen3-vl
lora
conclick
Model card Files
xet
 Community
1
ruotian commited on
Commit
1620dc5
·
verified ·
1 Parent(s): 0c502b4

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -2
README.md CHANGED
@@ -16,8 +16,7 @@ tags:
16
 
17
  # ConClick-8B
18
 
19
- Model for **ConClick: Element-wise Contrastive Learning Improves GUI Grounding**. ConClick trains on target–distractor examples with coordinate supervision, listwise ranking, and pairwise margin losses. ReClick revisits promising regions and selects a click using equally weighted cross-view support and spatial agreement.
20
-
21
  | Inference | ScreenSpot-Pro | MMBench-GUI L2 | OSWorld-G |
22
  |---|---:|---:|---:|
23
  | Direct | 66.03 | 86.28 | 70.20 |
 
16
 
17
  # ConClick-8B
18
 
19
+ Model for **ConClick: Element-wise Contrastive Learning Improves GUI Grounding**. ConClick trains on target–distractor examples with coordinate supervision, listwise ranking, and pairwise margin losses.
 
20
  | Inference | ScreenSpot-Pro | MMBench-GUI L2 | OSWorld-G |
21
  |---|---:|---:|---:|
22
  | Direct | 66.03 | 86.28 | 70.20 |