Downloads · 30 days
6
9% of all-time downloads
qwen4b-reward-pairwise-v1
flavianv/qwen4b-reward-pairwise-v1
qwen4b-reward-pairwise-v1 is a text classification model from flavianv. Use it when you need a label for a piece of text. The card lists the license as apache-2.0.
Reward V1: a Qwen3-4B AutoModelForSequenceClassification (scalar value head) warm-started from V0 and trained with margin Bradley-Terry loss on gold-vs-corrupted preference pairs — to discriminate good outfits, fixing…
All-time downloads
68
Public
Parameters
4B
8.1 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors8 GB · 100%
At a glance
- Task
- Text Classification
- License
- apache-2.0
- Model type
- qwen3
- Access
- Public
- Created
- Jun 30, 2026
- Updated
- Jun 30, 2026
- SHA
- 90404eee
Try a prompt
Base models
Model card
From the Hugging Face model README
DeepShopper Reward V1 — pairwise Bradley-Terry outfit reward (Qwen3-4B)
Reward V1: a Qwen3-4B AutoModelForSequenceClassification (scalar value head)
warm-started from V0
and trained with margin Bradley-Terry loss on gold-vs-corrupted preference pairs —
to discriminate good outfits, fixing V0's gender/structure blindness.
- Data:
flavianv/deepshopper-reward-pairs— 93k train pairs, 5 corruption types (gender_flip, item_swap, duplicate_role, count_drop, cross_need); larger margin for gender_flip. - Held-out ranking accuracy: overall 0.995, gender_flip 0.999 (V0 was flat on gender), duplicate_role 1.0, count_drop 0.998, cross_need 0.993, item_swap 0.987.
- Use: score an outfit (system bundle prompt + need +
### Item i: role\ntitle) → scalar reward (higher = better). - Note: "good" = resembling observed/sold bundles (purchase-likeness), the intended objective; still a learned proxy for real sales.
Code: https://github.com/clijo/reco-rl (branch outfit_bundle).
- Task
- Text Classification
- Type
- qwen3
- License
- apache-2.0
- Created
- Jun 30, 2026
- Updated
- Jun 30, 2026