Downloads · 30 days
1
5% of all-time downloads
rb-e5-residual-adapter
NoamBenzimra/rb-e5-residual-adapter
rb-e5-residual-adapter is a machine learning model from NoamBenzimra. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for peft.
A representative LoRA adapter from Cross-Modal Residual Boosting (RB), a text-tabular learning method (NLP Course 097215 final project, Technion).
All-time downloads
22
Public
Repo size
224 KB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors224 KB · 98%
At a glance
- Library
- peft
- Access
- Public
- Created
- Aug 26, 2026
- Updated
- Aug 26, 2026
- SHA
- 2f5010f3
Base models
Model card
From the Hugging Face model README
RB Residual Encoder — E5-small-v2 LoRA adapter
A representative LoRA adapter from Cross-Modal Residual Boosting (RB), a text-tabular learning method (NLP Course 097215 final project, Technion).
The adapter is a residual encoder: E5-small-v2 fine-tuned (LoRA, rank 16, layers 9–11) to predict the residuals of a tabular CatBoost baseline rather than the target directly, so the text representation is complementary by construction to the tabular features.
This adapter was trained on the MulTaBench MUL_TEXT_PRODUCT_SENTIMENT dataset,
fold 0. Usage (inference, no training) is documented in the project repository.
- Library
- peft
- Created
- Aug 26, 2026
- Updated
- Aug 26, 2026