Simplifying Alignment: From RLHF to Direct Preference Optimization (DPO)
•
13
Aritra Roy Gosthipaty PRO
ariG23498
AI & ML interests
Deep Representation Learning
Recent Activity
commented on
an
article
about 21 hours ago
SmolVLM Grows Smaller – Introducing the 250M & 500M Models!
upvoted
an
article
about 21 hours ago
SmolVLM Grows Smaller – Introducing the 250M & 500M Models!
Articles
Organizations
Posts
6
Post
1800
Tried my hand at simplifying the derivations of Direct Preference Optimization.
I cover how one can reformulate RLHF into DPO. The idea of implicit reward modeling is chef's kiss.
Blog: https://huggingface.co/blog/ariG23498/rlhf-to-dpo
I cover how one can reformulate RLHF into DPO. The idea of implicit reward modeling is chef's kiss.
Blog: https://huggingface.co/blog/ariG23498/rlhf-to-dpo
Post
1767
Timm ❤️ Transformers
Wtih the latest version of transformers you can now use any timm model with the familiar transformers API.
Blog Post: https://huggingface.co/blog/timm-transformers
Repository with examples: https://github.com/ariG23498/timm-wrapper-examples
Collection: ariG23498/timmwrapper-6777b85f1e8d085d3f1374a1
Wtih the latest version of transformers you can now use any timm model with the familiar transformers API.
Blog Post: https://huggingface.co/blog/timm-transformers
Repository with examples: https://github.com/ariG23498/timm-wrapper-examples
Collection: ariG23498/timmwrapper-6777b85f1e8d085d3f1374a1
Collections
9
This is a collection of models to showcase various goodies with the `TimmWrapper`.
-
1🦀
Food Classification
Food classifier
-
4😻
Timm + Transformers
Any timm model with the transformers integration
-
ariG23498/vit_base_patch16_224.orig_in21k_ft_in1k.ft_food101
Image Classification • Updated • 212 -
ariG23498/vit_base_patch16_224.augreg2_in21k_ft_in1k.ft_food101
Image Classification • Updated • 56
Papers
1
spaces
9
models
41
ariG23498/layer-skip
Text Generation
•
Updated
•
2
ariG23498/layer-skip-vanill-3.2-1b
Text Generation
•
Updated
•
9
ariG23498/tw.resnet50.a1_in1k
Image Classification
•
Updated
•
7
ariG23498/vit_base_patch16_224.augreg2_in21k_ft_in1k.ft_food101
Image Classification
•
Updated
•
56
ariG23498/vit_base_patch16_224.augreg2_in21k_ft_in1k.lora_ft_food101
Image Classification
•
Updated
ariG23498/timm_wrapper_vit_base_patch16_224.augreg2_in21k_ft_in1k
Image Classification
•
Updated
•
22
ariG23498/vit_base_patch16_224.orig_in21k_ft_in1k.ft_food101
Image Classification
•
Updated
•
212
ariG23498/layer-skip-v1
Text Generation
•
Updated
•
31
ariG23498/layer-skip-v0
Text Generation
•
Updated
•
33
ariG23498/trained-sana-lora
Text-to-Image
•
Updated
•
12
datasets
15
ariG23498/quick-notebooks
Updated
•
109
•
12
ariG23498/intersection-dataset
Viewer
•
Updated
•
3.3k
•
52
ariG23498/layer-skip-assets
Viewer
•
Updated
•
18
•
40
•
1
ariG23498/sukumar-ray-painting
Viewer
•
Updated
•
8
•
45
ariG23498/images
Viewer
•
Updated
•
28
•
358
•
1
ariG23498/colab-snippets
Viewer
•
Updated
•
1
•
40
ariG23498/community-blogs
Viewer
•
Updated
•
428
•
41
•
1
ariG23498/hi-en-dataset
Viewer
•
Updated
•
1.66M
•
62
ariG23498/samsum-tokenized-flan-t5-xxl
Viewer
•
Updated
•
16.4k
•
54
ariG23498/test-en-embed
Viewer
•
Updated
•
1
•
46