Owner
ExplainableML
4 indexed repositories · View on GitHub
-
sae-for-vlm
[NeurIPS 2025] Sparse Autoencoders Learn Monosemantic Features in Vision-Language Models
Python · 96 stars
-
EgoCVR
[ECCV 2024] EgoCVR: An Egocentric Benchmark for Fine-Grained Composed Video Retrieval
Python · 41 stars
-
SOTAlign
[ICML 2026] SOTAlign: Semi-Supervised Alignment of Unimodal Vision and Language Models via Optimal Transport
Python · 11 stars
-
SAE-TM
[ICML 2026] Sparse Autoencoders are Topic Models
Jupyter Notebook · 10 stars
-
Difficulty 4/5 3-5 days Newbie friendliness 35/100
-
ExplainableML/SAE-TM#2 · 5 comments · 1 assignee ·
-
ExplainableML/SAE-TM#1 · 1 comment · 1 assignee ·
-
Pre-trained SAEs Open
Difficulty 5/5 Over a week Newbie friendliness 35/100
ExplainableML/sae-for-vlm#5 · 8 comments ·
-
Difficulty 5/5 Over a week Newbie friendliness 20/100
ExplainableML/EgoCVR#6 ·
-
Difficulty 4/5 3-5 days Newbie friendliness 25/100
ExplainableML/sae-for-vlm#1 · 1 comment · 1 reaction ·
-
Difficulty 2/5 1-3 hours Newbie friendliness 35/100
ExplainableML/EgoCVR#3 ·
-
Difficulty 4/5 3-5 days Newbie friendliness 20/100
ExplainableML/EgoCVR#2 · 2 comments ·