A vision benchmark for testing whether models can infer hierarchical containment trees from images of nested, non-intersecting curves.
Amir Mohseni PRO
AmirMohseni
AI & ML interests
LLMs, VLMs, VLAs
Recent Activity
updated a dataset 3 days ago
AmirMohseni/WildChat-Legal-Questions updated a dataset 3 days ago
AmirMohseni/WildChat-Legal-Classification-V2-LegalOnly updated a dataset 3 days ago
AmirMohseni/WildChat-Legal-Classification-V2-BalancedOrganizations
-
Legal Conversation Explorer
⚖3Explore legal conversations on an interactive topic map
-
Legal Classifier
⚖Route WildChat conversations with ModernBERT encoders
-
AmirMohseni/WildChat-Legal-Classification-V2-Balanced
Viewer • Updated • 4.24k • 279 -
AmirMohseni/WildChat-Legal-Classification-V2-LegalOnly
Viewer • Updated • 2.12k • 193
Route between “thinking” and “no-thinking” modes for hybrid models like Qwen3. Blog: https://huggingface.co/blog/AmirMohseni/reasoning-router
A vision benchmark for testing whether models can infer hierarchical containment trees from images of nested, non-intersecting curves.
-
Legal Conversation Explorer
⚖3Explore legal conversations on an interactive topic map
-
Legal Classifier
⚖Route WildChat conversations with ModernBERT encoders
-
AmirMohseni/WildChat-Legal-Classification-V2-Balanced
Viewer • Updated • 4.24k • 279 -
AmirMohseni/WildChat-Legal-Classification-V2-LegalOnly
Viewer • Updated • 2.12k • 193
https://github.com/Amir-Mohseni/Whisper-Persian
Route between “thinking” and “no-thinking” modes for hybrid models like Qwen3. Blog: https://huggingface.co/blog/AmirMohseni/reasoning-router