jqtangust/Robust-R1
🔥🔥🔥[AAAI 2026 Oral] Official Implementation of Robust-R1: Degradation-Aware Reasoning for Robust Visual Understanding
This project helps anyone working with visual data ensure their AI models can accurately understand images, even when those images are blurry, noisy, or damaged. It takes in degraded images and a question about them, then produces a more reliable answer. This is for professionals who need AI to interpret visual information accurately in real-world conditions, like quality control inspectors or autonomous system developers.
520 stars.
Use this if your image analysis or visual question-answering systems struggle with real-world image imperfections like blur, noise, or compression artifacts.
Not ideal if your primary concern is generating perfect images or if you are working with pristine, high-quality visual data exclusively.
Stars
520
Forks
18
Language
Python
License
—
Category
Last pushed
Jan 20, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/jqtangust/Robust-R1"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
cvs-health/uqlm
UQLM: Uncertainty Quantification for Language Models, is a Python package for UQ-based LLM...
PRIME-RL/TTRL
[NeurIPS 2025] TTRL: Test-Time Reinforcement Learning
sapientinc/HRM
Hierarchical Reasoning Model Official Release
tigerchen52/query_level_uncertainty
query-level uncertainty in LLMs
reasoning-survey/Awesome-Reasoning-Foundation-Models
✨✨Latest Papers and Benchmarks in Reasoning with Foundation Models