Urbex: Agentic Spatial Grounding for City-Scale 3D Scenes
Paper • 2610.18356 • Published • 121
Rania Bouazizi
rbouazizi
AI & ML interests
reward hacking, speech translation, curriculum learning
Recent Activity
upvoted a paper about 16 hours ago
Urbex: Agentic Spatial Grounding for City-Scale 3D Scenes upvoted a paper about 16 hours ago
Scoring Higher, Answering Worse: Mitigating Reward Hacking in Rubric-Based RL via Protocol-Level Rubrics upvoted a paper about 16 hours ago
Vela: User Behavior as Supervision Sparks LLM IntelligenceOrganizations
None yet
upvoted 5
papers about 16 hours ago
Scoring Higher, Answering Worse: Mitigating Reward Hacking in Rubric-Based RL via Protocol-Level Rubrics
Paper • 2609.38847 • Published • 67
Vela: User Behavior as Supervision Sparks LLM Intelligence
Paper • 2610.14709 • Published • 78
X-Ragent: Tool-Augmented Reinforcement Learning for Evidence-Grounded Chest X-Ray Reasoning
Paper • 2610.16811 • Published • 72
AesCode: Aesthetic Code Generation with Decoupled Cross-Modal Rewards
Paper • 2610.18860 • Published • 122
upvoted 2
papers 1 day ago
upvoted 2
papers 2 days ago
upvoted 3
papers 3 days ago
upvoted 4
papers 5 days ago
Semifactual Credit-Augmented Policy Optimization
Paper • 2609.40360 • Published • 38
Self-Retrospection Distillation: Turning Post-hoc Experiences into Prior Foresight
Paper • 2610.08077 • Published • 194
On KL-Regularized Policy Optimization
Paper • 2610.08963 • Published • 26
VIEScore2: Unified Image Evaluation with Spatially Grounded Explanations
Paper • 2610.00994 • Published • 32
upvoted 4
papers 7 days ago
On-Policy or Off-Policy Learning? A Systematic Study of Distillation Dynamics
Paper • 2609.35259 • Published • 205
Sharpening Tax in Post-Training
Paper • 2610.01509 • Published • 111
Thinking Outside the Box: Can Language Models Rely on External Guidance Selectively?
Paper • 2609.39578 • Published • 72
TERRA: Terrain-Aware Reconstruction, Retargeting and Control for Musculoskeletal Locomotion
Paper • 2609.38653 • Published • 40