LoGRA: Scaling LLM Reinforcement Learning with Low-Rank Gradient Sketches
Paper • 2610.06647 • Published • 92
Caleb JONES
calebjonesler
·
AI & ML interests
Learning ML every day. Love open source and cool models.
Recent Activity
updated a model about 19 hours ago
calebjonesler/visual-question-answering-warmup published a model about 19 hours ago
calebjonesler/visual-question-answering-warmup upvoted a paper about 19 hours ago
LoGRA: Scaling LLM Reinforcement Learning with Low-Rank Gradient SketchesOrganizations
None yet
upvoted 2
papers about 19 hours ago
upvoted 3
papers 8 days ago
The Teacher Is a Direction, Not a Destination: Extrapolating RL-Induced Representation Residuals in On-Policy Distillation
Paper • 2609.36484 • Published • 511
Transformers Stop Thinking Too Early, and a Tiny LoRA Fixes It
Paper • 2609.36585 • Published • 70
False Frontiers: Diagnosing and Mitigating Co-Cheating in Self-Evolving Search Agents
Paper • 2609.39102 • Published • 506