Trust Region On-Policy Distillation
Paper • 2606.01249 • Published • 42
Jeongjae Park
jjp97
AI & ML interests
I’m interested in the latest NLP and AI technologies, such as uncertainty, retrieval, agentic approaches, and long-context models!
Recent Activity
upvoted a paper 5 days ago
Trust Region On-Policy Distillation liked a model 20 days ago
zai-org/GLM-4.7-Flash upvoted a paper 21 days ago
Post-Trained MoE Can Skip Half Experts via Self-DistillationOrganizations
upvoted a paper 5 days ago
upvoted a paper 21 days ago
upvoted a paper 29 days ago
upvoted 3
papers about 2 months ago
upvoted 3
papers 2 months ago
upvoted 6
papers 3 months ago
Efficient Reasoning with Balanced Thinking
Paper • 2603.12372 • Published • 151
Attention Residuals
Paper • 2603.15031 • Published • 187
Grounding World Simulation Models in a Real-World Metropolis
Paper • 2603.15583 • Published • 154
Thinking to Recall: How Reasoning Unlocks Parametric Knowledge in LLMs
Paper • 2603.09906 • Published • 76
IndexCache: Accelerating Sparse Attention via Cross-Layer Index Reuse
Paper • 2603.12201 • Published • 54
Does Your Reasoning Model Implicitly Know When to Stop Thinking?
Paper • 2602.08354 • Published • 266
upvoted 5
papers 4 months ago
FASA: Frequency-aware Sparse Attention
Paper • 2602.03152 • Published • 154
Weak-Driven Learning: How Weak Agents make Strong Agents Stronger
Paper • 2602.08222 • Published • 290
Shaping capabilities with token-level data filtering
Paper • 2601.21571 • Published • 29
A-RAG: Scaling Agentic Retrieval-Augmented Generation via Hierarchical Retrieval Interfaces
Paper • 2602.03442 • Published • 21
PaperBanana: Automating Academic Illustration for AI Scientists
Paper • 2601.23265 • Published • 228