From Production Traffic to Post-Training: Building a Self-Hosted LLM That Covers the Corporate Request Mix
Paper • 2609.01572 • Published • 33
🔄 In a Training Loop
Urro
urroxyz
AI & ML interests
computational linguistics major 🤖🔎ðŸ”
i am autistic. if i come off rude, i probably didn't mean to. please feel free to ask me for clarification.
Recent Activity
View all activity
Organizations
upvoted 2
papers 1 day ago
upvoted 4
papers 2 days ago
Verification-Aware Training for Speculative Decoding
Paper • 2608.30135 • Published • 8
Normalized Low-Rank Adaptation
Paper • 2608.31036 • Published • 50
On the Design of Qwen3.8-Next Architecture: Evaluation, Efficiency, and Training Stability
Paper • 2608.30320 • Published • 50
Does On-Policy Distillation Really Distill? From Noisy Teacher to Self-Improvement
Paper • 2608.31046 • Published • 135
upvoted 2
papers 3 days ago
upvoted a paper 5 days ago
upvoted 5
papers 6 days ago
Prefix Sliding for efficient test-time scaling
Paper • 2608.26070 • Published • 5
Gated Recurrent Transformers: Expressive Depth through Recurrent Modulation
Paper • 2608.15062 • Published • 10
D^3-MOPD: Adaptive Dynamic Domain ScheDuling for Efficient Multi-Teacher Distillation
Paper • 2608.24987 • Published • 25
Understanding Evolution Strategies for LLM Reasoning: Broader Reasoning Coverage than GRPO
Paper • 2608.27351 • Published • 21
TTPO: Test-Time Policy Optimization
Paper • 2608.27448 • Published • 81
upvoted 5
papers 7 days ago
Meta^n: Recursive Self-Improvement through Emergent Depth
Paper • 2608.24735 • Published • 15
Best Practice Critic Optimization
Paper • 2608.23566 • Published • 18
On-policy Distillation with Verifiable Reward
Paper • 2608.24696 • Published • 17
Expert Upcycling: Shifting the Compute-Efficient Frontier of Mixture-of-Experts
Paper • 2604.19835 • Published • 22
Why Does Self-Distillation (Sometimes) Degrade the Reasoning Capability of LLMs?
Paper • 2603.24472 • Published • 59
upvoted a paper 9 days ago