When2Think: Learning Difficulty-Aware Length Control for Efficient Hybrid Reasoning Models Paper • 2609.19671 • Published 5 days ago • 33
ProgramDistill: From Interactive Web Apps to Verifiable Reference-Guided SWE Tasks Paper • 2609.18805 • Published 6 days ago • 61
Eliciting Weak-to-Strong Generalization with On-Policy Reverse Distillation Paper • 2609.08798 • Published 14 days ago • 81
Temporal Preference Optimization for Unsupervised Retrieval Paper • 2606.17664 • Published Jun 16 • 1
Scaling Reasoning Efficiently via Relaxed On-Policy Distillation Paper • 2603.11137 • Published Mar 11 • 1
Decoder-Hybrid-Decoder Architecture for Efficient Reasoning with Long Generation Paper • 2507.06607 • Published Jul 9, 2025 • 10
Decoder-Hybrid-Decoder Architecture for Efficient Reasoning with Long Generation Paper • 2507.06607 • Published Jul 9, 2025 • 10
SlimMoE: Structured Compression of Large MoE Models via Expert Slimming and Distillation Paper • 2506.18349 • Published Jun 23, 2025 • 13
SlimMoE: Structured Compression of Large MoE Models via Expert Slimming and Distillation Paper • 2506.18349 • Published Jun 23, 2025 • 13
Phi-4-Mini-Reasoning: Exploring the Limits of Small Reasoning Language Models in Math Paper • 2504.21233 • Published Apr 30, 2025 • 50
Phi-4-Mini-Reasoning: Exploring the Limits of Small Reasoning Language Models in Math Paper • 2504.21233 • Published Apr 30, 2025 • 50