Thinking with Reasoning Skills: Fewer Tokens, More Accuracy Paper β’ 2604.21764 β’ Published Apr 23 β’ 1
When Good OCR Is Not Enough: Benchmarking OCR Robustness for Retrieval-Augmented Generation Paper β’ 2605.00911 β’ Published Apr 29 β’ 1
When Good OCR Is Not Enough: Benchmarking OCR Robustness for Retrieval-Augmented Generation Paper β’ 2605.00911 β’ Published Apr 29 β’ 1
TriPlay-RL: Tri-Role Self-Play Reinforcement Learning for LLM Safety Alignment Paper β’ 2601.18292 β’ Published Jan 26 β’ 12
FABLE: Forest-Based Adaptive Bi-Path LLM-Enhanced Retrieval for Multi-Document Reasoning Paper β’ 2601.18116 β’ Published Jan 26 β’ 13
TriPlay-RL: Tri-Role Self-Play Reinforcement Learning for LLM Safety Alignment Paper β’ 2601.18292 β’ Published Jan 26 β’ 12
Router Upcycling: Leveraging Mixture-of-Routers in Mixture-of-Experts Upcycling Paper β’ 2509.00679 β’ Published Aug 31, 2025
TextlessRAG: End-to-End Visual Document RAG by Speech Without Text Paper β’ 2509.07538 β’ Published Sep 9, 2025
FABLE: Forest-Based Adaptive Bi-Path LLM-Enhanced Retrieval for Multi-Document Reasoning Paper β’ 2601.18116 β’ Published Jan 26 β’ 13
TriPlay-RL: Tri-Role Self-Play Reinforcement Learning for LLM Safety Alignment Paper β’ 2601.18292 β’ Published Jan 26 β’ 12
FABLE: Forest-Based Adaptive Bi-Path LLM-Enhanced Retrieval for Multi-Document Reasoning Paper β’ 2601.18116 β’ Published Jan 26 β’ 13
Efficient Switchable Safety Control in LLMs via Magic-Token-Guided Co-Training Paper β’ 2508.14904 β’ Published Aug 12, 2025 β’ 2