Submitted by daxiang 155 Qianfan-OCR: A Unified End-to-End Model for Document Intelligence BAIDU 421 4
Submitted by YijuGuo 12 Less Noise, More Voice: Reinforcement Learning for Reasoning via Instruction Purification BAIDU 4
4 LoongFlow: Directed Evolutionary Search via a Cognitive Plan-Execute-Summarize Paradigm BAIDU 467 2
3 GlobalRAG: Enhancing Global Reasoning in Multi-hop Question Answering via Reinforcement Learning BAIDU 16
3 Knowledge-Level Consistency Reinforcement Learning: Dual-Fact Alignment for Long-Form Factuality BAIDU 14