From the 1 of 5 linked papers with an AI index.
5 papers
FinixDoc: Rethinking Financial Document Parsing Beyond Saturated Benchmarks
Hang Wang, Jin Zhang, Guoliang Xu +13
Financial document parsing requires accuracy, structural consistency, and verifiability that current benchmarks often fail to reflect. We present FinixDoc, an end-to-end agentic pa…
FAST: A Framework for Aligned Sampling and Training in Parallel Reinforcement Learning for Autonomous Driving
Bonan Wang, Letian Tao, Bin Shuai +7
The paper introduces FAST, a synchronous parallel framework that improves sampling efficiency for deep reinforcement learning in autonomous driving by aligning parallel simulations…
Neuro-Symbolic Drive: Rule-Grounded Faithful Reasoning for Driving VLAs
Xiangbo Gao, Xiukun Huang, Boyu Lu +5
Driving VLA models incorporating Chain-of-Thought (CoT) reasoning are attractive because they leverage pretrained VLM representations and expose intermediate decisions in natural l…
UCPO: Uncertainty-Aware Policy Optimization
Xianzhou Zeng, Jing Huang, Chunmei Xie +9
The key to building trustworthy large language models (LLMs) lies in endowing them with inherent uncertainty expression capabilities, thereby mitigating overconfident errors in hig…
NGRPO: Negative-enhanced Group Relative Policy Optimization
Gongrui Nan, Siye Chen, Jing Huang +8
RLVR has enhanced the reasoning capabilities of Large Language Models (LLMs) across various tasks. However, GRPO, a representative RLVR algorithm, suffers from a critical limitatio…