作者位次 #5/9ACL 2026Poster· Natural Language Processing
One Cognitive Loop Is Enough: SODA unlocks Pure-Text Spatial Reasoning in Large Language Models
Shunwen Bai · Jiahuan Zhang · Haoran Huang · Yurun Wang · Jiale Liu · Yanxi Wu · Ningzhe Yu · Yudong Gao · Mingjun Cheng
摘要
Currently, large language models (LLMs) have significant limitations in spatial reasoning, particularly in the absence of visual input. To address this issue, we introduce SODA (Spatial OODA), which draws inspiration from the OODA cognitive loop (Observe, Orient, Decide, Act), originally designed...