SheepNav
精选24天前0 投票

PRISM: Perception Reasoning Interleaved for Sequential Decision Making

arXiv:2605.05407v1 Announce Type: new Abstract: Scaling LLM-based embodied agents from text-only environments to complex multimodal settings remains a major challenge. Recent work identifies a perception-reasoning-decision gap in standalone Vision-Language Models (VLMs), which often overlook task-critical information. In this paper, we introduce PRISM, a framework that tightly couples perception (VLM) and decision (LLM) through a dynamic question-answer (DQA) pipeline. Instead of passively accep

延伸阅读

  1. 结构诱导信息助力重根列文树搜索:隐式子任务分解的新范式
  2. EHRBench:基于电子健康记录的大模型临床决策能力自动化评估基准
  3. 不确定性感知与时间调控的专家建议:让自动驾驶强化学习更安全
查看原文