现有的智能体框架和推理时间算法在处理复杂规划问题时往往面临挑战,原因在于验证生成的规划或推理过程存在局限,以及单个任务中实例复杂度的差异性。许多现有方法要么执行任务级验证而不考虑约束条件,要么应用推理时间算法而不适应实例级复杂度。为解决这些局限,我们提出了PlanGEN,一个与模型无关且易于扩展的智能体框架,包含三个关键组件:约束智能体、验证智能体和选择智能体。具体而言,我们的方法提出约束引导的迭代验证,以增强推理时间算法(Best of N、思维树和REBASE)的性能。在PlanGEN框架中,选择智能体根据实例复杂度优化算法选择,确保对复杂规划问题具有更好的适应性。实验结果表明,在多个基准上相比最强基线有显著提升,在NATURAL PLAN(约8%提升)、OlympiadBench(约4%提升)、DocFinQA(约7%提升)和GPQA(约1%提升)上取得了最先进的结果。我们的关键发现强调,约束引导的迭代验证改进了推理时间算法,而自适应选择进一步提升了复杂规划和推理问题的性能。
英文摘要:
Recent agent frameworks and inference-time algorithms often struggle with complex planning problems due to limitations in verifying generated plans or reasoning and varying complexity of instances within a single task. Many existing methods for these tasks either perform task-level verification without considering constraints or apply inference-time algorithms without adapting to instance-level complexity. To address these limitations, we propose PlanGEN, a model-agnostic and easily scalable agent framework with three key components: constraint, verification, and selection agents. Specifically, our approach proposes constraint-guided iterative verification to enhance performance of inference-time algorithms--Best of N, Tree-of-Thought, and REBASE. In PlanGEN framework, the selection agent optimizes algorithm choice based on instance complexity, ensuring better adaptability to complex planning problems. Experimental results demonstrate significant improvements over the strongest baseline across multiple benchmarks, achieving state-of-the-art results on NATURAL PLAN ($\sim$8%$\uparrow$), OlympiadBench ($\sim$4%$\uparrow$), DocFinQA ($\sim$7%$\uparrow$), and GPQA ($\sim$1%$\uparrow$). Our key finding highlights that constraint-guided iterative verification improves inference-time algorithms, and adaptive selection further boosts performance on complex planning and reasoning problems.
AnchorVLA: Bridging Discrete Decisions and Continuous Trajectories for Vision-Language-Action Planning
School of Vehicle and Mobility, Tsinghua University(清华大学车辆与运载学院);State Key Laboratory of Intelligent Green Vehicle and Mobility, Tsinghua University(清华大学智能绿色车辆与交通国家重点实验室);School of Computer Science and Engineering, Southeast University(东南大学计算机科学与工程学院);Meituan Inc.(美团公司);Dongfeng Motor Corporation(东风汽车公司)
Augmenting GRIPS with Heuristic Sampling for Planning Feasible Trajectories of a Car-Like Robot
Moscow Institute of Physics and Technology(莫斯科物理技术学院);LLC “Integrant”(Integrant有限责任公司);Federal Research Center “Computer Science and Control” RAS(俄罗斯科学院联邦研究中心“计算机科学与控制”)
Towards Bio-Inspired Robotic Trajectory Planning via Self-Supervised RNN
Faculty of Mathematics, Physics and Informatics, Comenius University Bratislava(数学、物理与信息学系,科门尼乌斯大学布拉迪斯拉发分校);Department of Informatics, University of Hamburg(信息学系,汉堡大学)