Skip to content

PhysCaP: Grounding Code-as-Policy Agent with Physics-Informed Exploration

Aug 2026 · arXiv.org · Vol abs/2608.21031 · 3 citations · 76 references
Computer Science

TL;DR

PhysCaP augments code-as-policy frameworks with a physics-informed exploration layer that enables explicit information-seeking through interaction, and introduces training-free physical property extraction modules that estimate object mass and stiffness from robot proprioception without additional sensors.

Abstract

We present PhysCaP, a Physics-Informed Code-as-Policy agent system for active perception in robotic manipulation. While vision-language-action policies excel at imitating demonstrations, they rely on passive observation and fail to infer latent physical properties critical for manipulation. PhysCaP augments code-as-policy frameworks with a physics-informed exploration layer that enables explicit information-seeking through interaction. Our method introduces training-free physical property extraction modules that estimate object mass and stiffness from robot proprioception without additional sensors. To balance exploration costs and the efficiency of information obtained, PhysCaP employs a multi-agent design: a Planner that decides when to explore and when to stop, and a Prioritizer that filters implausible interactions and ranks the remainder using a heuristic priority score, enabling efficient, targeted exploration. We evaluate PhysCaP on three real-world tabletop manipulation tasks and a simulated task in LIBERO. The results show that existing passive and naive interactive baselines either fail when physical properties are hidden or over-explore, whereas PhysCaP achieves comparable performance with fewer interactions and reduced execution time. Ablation studies further validate the effectiveness of the proposed physical property extraction modules. Project page: https://physcap.github.io

View source

Similar papers

#artificial intelligence Preprint Nov 2026

Deploying Physically and Epistemically Compliant Agents by Distilling Belief-Space Task and Motion Planners

In this work, we train agents to act under uncertainty according to specified, belief-dependent rules. Deep reinforcement learning (RL) can in theory train agents to follow rules if encoded into well-designed rewards, but in practice, RL fails to perfectly optimize these rewards, leading to ineffective or noncompliant...

Phillip J. K. Christoffersen, Rachel Ma, Patrick R. Barragán et al. · 0 citations

CodeActionBench: Evaluating Agentic Code-as-Policy for Embodied Manipulation

CodeActionBench is introduced, a benchmark of 25 manipulation tasks that evaluates this capability through agentic Code-as-Policy and provides a controlled testbed for measuring how general-purpose models translate their capabilities into manipulation behavior and for examining typical failure scenarios in that process...

Yiheng Lyu, Xueying Jiang, Wen-Hao Li et al. · 3 citations
Preprint Oct 2026

RoboDoctor: Training-Free Recursive Self-Improvement for Robot Policies

Generalist robot policies and language-model agents provide complementary capabilities, but completing a task also requires adapting their use to the consequences of physical interaction. We study execution as a locus of adaptation: feedback guides the current action and supplies evidence for revising subsequent decisi...

He-Yang Yu, Hao-Min Li, Si-Hang Li et al. · 1 citation
Preprint Oct 2026

EmbodiedEvo: Self-Evolving Coding Agent For Physical Intelligence

EmbodiedEvo is a self-evolving coding agent that distills physical feedback into reusable code Skills and diagnostic Lessons that improves weaker executors and shows the largest gains on memory-dependent tasks and successful reuse of coordinated procedures.

Wen-Yi Li, Yi-Zhe Chi, You-Jie Zheng et al. · 7 citations · ⚡3
#artificial intelligence Preprint Oct 2026

RoboAuto: Autonomous Robot Program Search through Multi-Level Physical Understanding

This work introduces RoboAuto, a framework for robot program search that explicitly organizes physical understanding at the object, scene, and interaction levels through three components and shows that the knowledge base transfers across simulators and embodiments, and that program search performance scales with its si...

Bu-Zhao Liu, Jia-Ning Ye, Xiu Yuan et al. · 1 citation

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.