Flame3D enables zero-shot compositional 3D scene reasoning by representing scenes as editable visual-textual memories exposed to agentic MLLMs through composable and synthesizable spatial tools.
hub
CaP-X: A framework for benchmarking and improving coding agents for robot manipulation
15 Pith papers cite this work. Polarity classification is still indexing.
15
Pith papers citing it
abstract
"Code-as-Policy" considers how executable code can complement data-intensive Vision-Language-Action (VLA) methods, yet their effectiveness as autonomous controllers for embodied manipulation remains underexplored. We present CaP-X, an open-access framework for systematically studying Code-as-Policy agents in robot manipulation. At its core is CaP-Gym, an interactive environment in which agents control robots by synthesizing and executing programs that compose perception and control primitives. Building on this foundation, CaP-Bench evaluates frontier language and vision-language models across varying levels of abstraction, interaction, and perceptual grounding. Across 12 models, CaP-Bench reveals a consistent trend: performance improves with human-crafted abstractions but degrades as these priors are removed, exposing a dependence on designer scaffolding. At the same time, we observe that this gap can be mitigated through scaling agentic test-time computation--through multi-turn interaction, structured execution feedback, visual differencing, automatic skill synthesis, and ensembled reasoning--substantially improves robustness even when agents operate over low-level primitives. These findings allow us to derive CaP-Agent0, a training-free framework that recovers human-level reliability on several manipulation tasks in simulation and on real embodiments. We further introduce CaP-RL, showing reinforcement learning with verifiable rewards improves success rates and transfers from sim2real with minimal gap. Together, CaP-X provides a principled, open-access platform for advancing embodied coding agents.
hub tools
citation-role summary
background 2
citation-polarity summary
years
2026 15roles
background 2polarities
background 2representative citing papers
ENPIRE supplies four modules (Environment, Policy Improvement, Rollout, Evolution) that turn real-world robot training into an autonomous optimization loop driven by coding agents.
ReSYNC learns recovery skills via RL then discovers and refines relational predicates to enable abstract planning that generalizes failure avoidance to unseen long-horizon tasks, outperforming baselines by over 50% in simulation and transferring to real robots.
Flow Reversal Steering steers flow matching generalist policies by reversing suboptimal actions to nearby better modes, enabling improved zero-shot control, quick distillation, and RL bootstrapping in robotic manipulation.
VoLoAgent uses a VLM to steer heterogeneous robot capabilities as interruptible tools for long-horizon manipulation and introduces the RoboVoLo benchmark, claiming substantial outperformance over single VLA/VLM or tool-based systems with real-robot validation.
A memory-guided LLM harness that calls a frozen VLA only for contact-rich phases lifts success to 82.4% on LIBERO-Pro, 55.4% on RoboCasa365, and 58.4% on RoboTwin C2R with no policy finetuning.
VLA skills that score 77-100% in isolation stall from chained states in BEHAVIOR-1K; failures trace to next-skill readiness, target grounding, and control execution.
SPARK reaches 43.7% success on six LIBERO-PRO cells by LLM-generated typed behavior trees plus multi-prompt perception and recovery, more than doubling CaP-Agent0 and VLA baselines.
RATs agents generate and solve their own exploratory tasks during play, distill successful code into a skill library, and reuse it to improve held-out task performance by 20.6 and 17.0 points on two benchmarks.
Guava harness enables 4B open-source models to achieve performance comparable to frontier models on embodied manipulation tasks by distilling capabilities from under 2K simulation trajectories using three identified design principles.
APT pretrains the action expert as a vision-action prior on frozen VLM features then adds language through gated fusion to improve OOD instruction generalization in continuous-action VLA policies.
GraspGen-X extends diffusion 6-DOF grasping to cross-embodiment via swept-volume gripper encoding, trained on procedural grippers and 2B grasps, claiming best zero-shot generalization to novel grippers in sim and real tests.
InSight enables autonomous acquisition of manipulation primitives in VLAs via automated segmentation for steerability and a VLM-guided data flywheel that generates and integrates new demonstrations for tasks like pouring and sweeping.
Survey framing LLM agents as model-plus-harness systems, decomposing harness responsibilities, mapping them to tasks, and highlighting open challenges in evaluation, safety, and co-evolution.
ABot-Claw is an embodied software layer that adds unified robot scheduling, cross-embodiment visual memory, and critic-driven replanning on top of OpenClaw to support persistent multi-robot execution from natural-language goals.
citing papers explorer
-
Flame3D: Zero-shot Compositional Reasoning of 3D Scenes with Agentic Language Models
Flame3D enables zero-shot compositional 3D scene reasoning by representing scenes as editable visual-textual memories exposed to agentic MLLMs through composable and synthesizable spatial tools.
-
ENPIRE: Agentic Robot Policy Self-Improvement in the Real World
ENPIRE supplies four modules (Environment, Policy Improvement, Rollout, Evolution) that turn real-world robot training into an autonomous optimization loop driven by coding agents.
-
Recover, Discover, Plan: Learning Skills and Concepts from Robot Failures
ReSYNC learns recovery skills via RL then discovers and refines relational predicates to enable abstract planning that generalizes failure avoidance to unseen long-horizon tasks, outperforming baselines by over 50% in simulation and transferring to real robots.
-
Improving Robotic Generalist Policies via Flow Reversal Steering
Flow Reversal Steering steers flow matching generalist policies by reversing suboptimal actions to nearby better modes, enabling improved zero-shot control, quick distillation, and RL bootstrapping in robotic manipulation.
-
VoLo: A Physical Orchestrator for Open-Vocabulary Long-Horizon Manipulation
VoLoAgent uses a VLM to steer heterogeneous robot capabilities as interruptible tools for long-horizon manipulation and introduces the RoboVoLo benchmark, claiming substantial outperformance over single VLA/VLM or tool-based systems with real-robot validation.
-
Harness VLA: Steering Frozen VLAs into Reliable Manipulation Primitives via Memory-Guided Agents
A memory-guided LLM harness that calls a frozen VLA only for contact-rich phases lifts success to 82.4% on LIBERO-Pro, 55.4% on RoboCasa365, and 58.4% on RoboTwin C2R with no policy finetuning.
-
Diagnosing Semantic Handoff Failures in Agent-Orchestrated Vision-Language-Action Skill Composition
VLA skills that score 77-100% in isolation stall from chained states in BEHAVIOR-1K; failures trace to next-skill readiness, target grounding, and control execution.
-
Sequential Planning via Anchored Robotic Keypoints
SPARK reaches 43.7% success on six LIBERO-PRO cells by LLM-generated typed behavior trees plus multi-prompt perception and recovery, more than doubling CaP-Agent0 and VLA baselines.
-
Playful Agentic Robot Learning
RATs agents generate and solve their own exploratory tasks during play, distill successful code into a skill library, and reuse it to improve held-out task performance by 20.6 and 17.0 points on two benchmarks.
-
Guava: An Effective and Universal Harness for Embodied Manipulation
Guava harness enables 4B open-source models to achieve performance comparable to frontier models on embodied manipulation tasks by distilling capabilities from under 2K simulation trajectories using three identified design principles.
-
APT: Action Expert Pretraining Improves Instruction Generalization of Vision-Language-Action Policies
APT pretrains the action expert as a vision-action prior on frozen VLM features then adds language through gated fusion to improve OOD instruction generalization in continuous-action VLA policies.
-
GraspGen-X: Cross-Embodiment 6-DOF Diffusion-based Grasping
GraspGen-X extends diffusion 6-DOF grasping to cross-embodiment via swept-volume gripper encoding, trained on procedural grippers and 2B grasps, claiming best zero-shot generalization to novel grippers in sim and real tests.
-
InSight: Self-Guided Skill Acquisition via Steerable VLAs
InSight enables autonomous acquisition of manipulation primitives in VLAs via automated segmentation for steerability and a VLM-guided data flywheel that generates and integrates new demonstrations for tasks like pouring and sweeping.
-
From Question Answering to Task Completion: A Survey on Agent System and Harness Design
Survey framing LLM agents as model-plus-harness systems, decomposing harness responsibilities, mapping them to tasks, and highlighting open challenges in evaluation, safety, and co-evolution.
-
ABot-Claw: A Foundation for Persistent, Cooperative, and Self-Evolving Robotic Agents
ABot-Claw is an embodied software layer that adds unified robot scheduling, cross-embodiment visual memory, and critic-driven replanning on top of OpenClaw to support persistent multi-robot execution from natural-language goals.