CometVLA: Co-Training on an Embodied Data Pyramid towards Physical Understanding

• URL: http://arxiv.org/abs/2608.30289v1
• Date: Mon, 31 Aug 2026 05:55:34 GMT
• Title: CometVLA: Co-Training on an Embodied Data Pyramid towards Physical Understanding
• Abstract summary: Vision-language-action (VLA) models remain brittle in manipulation tasks that require physical commonsense.<n>It remains unclear whether improved VLM physical understanding actually benefits downstream action generation.<n>We present CometVLA, an embodied physical VQA corpus and benchmark strictly aligned with the robot's action data and embodiment.
• Score: 56.55196056616493
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Vision-language-action (VLA) models remain brittle in manipulation tasks that require physical commonsense. Current physical VQA data is typically disembodied and misaligned with robot action domains. Egocentric videos are used only as auxiliary pre-training. It remains unclear whether improved VLM physical understanding actually benefits downstream action generation. Therefore, we present CometVLA to close this gap. We construct CometData and CometBench, an embodied physical VQA corpus and benchmark strictly aligned with the robot's action data and embodiment. We introduce Global Action Prior (GAP) tokens, a compact learnable bottleneck that isolates task-agnostic motion regularities and lets the action head consume physical commonsense without corrupting the pre-trained VLM backbone. We co-train CometVLA across the embodied data pyramid, spanning teleoperation, simulation, egocentric trajectories, and VQA layers. On real-world manipulation tasks and RoboTwin simulation, CometVLA consistently outperforms strong VLA baselines. Correlation analysis shows that stronger VLM performance on CometBench indicates higher VLA success rates. Results demonstrate that physical understanding pre-training genuinely benefits downstream manipulation.

Related papers

• Beyond Data Scaling: Representation-Centric Continued Pre-training for Vision-Language-Action Models [120.24139942108405]
  Scaling robot data is crucial for building generalist Vision-Language-Action (VLA) models.<n>Under a fixed robot-data budget, continued pre-training must turn limited trajectories into transferable visual-action knowledge.<n>We propose VLAct, a VLA-oriented VLM backbone trained on broad, heterogeneous, multi-embodiment robot data.
  arXiv  Detail & Related papers  (2026-08-27T17:59:40Z)
• Does VLA Even Know the Basics? Measuring Commonsense and World Knowledge Retention in Vision-Language-Action Models [42.38903028184938]
  We introduce Act2Answer, a protocol that adapts VLM knowledge benchmarks to VLA evaluation by requiring agents to answer through action.<n>Each question becomes a short tabletop episode where the agent performs a single object-placement action to select among candidate answers, yielding an action-grounded success rate with reduced control confounds.<n>In a large-scale study of 7 VLA models and 9 VLM baselines, we systematically rank models across categories, finding that VLAs show solid performance on simple concepts while exhibiting larger gaps on richer semantic categories relative to their source VLMs.
  arXiv  Detail & Related papers  (2026-06-17T17:20:46Z)
• Two Bridges, One Pathway: From VLMs to Generalizable VLAs with Embodied Trajectory-Coupled Data [93.7685703383343]
  Vision-language models (VLMs) are powerful general-purpose reasoners, yet converting them into robot control policies is surprisingly difficult.<n>We argue that this gap can be bridged gradually with the right intermediate data.<n>We show that mixing task-relevant out-of-distribution ETC data with a small amount of action data enables the model to generalize to novel visual-language conditions.
  arXiv  Detail & Related papers  (2026-06-07T08:57:51Z)
• QuoVLA: Quotient Space for Vision-Language-Action Models [51.02329790939691]
  Vision-Language-Action (VLA) models commonly adapt pretrained VLM latents to robot control by mapping visual observations and language instructions to continuous actions.<n>Our textitQuotient Theory for VLA shows that pretrained VLM latents are not action-insufficient but action-sufficient.<n>We propose QuoVLA, a quotient-space framework for VLA that compresses pretrained VLM latents into action-sufficient representations.
  arXiv  Detail & Related papers  (2026-05-24T06:28:53Z)
• UAM: A Dual-Stream Perspective on Forgetting in VLA Training [20.50709311546184]
  Vision--language--action (VLA) models are typically built by fine-tuning a pretrained vision--language model (VLM) on action data.<n>We show that this standard recipe systematically erodes the VLM's multimodal competence.<n>Inspired by the two-stream organization of biological vision, we trace this degradation to a structural bottleneck.
  arXiv  Detail & Related papers  (2026-05-15T08:45:37Z)
• mimic-video: Video-Action Models for Generalizable Robot Control Beyond VLAs [5.109732854501585]
  We introduce mimic-video, a novel Video-Action Model (VAM) that pairs a pretrained Internet-scale video model with a flow matching-based action decoder conditioned on its latent representations.<n>Our approach achieves state-of-the-art performance on simulated and real-world robotic manipulation tasks, improving sample efficiency by 10x and convergence speed by 2x compared to traditional VLA architectures.
  arXiv  Detail & Related papers  (2025-12-17T18:47:31Z)
• SIMPACT: Simulation-Enabled Action Planning using Vision-Language Models [60.80050275581661]
  Vision-Language Models (VLMs) exhibit remarkable common-sense and semantic reasoning capabilities.<n>They lack a grounded understanding of physical dynamics.<n>We present S, a test-time, SIMulation-enabled ACTion Planning framework.<n>Our method demonstrates state-of-the-art performance on five challenging, real-world rigid-body and deformable manipulation tasks.
  arXiv  Detail & Related papers  (2025-12-05T18:51:03Z)
• CoT-VLA: Visual Chain-of-Thought Reasoning for Vision-Language-Action Models [89.44024245194315]
  We introduce a method that incorporates explicit visual chain-of-thought (CoT) reasoning into vision-language-action models (VLAs)<n>We introduce CoT-VLA, a state-of-the-art 7B VLA that can understand and generate visual and action tokens.<n>Our experimental results demonstrate that CoT-VLA achieves strong performance, outperforming the state-of-the-art VLA model by 17% in real-world manipulation tasks and 6% in simulation benchmarks.
  arXiv  Detail & Related papers  (2025-03-27T22:23:04Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.