Pith. sign in

REVIEW 12 cited by

Scaling Proprioceptive-Visual Learning with Heterogeneous Pre-trained Transformers

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2409.20537 v1 pith:HD3ORU72 submitted 2024-09-30 cs.RO cs.CVcs.LG

classification cs.ROcs.CVcs.LG
keywords datasetsheterogeneouslearningpolicytasksacrossdatadifferent
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

One of the roadblocks for training generalist robotic models today is heterogeneity. Previous robot learning methods often collect data to train with one specific embodiment for one task, which is expensive and prone to overfitting. This work studies the problem of learning policy representations through heterogeneous pre-training on robot data across different embodiments and tasks at scale. We propose Heterogeneous Pre-trained Transformers (HPT), which pre-train a large, shareable trunk of a policy neural network to learn a task and embodiment agnostic shared representation. This general architecture aligns the specific proprioception and vision inputs from distinct embodiments to a short sequence of tokens and then processes such tokens to map to control robots for different tasks. Leveraging the recent large-scale multi-embodiment real-world robotic datasets as well as simulation, deployed robots, and human video datasets, we investigate pre-training policies across heterogeneity. We conduct experiments to investigate the scaling behaviors of training objectives, to the extent of 52 datasets. HPTs outperform several baselines and enhance the fine-tuned policy performance by over 20% on unseen tasks in multiple simulator benchmarks and real-world settings. See the project website (https://liruiw.github.io/hpt/) for code and videos.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 12 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. What Can Latent World Models Know? Physical Parameter Identifiability in Multimodal Predictive Representations

    cs.LG 2026-07 conditional novelty 7.0 of 10

    Prediction targets, not inputs, decide which physical parameters a latent world model acquires; a certified-recoverable drag parameter stays unlearned under every deterministic prediction objective tested.

  2. EgoWAM: World Action Models Beyond Pixels with In-the-Wild Egocentric Human Data

    cs.RO 2026-07 accept novelty 6.5 of 10

    World Action Model co-training with DINO or 3D-flow targets scales human-to-robot transfer on bimanual tasks far better than behavior cloning, while pixel prediction transfers weakly.

  3. EgoVerse: An Egocentric Human Dataset for Robot Learning from Around the World

    cs.RO 2026-04 unverdicted novelty 6.5 of 10

    EgoVerse releases 1,362 hours of standardized egocentric human data across 1,965 tasks and shows via multi-lab experiments that robot policy performance scales with human data volume when the data aligns with robot ob...

  4. Weights or Skills? A Survey of Robot-Learning Techniques: from Action-Predicting Weights to Robots that Write their Own Skills

    cs.RO 2026-08 conditional novelty 6.0 of 10

    A taxonomy of robot learning on a weights-versus-skills axis, with a five-rung self-improvement ladder whose top cell (feedback plus memory plus search) holds only a few recent systems.

  5. Kepler-Encoder-v0.1: Towards a Multimodal Embedding Model for Robots

    cs.RO 2026-07 conditional novelty 6.0 of 10

    A self-supervised multimodal encoder trained with vision, proprioception, and force yields a vision-only latent that recovers end-effector state and force above vision baselines on RH20T, with modest absolute force accuracy.

  6. Knowledge-Driven Imitation Learning: Enabling Generalization Across Diverse Conditions

    cs.RO 2025-06 conditional novelty 6.0 of 10

    A semantic keypoint graph matched to novel objects lets imitation-learned manipulation policies generalize with a quarter of the demonstrations.

  7. Human2LocoMan: Learning Versatile Quadrupedal Manipulation with Human Pretraining

    cs.RO 2025-06 conditional novelty 6.0 of 10

    Pretraining a modular transformer policy on human demonstrations then finetuning on a small robot dataset improves success on six real quadruped manipulation tasks, including out-of-distribution objects.

  8. Learning Real-World Action-Video Dynamics with Heterogeneous Masked Autoregression

    cs.RO 2025-02 conditional novelty 6.0 of 10

    HMA is a masked autoregressive transformer that predicts future video and actions across many robot embodiments, running up to 15x faster than prior diffusion-based video simulators while matching or improving visual ...

  9. Universal Actions for Enhanced Embodied Foundation Models

    cs.RO 2025-01 conditional novelty 6.0 of 10

    UniAct learns a shared discrete codebook of universal actions for many robots, decodes them with per-robot heads, and reports gains over larger baselines in robot manipulation.

  10. From Multimodal LLMs to Generalist Embodied Agents: Methods and Lessons

    cs.LG 2024-12 conditional novelty 6.0 of 10

    A single MLLM-based agent, finetuned with cross-domain supervision and online RL, achieves strong zero-shot generalization across manipulation, navigation, games, UI control, and planning.

  11. AnyBimanual: Transferring Unimanual Policy for General Bimanual Manipulation

    cs.RO 2024-12 conditional novelty 6.0 of 10

    AnyBimanual transfers pretrained unimanual robot policies to bimanual manipulation via a skill manager and a visual aligner, achieving 32.00% average success on 12 RLBench2 tasks.

  12. Tra-MoE: Learning Trajectory Prediction Model from Multiple Domains for Adaptive Policy Conditioning

    cs.RO 2024-11 conditional novelty 6.0 of 10

    Tra-MoE shows that a sparsely-gated MoE version of the ATM trajectory model improves with out-of-domain video data, while the dense baseline degrades.

Pith tools