Pith. sign in

REVIEW 14 cited by

Behavior Transformers: Cloning k modes with one stone

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 2206.11251 v2 pith:MS7VSGNK submitted 2022-06-22 cs.LG cs.AIcs.CVcs.RO

Behavior Transformers: Cloning k modes with one stone

classification cs.LG cs.AIcs.CVcs.RO
keywords behaviordatasetsmodesactioncloninghumanlargeleverage
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

While behavior learning has made impressive progress in recent times, it lags behind computer vision and natural language processing due to its inability to leverage large, human-generated datasets. Human behaviors have wide variance, multiple modes, and human demonstrations typically do not come with reward labels. These properties limit the applicability of current methods in Offline RL and Behavioral Cloning to learn from large, pre-collected datasets. In this work, we present Behavior Transformer (BeT), a new technique to model unlabeled demonstration data with multiple modes. BeT retrofits standard transformer architectures with action discretization coupled with a multi-task action correction inspired by offset prediction in object detection. This allows us to leverage the multi-modal modeling ability of modern transformers to predict multi-modal continuous actions. We experimentally evaluate BeT on a variety of robotic manipulation and self-driving behavior datasets. We show that BeT significantly improves over prior state-of-the-art work on solving demonstrated tasks while capturing the major modes present in the pre-collected datasets. Finally, through an extensive ablation study, we analyze the importance of every crucial component in BeT. Videos of behavior generated by BeT are available at https://notmahi.github.io/bet

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 14 Pith papers

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

  1. Action-Prior Denoising for Smooth Real-Time Chunking

    cs.RO 2026-05 unverdicted novelty 7.0

    Soft RTC uses partially denoised states for overlap tokens and token-wise blending to reduce action delta and jerk by ~9% versus hard RTC while matching solve rates on Kinetix levels.

  2. SkiP: When to Skip and When to Refine for Efficient Robot Manipulation

    cs.RO 2026-05 unverdicted novelty 7.0

    SkiP introduces action relabeling and Motion Spectrum Keying to skip redundant steps in robot trajectories, cutting executed steps by 15-40% while maintaining success rates across 72 simulated and 3 real tasks.

  3. DSSP: Diffusion State Space Policy with Full-History Encoding

    cs.RO 2026-05 conditional novelty 7.0

    DSSP is a history-conditioned diffusion state space policy that uses SSMs to encode full observation streams with an auxiliary dynamics objective and hierarchical fusion, achieving SOTA results with reduced model size...

  4. Referring-Aware Visuomotor Policy Learning for Closed-Loop Manipulation

    cs.RO 2026-04 unverdicted novelty 7.0

    ReV is a referring-aware visuomotor policy using coupled diffusion heads for real-time trajectory replanning in robotic manipulation, trained solely via targeted perturbations to expert demonstrations and achieving hi...

  5. Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware

    cs.RO 2023-04 conditional novelty 7.0

    Low-cost imprecise robots achieve 80-90% success on six fine bimanual manipulation tasks using imitation learning with a new Action Chunking with Transformers algorithm trained on only 10 minutes of demonstrations.

  6. Diffusion Policies as an Expressive Policy Class for Offline Reinforcement Learning

    cs.LG 2022-08 unverdicted novelty 7.0

    Diffusion-QL uses conditional diffusion models as expressive policies in offline RL by coupling behavior cloning with Q-value maximization, achieving SOTA on most D4RL tasks.

  7. Patch Policy: Efficient Embodied Control via Dense Visual Representations

    cs.RO 2026-07 conditional novelty 6.0

    Patch Policy shows that frozen dense ViT patch tokens, consumed through a block-causal attention mask, let lightweight robot policies beat pooled-feature policies and even a fine-tuned 7B vision-language-action model.

  8. Continuous Reasoning for Vision-Language-Action

    cs.RO 2026-05 unverdicted novelty 6.0

    Continuous Reasoning for VLA introduces a shared Gaussian latent for continuous thoughts, trained with self-verification to improve action prediction on LIBERO-PRO and real robots.

  9. LLM-Guided Future Hypotheses for Horizon-Aware Exploration in Multi-Step Robot Manipulation

    cs.RO 2026-05 unverdicted novelty 6.0

    FEC conditions policies on LLM-guided short-horizon future videos via a three-stage pipeline, yielding performance gains for BC+RL over no-future baselines on RoboCasa and CALVIN while mismatched futures degrade results.

  10. SID: Sliding into Distribution for Robust Few-Demonstration Manipulation

    cs.RO 2026-05 unverdicted novelty 6.0

    SID achieves approximately 90% success on six real-world manipulation tasks with only two demonstrations under out-of-distribution initializations, with less than 10% performance drop under distractors and disturbances.

  11. AFFORD2ACT: Affordance-Guided Automatic Keypoint Selection for Generalizable and Lightweight Robotic Manipulation

    cs.RO 2025-10 unverdicted novelty 6.0

    AFFORD2ACT distills a minimal set of affordance-guided 2D keypoints from text and a single image to train a 38-dimensional gated transformer policy that achieves 82% success on unseen objects and scenes.

  12. Unified World Models: Coupling Video and Action Diffusion for Pretraining on Large Robotic Datasets

    cs.RO 2025-04 unverdicted novelty 6.0

    Unified World Models couple video and action diffusion inside one transformer with independent timesteps, enabling pretraining on heterogeneous robot datasets that include action-free video and producing more generali...

  13. Mobile ALOHA: Learning Bimanual Mobile Manipulation with Low-Cost Whole-Body Teleoperation

    cs.RO 2024-01 conditional novelty 6.0

    A low-cost whole-body teleoperation system enables effective imitation learning for complex bimanual mobile manipulation by co-training on mobile and static demonstration datasets.

  14. Can Predicted Dynamics Exist in the Physical World?

    cs.RO 2026-05 unverdicted novelty 4.0

    Physical admissibility is defined as a prediction-control interface using kinematic, dynamic, and composed-horizon conditions to reject invalid dynamics proposals, with AUC 0.957 on LeRobot PushT and 87-89% prevention...