Data-Asymmetric Latent Imagination and Reranking for 3D Robotic Imitation Learning

• URL: http://arxiv.org/abs/2605.10166v1
• Date: Mon, 11 May 2026 08:16:13 GMT
• Title: Data-Asymmetric Latent Imagination and Reranking for 3D Robotic Imitation Learning
• Abstract summary: We propose DALI-R, a framework for 3D robotic imitation learning from mixed-quality trajectories.<n>It learns a Latent World Model over 3D point clouds for imagined rollouts and a Task Completion Scorer that reranks candidate action chunks.<n>Across the two evaluated 3D base policies, DALI-R achieves an average $6.8$% improvement in success rate.
• Score: 15.264898640069193
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Robotic imitation learning typically assumes access to optimal demonstrations, yet real-world data collection often yields suboptimal, exploratory, or even failed trajectories. Discarding such data wastes valuable information about environment dynamics and failure modes, which can instead be leveraged to improve decision-making. While 3D policies reduce reliance on high-quality demonstrations through strong spatial generalization, they still require large-scale data to achieve high task success. To address this, we propose DALI-R, a Data-Asymmetric Latent Imagination and Reranking framework for 3D robotic imitation learning from mixed-quality trajectories. It learns a Latent World Model over 3D point clouds for imagined rollouts and a Task Completion Scorer that reranks candidate action chunks, improving decision-making without additional high-quality demonstrations. We instantiate DALI-R with both diffusion and efficient flow-matching policies and evaluate it on Adroit and MetaWorld benchmarks. Across the two evaluated 3D base policies, DALI-R achieves an average $6.8$\% improvement in success rate while incurring less than $0.7\times$ additional inference overhead.

Related papers

• Masked Diffusion Language Models are Strong and Steerable Text-Based World Models for Agentic RL [0.8546394944138487]
  We formalize text-based world modeling as a steerable transition-dynamics problem decomposed into initial state.<n>We curate 239,403 grounded state-action trajectories spanning nine open-source environments and twelve frontier model families.<n>We introduce a plug-and-play GRPO training framework with deterministic state checks, and perform zero-shot transfer ablations on three OOD environments.
  arXiv  Detail & Related papers  (2026-05-07T00:40:32Z)
• An Efficient Metric for Data Quality Measurement in Imitation Learning [1.5469452301122175]
  Fine-tuning pre-trained policies with end-user demonstrations collected in deployment environments is a promising strategy to address this challenge.<n>Existing automated approaches for curating demonstration data require policy rollouts in the environment.<n>We propose a fast, efficient, and fully automated demonstration ranking metric based on the power spectral density (PSD) of demonstration trajectories.
  arXiv  Detail & Related papers  (2026-05-02T17:16:50Z)
• AffordGen: Generating Diverse Demonstrations for Generalizable Object Manipulation with Afford Correspondence [40.46817120361278]
  AffordGen is a framework to generate new robot manipulation trajectories.<n>It combines the semantic generalizability of affordances with the reactive robustness of end-to-end learning.<n>Experiments show that policies trained with AffordGen achieve high success rates and enable zero-shot generalization to truly unseen objects.
  arXiv  Detail & Related papers  (2026-04-12T10:56:31Z)
• RISE: Self-Improving Robot Policy with Compositional World Model [52.227523057681786]
  We present RISE, a scalable framework of robotic reinforcement learning via imagination.<n>At its core is a Compositional World Model that predicts multi-view future via a controllable dynamics model, and (ii) evaluates imagined outcomes with a progress value model.<n>These components are integrated into a closed-loop self-improving pipeline that continuously generates imaginary rollouts, estimates advantages, and updates the policy in imaginary space without costly physical interaction.
  arXiv  Detail & Related papers  (2026-02-11T17:43:36Z)
• World Models for Autonomous Navigation of Terrestrial Robots from LIDAR Observations [0.7239024032079358]
  This paper presents a novel model-based RL framework built on top of the DreamerV3 algorithm.<n>It integrates a Multi-Layer Perceptron Variational Autoencoder (MLP-VAE) within a world model to encode high-dimensional LIDAR readings into compact latent representations.<n>Experiments on simulated TurtleBot3 navigation tasks demonstrate that the proposed architecture achieves faster convergence and higher success rate.
  arXiv  Detail & Related papers  (2025-12-03T04:15:31Z)
• Robust3D-CIL: Robust Class-Incremental Learning for 3D Perception [7.507868991415516]
  3D perception plays a crucial role in real-world applications such as autonomous driving, robotics, and AR/VR.<n>Adopting class-incremental learning (CIL) becomes particularly essential.<n>We introduce a point cloud downsampling-based replay method to utilize the limited replay buffer memory more efficiently.<n>Our method improves the performance of replay-based CIL baselines by 2% to 11%, proving its effectiveness and promising potential for real-world 3D applications.
  arXiv  Detail & Related papers  (2025-03-18T03:36:17Z)
• GEAL: Generalizable 3D Affordance Learning with Cross-Modal Consistency [50.11520458252128]
  Existing 3D affordance learning methods struggle with generalization and robustness due to limited annotated data.<n>We propose GEAL, a novel framework designed to enhance the generalization and robustness of 3D affordance learning by leveraging large-scale pre-trained 2D models.<n>GEAL consistently outperforms existing methods across seen and novel object categories, as well as corrupted data.
  arXiv  Detail & Related papers  (2024-12-12T17:59:03Z)
• Enhancing Generalizability of Representation Learning for Data-Efficient 3D Scene Understanding [50.448520056844885]
  We propose a generative Bayesian network to produce diverse synthetic scenes with real-world patterns. A series of experiments robustly display our method's consistent superiority over existing state-of-the-art pre-training approaches.
  arXiv  Detail & Related papers  (2024-06-17T07:43:53Z)
• FILP-3D: Enhancing 3D Few-shot Class-incremental Learning with Pre-trained Vision-Language Models [59.13757801286343]
  Few-shot class-incremental learning aims to mitigate the catastrophic forgetting issue when a model is incrementally trained on limited data.<n>We introduce the FILP-3D framework with two novel components: the Redundant Feature Eliminator (RFE) for feature space misalignment and the Spatial Noise Compensator (SNC) for significant noise.
  arXiv  Detail & Related papers  (2023-12-28T14:52:07Z)
• Domain Adaptive 3D Pose Augmentation for In-the-wild Human Mesh Recovery [32.73513554145019]
  Domain Adaptive 3D Pose Augmentation (DAPA) is a data augmentation method that enhances the model's generalization ability in in-the-wild scenarios. We show quantitatively that finetuning with DAPA effectively improves results on benchmarks 3DPW and AGORA.
  arXiv  Detail & Related papers  (2022-06-21T15:02:31Z)
• What Stops Learning-based 3D Registration from Working in the Real World? [53.68326201131434]
  This work identifies the sources of 3D point cloud registration failures, analyze the reasons behind them, and propose solutions. Ultimately, this translates to a best-practice 3D registration network (BPNet), constituting the first learning-based method able to handle previously-unseen objects in real-world data. Our model generalizes to real data without any fine-tuning, reaching an accuracy of up to 67% on point clouds of unseen objects obtained with a commercial sensor.
  arXiv  Detail & Related papers  (2021-11-19T19:24:27Z)
• Secrets of 3D Implicit Object Shape Reconstruction in the Wild [92.5554695397653]
  Reconstructing high-fidelity 3D objects from sparse, partial observation is crucial for various applications in computer vision, robotics, and graphics. Recent neural implicit modeling methods show promising results on synthetic or dense datasets. But, they perform poorly on real-world data that is sparse and noisy. This paper analyzes the root cause of such deficient performance of a popular neural implicit model.
  arXiv  Detail & Related papers  (2021-01-18T03:24:48Z)

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.