journal-club

arXiv Paper Digest - 2026-03-30

Generated on 2026-03-30 10:17:09

Found 10 relevant papers.


1. Knowledge-Guided Manipulation Using Multi-Task Reinforcement Learning

Authors: Aditya Narendra, Mukhammadrizo Maribjonov, Dmitry Makarov, Dmitry Yudin, Aleksandr Panov

Published: 2026-03-25

Categories: cs.RO, cs.AI, cs.LG

Relevance Score: 0.737 (Keyword matches: 3)

Topic: Robotics Manipulation

Abstract: This paper introduces Knowledge Graph based Massively Multi-task Model-based Policy Optimization (KG-M3PO), a framework for multi-task robotic manipulation in partially observable settings that unifies Perception, Knowledge, and Policy. The method augments egocentric vision with an online 3D scene g…

PDF arXiv

2. Meta-Adaptive Beam Search Planning for Transformer-Based Reinforcement Learning Control of UAVs with Overhead Manipulators under Flight Disturbances

Authors: Hazim Alzorgan, Sayed Pedram Haeri Boroujeni, Abolfazl Razi

Published: 2026-03-27

Categories: cs.RO

Relevance Score: 0.647 (Keyword matches: 2)

Topic: Robotics Manipulation

Abstract: Drones equipped with overhead manipulators offer unique capabilities for inspection, maintenance, and contact-based interaction. However, the motion of the drone and its manipulator is tightly linked, and even small attitude changes caused by wind or control imperfections shift the end-effector away…

PDF arXiv

3. Closed-Loop Verbal Reinforcement Learning for Task-Level Robotic Planning

Authors: Dmitrii Plotnikov, Iaroslav Kolomiets, Dmitrii Maliukov, Dmitrij Kosenkov, Daniia Zinniatullina, et al. (+8 more)

Published: 2026-03-23

Categories: cs.RO

Relevance Score: 0.604 (Keyword matches: 2)

Topic: Robotics Manipulation

Abstract: We propose a new Verbal Reinforcement Learning (VRL) framework for interpretable task-level planning in mobile robotic systems operating under execution uncertainty. The framework follows a closed-loop architecture that enables iterative policy improvement through interaction with the physical envir…

PDF arXiv

4. DiffusionAnything: End-to-End In-context Diffusion Learning for Unified Navigation and Pre-Grasp Motion

Authors: Iana Zhura, Yara Mahmoud, Jeffrin Sam, Hung Khang Nguyen, Didar Seyidov, et al. (+2 more)

Published: 2026-03-27

Categories: cs.RO

Relevance Score: 0.602

Topic: Robotics Manipulation

Abstract: Efficiently predicting motion plans directly from vision remains a fundamental challenge in robotics, where planning typically requires explicit goal specification and task-specific design. Recent vision-language-action (VLA) models infer actions directly from visual input but demand massive computa…

PDF arXiv

5. Toward Generalist Neural Motion Planners for Robotic Manipulators: Challenges and Opportunities

Authors: Davood Soleymanzadeh, Ivan Lopez-Sanchez, Hao Su, Yunzhu Li, Xiao Liang, et al. (+1 more)

Published: 2026-03-25

Categories: cs.RO, cs.AI

Relevance Score: 0.600

Topic: Robotics Manipulation

Abstract: State-of-the-art generalist manipulation policies have enabled the deployment of robotic manipulators in unstructured human environments. However, these frameworks struggle in cluttered environments primarily because they utilize auxiliary modules for low-level motion planning and control. Motion pl…

PDF arXiv

6. LaMP: Learning Vision-Language-Action Policies with 3D Scene Flow as Latent Motion Prior

Authors: Xinkai Wang, Chenyi Wang, Yifu Xu, Mingzhe Ye, Fu-Cheng Zhang, et al. (+5 more)

Published: 2026-03-26

Categories: cs.CV, cs.RO

Relevance Score: 0.597

Topic: Robotics Manipulation

Abstract: We introduce \textbf{LaMP}, a dual-expert Vision-Language-Action framework that embeds dense 3D scene flow as a latent motion prior for robotic manipulation. Existing VLA models regress actions directly from 2D semantic visual features, forcing them to learn complex 3D physical interactions implicit…

PDF arXiv

7. SG-VLA: Learning Spatially-Grounded Vision-Language-Action Models for Mobile Manipulation

Authors: Ruisen Tu, Arth Shukla, Sohyun Yoo, Xuanlin Li, Junxi Li, et al. (+3 more)

Published: 2026-03-24

Categories: cs.RO

Relevance Score: 0.591 (Keyword matches: 1)

Topic: Robotics Manipulation

Abstract: Vision-Language-Action (VLA) models show promise for robotic control, yet performance in complex household environments remains sub-optimal. Mobile manipulation requires reasoning about global scene layout, fine-grained geometry, and high-dimensional continuous actions, making standard imitation lea…

PDF arXiv

8. Persistent Robot World Models: Stabilizing Multi-Step Rollouts via Reinforcement Learning

Authors: Jai Bardhan, Patrik Drozdik, Josef Sivic, Vladimir Petrik

Published: 2026-03-26

Categories: cs.RO, cs.CV

Relevance Score: 0.589 (Keyword matches: 2)

Topic: Robotics Manipulation

Abstract: Action-conditioned robot world models generate future video frames of the manipulated scene given a robot action sequence, offering a promising alternative for simulating tasks that are difficult to model with traditional physics engines. However, these models are optimized for short-term prediction…

PDF arXiv

9. DFM-VLA: Iterative Action Refinement for Robot Manipulation via Discrete Flow Matching

Authors: Jiayi Chen, Wenxuan Song, Shuai Chen, Jingbo Wang, Zhijun Li, et al. (+1 more)

Published: 2026-03-27

Categories: cs.RO, cs.CV

Relevance Score: 0.589 (Keyword matches: 1)

Topic: Robotics Manipulation

Abstract: Vision–Language–Action (VLA) models that encode actions using a discrete tokenization scheme are increasingly adopted for robotic manipulation, but existing decoding paradigms remain fundamentally limited. Whether actions are decoded sequentially by autoregressive VLAs or in parallel by discrete d…

PDF arXiv

10. SoftMimicGen: A Data Generation System for Scalable Robot Learning in Deformable Object Manipulation

Authors: Masoud Moghani, Mahdi Azizian, Animesh Garg, Yuke Zhu, Sean Huver, et al. (+1 more)

Published: 2026-03-26

Categories: cs.RO

Relevance Score: 0.585 (Keyword matches: 1)

Topic: Robotics Manipulation

Abstract: Large-scale robot datasets have facilitated the learning of a wide range of robot manipulation skills, but these datasets remain difficult to collect and scale further, owing to the intractable amount of human time, effort, and cost required. Simulation and synthetic data generation have proven to b…

PDF arXiv