# SkillMimic-V2: Learning Robust and Generalizable Interaction Skills from Sparse and Noisy Demonstrations

arXiv: https://arxiv.org/abs/2505.02094 · HTML: https://mlsanity.com/paper/2505.02094/

## Key facts

- Cited by 25 papers in the map (data through 14 Oct 2026).
- Placed in Deep RL algorithms and representations, one of 15 directions in Deep reinforcement learning methods.
- Builds on 49 of its 71 references that are in the map.
- First posted on arXiv 4 May 2025.

## Placed in

- Deep reinforcement learning methods / [Deep RL algorithms and representations](https://mlsanity.com/deeprl/deep-rl-algorithms-and-representations/)

## Builds on (most cited first, 30 of 49)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Generative Adversarial Imitation Learning](https://mlsanity.com/paper/1606.03476/) (2016)
- [Isaac Gym: High Performance GPU-Based Physics Simulation For Robot Learning](https://mlsanity.com/paper/2108.10470/) (2021)
- [Learning Dexterous In-Hand Manipulation](https://mlsanity.com/paper/1808.00177/) (2018)
- [Learning agile and dynamic motor skills for legged robots](https://mlsanity.com/paper/1901.08652/) (2019)
- [GRAB: A Dataset of Whole-Body Human Grasping of Objects](https://mlsanity.com/paper/2008.11200/) (2020)
- [Mobile ALOHA: Learning Bimanual Mobile Manipulation with Low-Cost Whole-Body Teleoperation](https://mlsanity.com/paper/2401.02117/) (2024)
- [HOI4D: A 4D Egocentric Dataset for Category-Level Human-Object Interaction](https://mlsanity.com/paper/2203.01577/) (2022)
- [MimicGen: A Data Generation System for Scalable Robot Learning using Human Demonstrations](https://mlsanity.com/paper/2310.17596/) (2023)
- [ARCTIC: A Dataset for Dexterous Bimanual Hand-Object Manipulation](https://mlsanity.com/paper/2204.13662/) (2022)
- [Perpetual Humanoid Control for Real-time Simulated Avatars](https://mlsanity.com/paper/2305.06456/) (2023)
- [Object Motion Guided Human Motion Synthesis](https://mlsanity.com/paper/2309.16237/) (2023)
- [OmniH2O: Universal and Dexterous Human-to-Humanoid Whole-Body Teleoperation and Learning](https://mlsanity.com/paper/2406.08858/) (2024)
- [InterDiff: Generating 3D Human-Object Interactions with Physics-Informed Diffusion](https://mlsanity.com/paper/2308.16905/) (2023)
- [DeepMimic: Example-Guided Deep Reinforcement Learning of Physics-Based Character Skills](https://mlsanity.com/paper/1804.02717/) (2018)
- [Hand-Object Contact Consistency Reasoning for Human Grasps Generation](https://mlsanity.com/paper/2104.03304/) (2021)
- [Scaling Up Dynamic Human-Scene Interaction Modeling](https://mlsanity.com/paper/2403.08629/) (2024)
- [THE COLOSSEUM: A Benchmark for Evaluating Generalization for Robotic Manipulation](https://mlsanity.com/paper/2402.08191/) (2024)
- [UniDexGrasp: Universal Robotic Dexterous Grasping via Learning Diverse Proposal Generation and Goal-Conditioned Policy](https://mlsanity.com/paper/2303.00938/) (2023)
- [MaskedMimic: Unified Physics-Based Character Control Through Masked Motion Inpainting](https://mlsanity.com/paper/2409.14393/) (2024)
- [HumanoidBench: Simulated Humanoid Benchmark for Whole-Body Locomotion and Manipulation](https://mlsanity.com/paper/2403.10506/) (2024)
- [Synthesizing Physical Character-Scene Interactions](https://mlsanity.com/paper/2302.00883/) (2023)
- [CALM: Conditional Adversarial Latent Models for Directable Virtual Characters](https://mlsanity.com/paper/2305.02195/) (2023)
- [Unified Human-Scene Interaction via Prompted Chain-of-Contacts](https://mlsanity.com/paper/2309.07918/) (2023)
- [HOVER: Versatile Neural Whole-Body Controller for Humanoid Robots](https://mlsanity.com/paper/2410.21229/) (2024)
- [MotionStreamer: Streaming Motion Generation via Diffusion-based Autoregressive Model in Causal Latent Space](https://mlsanity.com/paper/2503.15451/) (2025)
- [Move as You Say, Interact as You Can: Language-guided Human Motion Generation with Scene Affordance](https://mlsanity.com/paper/2403.18036/) (2024)
- [Full-Body Articulated Human-Object Interaction](https://mlsanity.com/paper/2212.10621/) (2022)
- [HOLD: Category-agnostic 3D Reconstruction of Interacting Hands and Objects from Video](https://mlsanity.com/paper/2311.18448/) (2023)
- [PhysHOI: Physics-Based Imitation of Dynamic Human-Object Interaction](https://mlsanity.com/paper/2312.04393/) (2023)

## Built on by (most cited first, 25 of 25)

- [BeyondMimic: From Motion Tracking to Versatile Humanoid Control via Guided Diffusion](https://mlsanity.com/paper/2508.08241/) (2025)
- [Go to Zero: Towards Zero-shot Motion Generation with Million-scale Data](https://mlsanity.com/paper/2507.07095/) (2025)
- [PhysHSI: Towards a Real-World Generalizable and Natural Humanoid-Scene Interaction System](https://mlsanity.com/paper/2510.11072/) (2025)
- [HumanX: Toward Agile and Generalizable Humanoid Interaction Skills from Human Videos](https://mlsanity.com/paper/2602.02473/) (2026)
- [KungfuBot2: Learning Versatile Motion Skills for Humanoid Whole-Body Control](https://mlsanity.com/paper/2509.16638/) (2025)
- [CoDA: Coordinated Diffusion Noise Optimization for Whole-Body Manipulation of Articulated Objects](https://mlsanity.com/paper/2505.21437/) (2025)
- [InterPrior: Scaling Generative Control for Physics-Based Human-Object Interactions](https://mlsanity.com/paper/2602.06035/) (2026)
- [InterAgent: Physics-based Multi-agent Command Execution via Diffusion on Interaction Graphs](https://mlsanity.com/paper/2512.07410/) (2025)
- [TeamHOI: Learning a Unified Policy for Cooperative Human-Object Interactions with Any Team Size](https://mlsanity.com/paper/2603.07988/) (2026)
- [Heracles: Bridging Precise Tracking and Generative Synthesis for General Humanoid Control](https://mlsanity.com/paper/2603.27756/) (2026)
- [FieldGen: From Teleoperated Pre-Manipulation Trajectories to Field-Guided Data Generation](https://mlsanity.com/paper/2510.20774/) (2025)
- [DiscreteRTC: Discrete Diffusion Policies are Natural Asynchronous Executors](https://mlsanity.com/paper/2604.25050/) (2026)
- [MaskedManipulator: Versatile Whole-Body Manipulation](https://mlsanity.com/paper/2505.19086/) (2025)
- [Learning Human-Like Badminton Skills for Humanoid Robots](https://mlsanity.com/paper/2602.08370/) (2026)
- [InfBaGel: Human-Object-Scene Interaction Generation with Dynamic Perception and Iterative Refinement](https://mlsanity.com/paper/2604.04843/) (2026)
- [SUGAR: A Scalable Human-Video-Driven Generalizable Humanoid Loco-Manipulation Learning Framework](https://mlsanity.com/paper/2605.20373/) (2026)
- [Learning Smooth Time-Varying Linear Policies with an Action Jacobian Penalty](https://mlsanity.com/paper/2602.18312/) (2026)
- [Switch: Learning Agile Skills Switching for Humanoid Robots](https://mlsanity.com/paper/2604.14834/) (2026)
- [TopoCap: Learning Topology-Agnostic Motion Priors for Monocular Video-to-Animation](https://mlsanity.com/paper/2606.12153/) (2026)
- [MOCHI: Motion Enhancement of Collaborative Human-object Interactions](https://mlsanity.com/paper/2606.18243/) (2026)
- [Dynamic Full-body Motion Agent with Object Interaction via Blending Pre-trained Modular Controllers](https://mlsanity.com/paper/2605.11369/) (2026)
- [Policy-as-Data: Learning Generalizable HOI Diffusion Models from Simulated Physics](https://mlsanity.com/paper/2606.22806/) (2026)
- [MinInter: Minimizing Trajectory Interpolation During Data Augmentation for Imitation Learning](https://mlsanity.com/paper/2606.24078/) (2026)
- [ComplexMimic: Human-Scene Interaction Imitation in Complex 3D Environments](https://mlsanity.com/paper/2607.02034/) (2026)
- [WristMimic: Full-Body Humanoid Control with Wrist-Guided Manipulation](https://mlsanity.com/paper/2607.06438/) (2026)

## Abstract (arXiv)

We address a fundamental challenge in Reinforcement Learning from Interaction Demonstration (RLID): demonstration noise and coverage limitations. While existing data collection approaches provide valuable interaction demonstrations, they often yield sparse, disconnected, and noisy trajectories that fail to capture the full spectrum of possible skill variations and transitions. Our key insight is that despite noisy and sparse demonstrations, there exist infinite physically feasible trajectories that naturally bridge between demonstrated skills or emerge from their neighboring states, forming a continuous space of possible skill variations and transitions. Building upon this insight, we present two data augmentation techniques: a Stitched Trajectory Graph (STG) that discovers potential transitions between demonstration skills, and a State Transition Field (STF) that establishes unique connections for arbitrary states within the demonstration neighborhood. To enable effective RLID with augmented data, we develop an Adaptive Trajectory Sampling (ATS) strategy for dynamic curriculum generation and a historical encoding mechanism for memory-dependent skill learning. Our approach enables robust skill acquisition that significantly generalizes beyond the reference demonstrations. Extensive experiments across diverse interaction tasks demonstrate substantial improvements over state-of-the-art methods in terms of convergence stability, generalization capability, and recovery robustness.