# Mitigating the Human-Robot Domain Discrepancy in Visual Pre-training for Robotic Manipulation

arXiv: https://arxiv.org/abs/2406.14235 · HTML: https://mlsanity.com/paper/2406.14235/

## Key facts

- Cited by 23 papers in the map (data through 14 Oct 2026).
- Builds on 77 of its 93 references that are in the map.
- First posted on arXiv 20 Jun 2024.

## Builds on (most cited first, 30 of 77)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Momentum Contrast for Unsupervised Visual Representation Learning](https://mlsanity.com/paper/1911.05722/) (2019)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter](https://mlsanity.com/paper/1910.01108/) (2019)
- [Playing Atari with Deep Reinforcement Learning](https://mlsanity.com/paper/1312.5602/) (2013)
- [Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset](https://mlsanity.com/paper/1705.07750/) (2017)
- [RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control](https://mlsanity.com/paper/2307.15818/) (2023)
- [Domain-Adversarial Training of Neural Networks](https://mlsanity.com/paper/1505.07818/) (2015)
- [Diffusion Policy: Visuomotor Policy Learning via Action Diffusion](https://mlsanity.com/paper/2303.04137/) (2023)
- [Non-local Neural Networks](https://mlsanity.com/paper/1711.07971/) (2017)
- [RT-1: Robotics Transformer for Real-World Control at Scale](https://mlsanity.com/paper/2212.06817/) (2022)
- [Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware](https://mlsanity.com/paper/2304.13705/) (2023)
- [Ego4D: Around the World in 3,000 Hours of Egocentric Video](https://mlsanity.com/paper/2110.07058/) (2021)
- [VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Training](https://mlsanity.com/paper/2203.12602/) (2022)
- [Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learning](https://mlsanity.com/paper/1910.10897/) (2019)
- [The "something something" video database for learning and evaluating visual common sense](https://mlsanity.com/paper/1706.04261/) (2017)
- [Open X-Embodiment: Robotic Learning Datasets and RT-X Models](https://mlsanity.com/paper/2310.08864/) (2023)
- [QT-Opt: Scalable Deep Reinforcement Learning for Vision-Based Robotic Manipulation](https://mlsanity.com/paper/1806.10293/) (2018)
- [Learning Complex Dexterous Manipulation with Deep Reinforcement Learning and Demonstrations](https://mlsanity.com/paper/1709.10087/) (2017)
- [R3M: A Universal Visual Representation for Robot Manipulation](https://mlsanity.com/paper/2203.12601/) (2022)
- [RLBench: The Robot Learning Benchmark & Learning Environment](https://mlsanity.com/paper/1909.12271/) (2019)
- [CLIPort: What and Where Pathways for Robotic Manipulation](https://mlsanity.com/paper/2109.12098/) (2021)
- [CALVIN: A Benchmark for Language-Conditioned Policy Learning for Long-Horizon Robot Manipulation Tasks](https://mlsanity.com/paper/2112.03227/) (2021)
- [Perceiver-Actor: A Multi-Task Transformer for Robotic Manipulation](https://mlsanity.com/paper/2209.05451/) (2022)
- [VideoMAE V2: Scaling Video Masked Autoencoders with Dual Masking](https://mlsanity.com/paper/2303.16727/) (2023)
- [Masked Autoencoders As Spatiotemporal Learners](https://mlsanity.com/paper/2205.09113/) (2022)
- [VIP: Towards Universal Visual Reward and Representation via Value-Implicit Pre-Training](https://mlsanity.com/paper/2210.00030/) (2022)

## Built on by (most cited first, 23 of 23)

- [Large VLM-based Vision-Language-Action Models for Robotic Manipulation: A Survey](https://mlsanity.com/paper/2508.13073/) (2025)
- [Robotic Manipulation by Imitating Generated Videos Without Physical Demonstrations](https://mlsanity.com/paper/2507.00990/) (2025)
- [Native Video-Action Pretraining for Generalizable Robot Control](https://mlsanity.com/paper/2607.08639/) (2026)
- [Exploring the Limits of Vision-Language-Action Manipulations in Cross-task Generalization](https://mlsanity.com/paper/2505.15660/) (2025)
- [An Economic Framework for 6-DoF Grasp Detection](https://mlsanity.com/paper/2407.08366/) (2024)
- [GLOVER++: Unleashing the Potential of Affordance Learning from Human Behaviors for Robotic Manipulation](https://mlsanity.com/paper/2505.11865/) (2025)
- [Zero-WAM: In-Context World-Action Modeling from Human Videos for Open-Ended Task Generalization](https://mlsanity.com/paper/2608.26103/) (2026)
- [Rethinking Bimanual Robotic Manipulation: Learning with Decoupled Interaction Framework](https://mlsanity.com/paper/2503.09186/) (2025)
- [Modeling Multiple Normal Action Representations for Error Detection in Procedural Tasks](https://mlsanity.com/paper/2503.22405/) (2025)
- [AnyTouch 2: General Optical Tactile Representation Learning For Dynamic Tactile Perception](https://mlsanity.com/paper/2602.09617/) (2026)
- [VLANeXt: Recipes for Building Strong VLA Models](https://mlsanity.com/paper/2602.18532/) (2026)
- [Panoptic Captioning: An Equivalence Bridge for Image and Text](https://mlsanity.com/paper/2505.16334/) (2025)
- [EgoTraj-Bench: Towards Robust Trajectory Prediction Under Ego-view Noisy Observations](https://mlsanity.com/paper/2510.00405/) (2025)
- [HARP-VLA: Human-Robot Aligned Representation Learning for Vision-Language-Action Model](https://mlsanity.com/paper/2605.31234/) (2026)
- [Provable Ordering and Continuity in Vision-Language Pretraining for Generalizable Embodied Agents](https://mlsanity.com/paper/2502.01218/) (2025)
- [MiVLA: Towards Generalizable Vision-Language-Action Model with Human-Robot Mutual Imitation Pre-training](https://mlsanity.com/paper/2512.15411/) (2025)
- [From Imagined Futures to Executable Actions: Mixture of Latent Actions for Robot Manipulation](https://mlsanity.com/paper/2605.12167/) (2026)
- [A Closed-Loop Multi-Agent Framework for Robust Multi-Robot Manipulation](https://mlsanity.com/paper/2607.06990/) (2026)
- [Structural Action Transformer for 3D Dexterous Manipulation](https://mlsanity.com/paper/2603.03960/) (2026)
- [Decompose and Recompose: Reasoning New Skills from Existing Abilities for Cross-Task Robotic Manipulation](https://mlsanity.com/paper/2605.01448/) (2026)
- [Holo-Captioning: Toward the Text Equivalent of 3D Scenes](https://mlsanity.com/paper/2607.02908/) (2026)
- [Exo2EgoPose: Leveraging Exocentric Demonstrations for Vision-Language guided Egocentric 3D Hand Pose Forecasting](https://mlsanity.com/paper/2607.15890/) (2026)
- [Frame-Level Temporal Alignment for Human-to-Robot Visual Adaptation](https://mlsanity.com/paper/2610.04372/) (2026)

## Abstract (arXiv)

Learning generalizable visual representations across different embodied environments is essential for effective robotic manipulation in real-world scenarios. However, the limited scale and diversity of robot demonstration data pose a significant challenge. Recent research has explored leveraging large-scale human activity data for pre-training, but the substantial morphological differences between humans and robots introduce a significant human-robot domain discrepancy, hindering the generalization of these models to downstream manipulation tasks. To overcome this, we propose a novel adaptation paradigm that leverages readily available paired human-robot video data to bridge the domain gap. Our method employs a human-robot contrastive alignment loss to align the semantics of human and robot videos, adapting pre-trained models to the robot domain in a parameter-efficient manner. Experiments on 20 simulated tasks across two different benchmarks and five real-world tasks demonstrate significant improvements. These results span both single-task and language-conditioned multi-task settings, evaluated using two different pre-trained models. Compared to existing pre-trained models, our adaptation method improves the average success rate by over 7% across multiple tasks on both simulated benchmarks and real-world evaluations.