# CRAFT: Adapting VLA Models to Contact-rich Manipulation via Force-aware Curriculum Fine-tuning arXiv: https://arxiv.org/abs/2602.12532 · HTML: https://mlsanity.com/paper/2602.12532/ ## Key facts - Cited by 8 papers in the map (data through 14 Oct 2026). - Placed in Tactile, force-aware and humanoid VLA, one of 13 directions in Vision-language-action models. - Builds on 33 of its 42 references that are in the map. - First posted on arXiv 13 Feb 2026. ## Placed in - Vision-language-action models / [Tactile, force-aware and humanoid VLA](https://mlsanity.com/vla/tactile-force-aware-and-humanoid-vla/) ## Builds on (most cited first, 30 of 33) - [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021) - [RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control](https://mlsanity.com/paper/2307.15818/) (2023) - [OpenVLA: An Open-Source Vision-Language-Action Model](https://mlsanity.com/paper/2406.09246/) (2024) - [Diffusion Policy: Visuomotor Policy Learning via Action Diffusion](https://mlsanity.com/paper/2303.04137/) (2023) - [π0: A Vision-Language-Action Flow Model for General Robot Control](https://mlsanity.com/paper/2410.24164/) (2024) - [Do As I Can, Not As I Say: Grounding Language in Robotic Affordances](https://mlsanity.com/paper/2204.01691/) (2022) - [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) - [Octo: An Open-Source Generalist Robot Policy](https://mlsanity.com/paper/2405.12213/) (2024) - [Deep Variational Information Bottleneck](https://mlsanity.com/paper/1612.00410/) (2016) - [VoxPoser: Composable 3D Value Maps for Robotic Manipulation with Language Models](https://mlsanity.com/paper/2307.05973/) (2023) - [CogACT: A Foundational Vision-Language-Action Model for Synergizing Cognition and Action in Robotic Manipulation](https://mlsanity.com/paper/2411.19650/) (2024) - [DexVLA: Vision-Language Model with Plug-In Diffusion Expert for General Robot Control](https://mlsanity.com/paper/2502.05855/) (2025) - [Graph Information Bottleneck](https://mlsanity.com/paper/2010.12811/) (2020) - [DeeR-VLA: Dynamic Inference of Multimodal Large Language Models for Efficient Robot Execution](https://mlsanity.com/paper/2411.02359/) (2024) - [Restricting the Flow: Information Bottlenecks for Attribution](https://mlsanity.com/paper/2001.00396/) (2020) - [Reactive Diffusion Policy: Slow-Fast Visual-Tactile Policy Learning for Contact-Rich Manipulation](https://mlsanity.com/paper/2503.02881/) (2025) - [ForceVLA: Enhancing VLA Models with a Force-aware MoE for Contact-rich Manipulation](https://mlsanity.com/paper/2505.22159/) (2025) - [OccLLaMA: An Occupancy-Language-Action Generative World Model for Autonomous Driving](https://mlsanity.com/paper/2409.03272/) (2024) - [VLA-Touch: Enhancing Vision-Language-Action Models with Dual-Level Tactile Feedback](https://mlsanity.com/paper/2507.17294/) (2025) - [FACTR: Force-Attending Curriculum Training for Contact-Rich Policy Learning](https://mlsanity.com/paper/2502.17432/) (2025) - [Learning Optimal Representations with the Decodable Information Bottleneck](https://mlsanity.com/paper/2009.12789/) (2020) - [ReVLA: Reverting Visual Domain Limitation of Robotic Foundation Models](https://mlsanity.com/paper/2409.15250/) (2024) - [EdgeVLA: Efficient Vision-Language-Action Models](https://mlsanity.com/paper/2507.14049/) (2025) - [SwitchVLA: Execution-Aware Task Switching for Vision-Language-Action Models](https://mlsanity.com/paper/2506.03574/) (2025) - [Dynamic Multimodal Information Bottleneck for Multimodality Classification](https://mlsanity.com/paper/2311.01066/) (2023) - [Learning from Imperfect Demonstrations with Self-Supervision for Robotic Manipulation](https://mlsanity.com/paper/2401.08957/) (2024) - [FreqPolicy: Efficient Flow-based Visuomotor Policy via Frequency Consistency](https://mlsanity.com/paper/2506.08822/) (2025) - [Towards Forceful Robotic Foundation Models: a Literature Survey](https://mlsanity.com/paper/2504.11827/) (2025) - [Learning Sparse Latent Representations with the Deep Copula Information Bottleneck](https://mlsanity.com/paper/1804.06216/) (2018) ## Built on by (most cited first, 8 of 8) - [Human-Centric Transferable Tactile Pre-Training for Dexterous Robotic Manipulation](https://mlsanity.com/paper/2607.01067/) (2026) - [Demystifying When and Why VLAs Fail in Contact-Rich Tasks and How to Fix Them](https://mlsanity.com/paper/2608.01402/) (2026) - [Learning Physical Interaction: A Survey of Tactile- and Force-aware Robot Learning](https://mlsanity.com/paper/2608.07558/) (2026) - [ForceU-VLA: A Force-Aware Vision-Language-Action Model for Embodied Ultrasound Scanning](https://mlsanity.com/paper/2608.15009/) (2026) - [US-VLA: An Ultrasound Vision-Language-Action Model for Embodied Abdomina](https://mlsanity.com/paper/2608.16074/) (2026) - [ForceDelta-VLA: Distilling Force-Conditioned ActionCorrections for Contact-Rich Manipulation](https://mlsanity.com/paper/2609.18242/) (2026) - [VT-Bridge: Bridging Pretrained Foundation VLAs to VTLAs via Lightweight Residual Adaptation](https://mlsanity.com/paper/2609.22606/) (2026) - [ForceRFT: Refining VLA Actions through Force-Guided Residual Reinforcement Learning](https://mlsanity.com/paper/2609.22840/) (2026) ## Abstract (arXiv) Vision-Language-Action (VLA) models have shown a strong capability in enabling robots to execute general instructions, yet they struggle with contact-rich manipulation tasks, where success requires precise alignment, stable contact maintenance, and effective handling of deformable objects. A fundamental challenge arises from the imbalance between high-entropy vision and language inputs and low-entropy but critical force signals, which often leads to over-reliance on perception and unstable control. To address this, we introduce CRAFT, a force-aware curriculum fine-tuning framework that integrates a variational information bottleneck module to regulate vision and language embeddings during early training. This curriculum strategy encourages the model to prioritize force signals initially, before progressively restoring access to the full multimodal information. To enable force-aware learning, we further design a homologous leader-follower teleoperation system that collects synchronized vision, language, and force data across diverse contact-rich tasks. Real-world experiments demonstrate that CRAFT consistently improves task success, generalizes to unseen objects and novel task variations, and adapts effectively across diverse VLA architectures, enabling robust and generalizable contact-rich manipulation.