# IMU-Bench: Benchmarking Wrist IMU Representations Across Kinematic, Semantic and Clinical Tasks

arXiv: https://arxiv.org/abs/2610.19481 · HTML: https://mlsanity.com/paper/2610.19481/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 48 of its 84 references that are in the map.
- First posted on arXiv 14 Oct 2026.

## Builds on (most cited first, 30 of 48)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [A Time Series is Worth 64 Words: Long-term Forecasting with Transformers](https://mlsanity.com/paper/2211.14730/) (2022)
- [Similarity of Neural Network Representations Revisited](https://mlsanity.com/paper/1905.00414/) (2019)
- [VGGT: Visual Geometry Grounded Transformer](https://mlsanity.com/paper/2503.11651/) (2025)
- [Understanding intermediate layers using linear classifier probes](https://mlsanity.com/paper/1610.01644/) (2016)
- [Designing and Interpreting Probes with Control Tasks](https://mlsanity.com/paper/1909.03368/) (2019)
- [Taskonomy: Disentangling Task Transfer Learning](https://mlsanity.com/paper/1804.08328/) (2018)
- [Beyond neural scaling laws: beating power law scaling via data pruning](https://mlsanity.com/paper/2206.14486/) (2022)
- [MOMENT: A Family of Open Time-series Foundation Models](https://mlsanity.com/paper/2402.03885/) (2024)
- [Which Tasks Should Be Learned Together in Multi-task Learning?](https://mlsanity.com/paper/1905.07553/) (2019)
- [Patch n' Pack: NaViT, a Vision Transformer for any Aspect Ratio and Resolution](https://mlsanity.com/paper/2307.06304/) (2023)
- [Motion-X: A Large-scale 3D Expressive Whole-body Human Motion Dataset](https://mlsanity.com/paper/2307.00818/) (2023)
- [Task2Vec: Task Embedding for Meta-Learning](https://mlsanity.com/paper/1902.03545/) (2019)
- [Physical Inertial Poser (PIP): Physics-aware Real-time Human Motion Tracking from Sparse Inertial Sensors](https://mlsanity.com/paper/2203.08528/) (2022)
- [Data Augmentation of Wearable Sensor Data for Parkinson's Disease Monitoring using Convolutional Neural Networks](https://mlsanity.com/paper/1706.00527/) (2017)
- [Exploring and Predicting Transferability across NLP Tasks](https://mlsanity.com/paper/2005.00770/) (2020)
- [Nymeria: A Massive Collection of Multimodal Egocentric Daily Motion in the Wild](https://mlsanity.com/paper/2406.09905/) (2024)
- [Multi-task Self-Supervised Learning for Human Activity Detection](https://mlsanity.com/paper/1907.11879/) (2019)
- [Large-scale Training of Foundation Models for Wearable Biosignals](https://mlsanity.com/paper/2312.05409/) (2023)
- [Transformer Inertial Poser: Real-time Human Motion Reconstruction from Sparse IMUs with Simultaneous Terrain Generation](https://mlsanity.com/paper/2203.15720/) (2022)
- [Self-supervised Learning for Human Activity Recognition Using 700,000 Person-days of Wearable Data](https://mlsanity.com/paper/2206.02909/) (2022)
- [RoNIN: Robust Neural Inertial Navigation in the Wild: Benchmark, Evaluations, and New Methods](https://mlsanity.com/paper/1905.12853/) (2019)
- [Assessing the State of Self-Supervised Human Activity Recognition using Wearables](https://mlsanity.com/paper/2202.12938/) (2022)
- [Scaling Wearable Foundation Models](https://mlsanity.com/paper/2410.13638/) (2024)
- [IONet: Learning to Cure the Curse of Drift in Inertial Odometry](https://mlsanity.com/paper/1802.02209/) (2018)
- [TLIO: Tight Learned Inertial Odometry](https://mlsanity.com/paper/2007.01867/) (2020)
- [IMUPoser: Full-Body Pose Estimation using IMUs in Phones, Watches, and Earbuds](https://mlsanity.com/paper/2304.12518/) (2023)
- [SensorLM: Learning the Language of Wearable Sensors](https://mlsanity.com/paper/2506.09108/) (2025)

## Abstract (arXiv)

Wrist-worn inertial measurement units (IMUs) provide a common sensing modality for tasks ranging from activity and gesture recognition to pose estimation, inertial odometry and clinical motor assessment. Existing wearable benchmarks evaluate performance across multiple tasks, but leave open how supervision for one task affects transfer to others and whether these tasks benefit from a shared representation. We present IMU-Bench, a benchmark of ten tasks from eight datasets for studying IMU multi-task performance and cross-task transfer. We compare six self-supervised methods pretrained on unlabeled recorded and simulated wrist motion, evaluating classification, pose and velocity estimation, and IMU-language retrieval. We find that our masked autoencoder, InrtlWrist, achieves the best or joint-best frozen-probe performance on nine of ten tasks, and fine-tuning improves on training from scratch on nine of ten tasks, including full-body pose, odometry and retrieval. Across transfer and joint-training experiments, supervision that benefits one target can reduce performance on others, with supervised sources transferring within kinematic estimation while, under frozen probing, none approaches the pretrained encoder on language retrieval, and joint-training gains that come at a cost to the paired task. These findings distinguish broad downstream usefulness from complementary task supervision. Code, task cards and a leaderboard are publicly available.