# CSI-JEPA: Towards Foundation Representations for Ubiquitous Sensing with Minimal Supervision

arXiv: https://arxiv.org/abs/2605.14171 · HTML: https://mlsanity.com/paper/2605.14171/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 38 references that are in the map.
- First posted on arXiv 13 May 2026.

## Builds on (most cited first, 22 of 22)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecture](https://mlsanity.com/paper/2301.08243/) (2023)
- [VL-JEPA: Joint Embedding Predictive Architecture for Vision-language](https://mlsanity.com/paper/2512.10942/) (2025)
- [SenseFi: A Library and Benchmark on Deep-Learning-Empowered WiFi Human Sensing](https://mlsanity.com/paper/2207.07859/) (2022)
- [AutoFi: Towards Automatic WiFi Human Sensing via Geometric Self-Supervised Learning](https://mlsanity.com/paper/2205.01629/) (2022)
- [LLM4WM: Adapting LLM for Wireless Multi-Tasking](https://mlsanity.com/paper/2501.12983/) (2025)
- [Denoising Diffusion Probabilistic Model for Radio Map Estimation in Generative Wireless Networks](https://mlsanity.com/paper/2501.06604/) (2025)
- [An Overview on IEEE 802.11bf: WLAN Sensing](https://mlsanity.com/paper/2310.17661/) (2023)
- [A Tutorial-cum-Survey on Self-Supervised Learning for Wi-Fi Sensing: Trends, Challenges, and Outlook](https://mlsanity.com/paper/2506.12052/) (2025)
- [CSI-Bench: A Large-Scale In-the-Wild Dataset for Multi-task WiFi Sensing](https://mlsanity.com/paper/2505.21866/) (2025)
- [WirelessJEPA: A Multi-Antenna Foundation Model using Spatio-temporal Wireless Latent Predictions](https://mlsanity.com/paper/2601.20190/) (2026)
- [Learning Latent Wireless Dynamics from Channel State Information](https://mlsanity.com/paper/2409.10045/) (2024)
- [CSI-MAE: A Masked Autoencoder-based Channel Foundation Model](https://mlsanity.com/paper/2601.03789/) (2026)
- [Toward Integrated Sensing and Communications in IEEE 802.11bf Wi-Fi Networks](https://mlsanity.com/paper/2212.13930/) (2022)
- [AM-FM: A Foundation Model for Ambient Intelligence Through WiFi](https://mlsanity.com/paper/2602.11200/) (2026)
- [Inferring Person-to-person Proximity Using WiFi Signals](https://mlsanity.com/paper/1610.04730/) (2016)
- [Contextual Combinatorial Beam Management via Online Probing for Multiple Access mmWave Wireless Networks](https://mlsanity.com/paper/2412.10385/) (2024)
- [Structured Latent Dynamics in Wireless CSI via Homomorphic World Models](https://mlsanity.com/paper/2603.20048/) (2026)
- [JEPA-MSAC: A Joint-Embedding Predictive Architecture for Multimodal Sensing-Assisted Communications](https://mlsanity.com/paper/2603.29796/) (2026)
- [Sensing Performance of the IEEE 802.11bf Protocol and Its Impact on Data Communication](https://mlsanity.com/paper/2403.19825/) (2024)
- [MMSense: Adapting Vision-based Foundation Model for Multi-task Multi-modal Wireless Sensing](https://mlsanity.com/paper/2511.12305/) (2025)

## Built on by (most cited first, 3 of 3)

- [Wireless Foundation Models: State-of-the-Art and Open Challenges](https://mlsanity.com/paper/2609.04707/) (2026)
- [Trajectory-Guided Tokenization of Complex CSI for Wi-Fi Sensing](https://mlsanity.com/paper/2610.06288/) (2026)
- [UniCSI Towards a Universal Wi-Fi CSI Encoder for Ubiquitous Human Sensing](https://mlsanity.com/paper/2610.09559/) (2026)

## Abstract (arXiv)

Channel state information (CSI) provides a widely available sensing modality for human and environment perception, but existing CSI sensing models usually rely on task-specific supervised training and require substantial labeled data for each task, device, user, or environment. This limits their scalability in practical deployments where unlabeled CSI is abundant but labeled data is costly to collect. In this paper, we present CSI-JEPA, a self-supervised predictive representation learning framework for label-efficient, multi-task Wi-Fi sensing. CSI-JEPA learns reusable temporal-spectral representations from unlabeled CSI samples by predicting latent features of masked channel regions from visible context. To better match the physical structure of CSI, CSI-JEPA tokenizes channel-response amplitude windows along the time and subcarrier dimensions. It then introduces a channel variation-aware masking strategy that samples predictive targets from regions with stronger local temporal and subcarrier-domain variations. After pretraining, the encoder is frozen and used as a backbone, with lightweight task-specific adapters added for downstream sensing tasks. We evaluate CSI-JEPA on seven real-world Wi-Fi sensing tasks spanning diverse objectives and deployment settings. The results show that CSI-JEPA improves downstream sensing performance over competitive baselines, achieving up to 10.64 percentage points mean accuracy gain over state-of-the-art supervised Transformer and matched-budget label savings of up to 98.0%.