# RadarLLM: Empowering Large Language Models to Understand Human Motion from Millimeter-Wave Point Cloud Sequence

arXiv: https://arxiv.org/abs/2504.09862 · HTML: https://mlsanity.com/paper/2504.09862/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 44 of its 79 references that are in the map.
- First posted on arXiv 14 Apr 2025.

## Builds on (most cited first, 30 of 44)

- [BERTScore: Evaluating Text Generation with BERT](https://mlsanity.com/paper/1904.09675/) (2019)
- [SimCSE: Simple Contrastive Learning of Sentence Embeddings](https://mlsanity.com/paper/2104.08821/) (2021)
- [CIDEr: Consensus-based Image Description Evaluation](https://mlsanity.com/paper/1411.5726/) (2014)
- [Video-LLaVA: Learning United Visual Representation by Alignment Before Projection](https://mlsanity.com/paper/2311.10122/) (2023)
- [AMASS: Archive of Motion Capture as Surface Shapes](https://mlsanity.com/paper/1904.03278/) (2019)
- [MVBench: A Comprehensive Multi-modal Video Understanding Benchmark](https://mlsanity.com/paper/2311.17005/) (2023)
- [Video-ChatGPT: Towards Detailed Video Understanding via Large Vision and Language Models](https://mlsanity.com/paper/2306.05424/) (2023)
- [VideoLLaMA 2: Advancing Spatial-Temporal Modeling and Audio Understanding in Video-LLMs](https://mlsanity.com/paper/2406.07476/) (2024)
- [VideoBERT: A Joint Model for Video and Language Representation Learning](https://mlsanity.com/paper/1904.01766/) (2019)
- [MotionGPT: Human Motion as a Foreign Language](https://mlsanity.com/paper/2306.14795/) (2023)
- [TM2T: Stochastic and Tokenized Modeling for the Reciprocal Generation of 3D Human Motions and Texts](https://mlsanity.com/paper/2207.01696/) (2022)
- [PointLLM: Empowering Large Language Models to Understand Point Clouds](https://mlsanity.com/paper/2308.16911/) (2023)
- [AudioGPT: Understanding and Generating Speech, Music, Sound, and Talking Head](https://mlsanity.com/paper/2304.12995/) (2023)
- [VTimeLLM: Empower LLM to Grasp Video Moments](https://mlsanity.com/paper/2311.18445/) (2023)
- [WHAM: Reconstructing World-grounded Humans with Accurate 3D Motion](https://mlsanity.com/paper/2312.07531/) (2023)
- [AlphaPose: Whole-Body Regional Multi-Person Pose Estimation and Tracking in Real-Time](https://mlsanity.com/paper/2211.03375/) (2022)
- [MotionBERT: A Unified Perspective on Learning Human Motion Representations](https://mlsanity.com/paper/2210.06551/) (2022)
- [LiDAR-LLM: Exploring the Potential of Large Language Models for 3D LiDAR Understanding](https://mlsanity.com/paper/2312.14074/) (2023)
- [Fast WordPiece Tokenization](https://mlsanity.com/paper/2012.15524/) (2020)
- [CRN: Camera Radar Net for Accurate, Robust, Efficient 3D Perception](https://mlsanity.com/paper/2304.00670/) (2023)
- [RCBEVDet: Radar-camera Fusion in Bird's Eye View for 3D Object Detection](https://mlsanity.com/paper/2403.16440/) (2024)
- [MM-Fi: Multi-Modal Non-Intrusive 4D Human Dataset for Versatile Wireless Sensing](https://mlsanity.com/paper/2305.10345/) (2023)
- [SVFormer: Semi-supervised Video Transformer for Action Recognition](https://mlsanity.com/paper/2211.13222/) (2022)
- [AvatarGPT: All-in-One Framework for Motion Understanding, Planning, Generation and Beyond](https://mlsanity.com/paper/2311.16468/) (2023)
- [RadarNet: Exploiting Radar for Robust Perception of Dynamic Objects](https://mlsanity.com/paper/2007.14366/) (2020)
- [RF-Diffusion: Radio Signal Generation via Time-Frequency Diffusion](https://mlsanity.com/paper/2404.09140/) (2024)
- [mRI: Multi-modal 3D Human Pose Estimation Dataset using mmWave, RGB-D, and Inertial Sensors](https://mlsanity.com/paper/2210.08394/) (2022)
- [SensorLLM: Aligning Large Language Models with Motion Sensors for Human Activity Recognition](https://mlsanity.com/paper/2410.10624/) (2024)
- [LLMs are Good Action Recognizers](https://mlsanity.com/paper/2404.00532/) (2024)
- [mmBody Benchmark: 3D Body Reconstruction Dataset and Analysis for Millimeter Wave Radar](https://mlsanity.com/paper/2209.05070/) (2022)

## Built on by (most cited first, 7 of 7)

- [RLM: A Vision-Language Model Approach for Radar Scene Understanding](https://mlsanity.com/paper/2511.21105/) (2025)
- [GeoMotionGPT: Geometry-Aligned Motion Understanding with Large Language Models](https://mlsanity.com/paper/2601.07632/) (2026)
- [Generative Latent Alignment for Interpretable Radar Based Occupancy Detection in Ambient Assisted Living](https://mlsanity.com/paper/2601.19853/) (2026)
- [RAGent: Physics-Aware Agentic Reasoning for Training-Free mmWave Human Activity Recognition](https://mlsanity.com/paper/2603.27571/) (2026)
- [DAP: Doppler-aware Point Network for Heterogeneous mmWave Action Recognition](https://mlsanity.com/paper/2605.09604/) (2026)
- [mmSimPrior: Learning Simulation Priors for Data-Efficient and Generalizable Real-World Radar-based Human Motion Reconstruction](https://mlsanity.com/paper/2607.22973/) (2026)
- [Can Language Models Understand mmWave Data? Benchmarking Large Language Models for mmWave Radar-Based Human Understanding](https://mlsanity.com/paper/2608.14179/) (2026)

## Abstract (arXiv)

Millimeter-wave radar offers a privacy-preserving and environment-robust alternative to vision-based sensing, enabling human motion analysis in challenging conditions such as low light, occlusions, rain, or smoke. However, its sparse point clouds pose significant challenges for semantic understanding. We present RadarLLM, the first framework that leverages large language models (LLMs) for human motion understanding from radar signals. RadarLLM introduces two key innovations: (1) a motion-guided radar tokenizer based on our Aggregate VQ-VAE architecture, integrating deformable body templates and masked trajectory modeling to convert spatial-temporal radar sequences into compact semantic tokens; and (2) a radar-aware language model that establishes cross-modal alignment between radar and text in a shared embedding space. To overcome the scarcity of paired radar-text data, we generate a realistic radar-text dataset from motion-text datasets with a physics-aware synthesis pipeline. Extensive experiments on both synthetic and real-world benchmarks show that RadarLLM achieves state-of-the-art performance, enabling robust and interpretable motion understanding under privacy and visibility constraints, even in adverse environments. This paper has been accepted for presentation at AAAI 2026. This is an extended version with supplementary materials.