---
title: "LLaSA — paper by Imran et al. (2024) — summary and ideas"
description: "Wearable IMU sensors are good at labeling activities, but current models cannot explain why a… 38 references, cited by 7 papers here."
authors: ["Sheikh Asif Imran", "Mohammad Nur Hossain Khan", "Subrata Biswas", "Bashima Islam"]
year: 2024
venue: "arXiv:2406.14498"
source: "https://arxiv.org/abs/2406.14498"
doi: "10.48550/arxiv.2406.14498"
arxiv: "2406.14498"
licence: "CC BY 4.0 (arXiv)"
licence_url: "http://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/imran-2024-llasa-sensor-aware-llm-natural/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/imran-2024-llasa-sensor-aware-llm-natural/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2024](https://md.research.elamur.ai/free/en/papers/2024/page/9/) / LLaSA: A Sensor-Aware LLM for Natural Language Reasoning of Human Activity from IMU Data

# LLaSA: A Sensor-Aware LLM for Natural Language Reasoning of Human Activity from IMU Data

**Authors:** Sheikh Asif Imran, Mohammad Nur Hossain Khan, Subrata Biswas and Bashima Islam · **Year:** 2024 · **Venue:** arXiv:2406.14498 · **Licence:** [CC BY 4.0 (arXiv)](http://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://arxiv.org/abs/2406.14498) · [PDF](https://arxiv.org/pdf/2406.14498) · [DOI 10.48550/arxiv.2406.14498](https://doi.org/10.48550/arxiv.2406.14498) · [arXiv 2406.14498](https://arxiv.org/abs/2406.14498) · [OpenAlex](https://openalex.org/W4399911924)

**Citations:** 4 citations (OpenAlex, 2026-10-14) · 7 in our library, of the 149,439 papers whose bibliographies we read

## Summary

Wearable IMU sensors are good at labeling activities, but current models cannot explain *why* a movement looks the way it does or answer free-form questions about it. This work targets sensor-based question answering (SQA) over raw accelerometer and gyroscope signals.

The authors built two resources from six public HAR datasets plus their own smartphone recordings collected in everyday settings. SensorCap pairs IMU windows with short captions and longer step-by-step narrations. OpenSQA adds question–answer pairs covering factual, descriptive, causal, diagnostic and predictive questions. Both were generated with GPT-4o-mini and checked by several LLM judges.

On this data they trained LLaSA, available in 7B and 13B sizes. It uses a frozen LIMU-BERT sensor encoder and a projection module that fits the signal into a 2048-token context. Model selection relied on GPT-based quality scoring rather than training loss.

LLaSA beats GPT-3.5, Vicuna and GPT-4o-mini on sensor-grounded QA and zero-shot activity recognition. This suggests that domain-specific alignment matters more than raw model size for reasoning over low-level sensor data.

## Key points

- **Datasets:**
  - SensorCap contains 35,960 IMU–caption pairs, with signals downsampled to 10 Hz.
  - OpenSQA contains 199,701 QA pairs, split into a 179,727-pair training fold and the Tune-OpenSQA fold (19,440 pairs from held-out PAMAP2).
  - A separate in-the-wild test set has 534 QA pairs, recorded by 10 users on varied phones.
- **Zero-shot classification:** LLaSA reaches F1 of 0.84 on HHAR and 0.83 on MotionSense. GPT-4o-mini scores ≤0.16 on these, and the best baseline scores 0.52. LLaSA does this with a 2k context, against 16k–128k for the baselines.
- **Tuning findings:** lower LoRA ranks gave better alignment. The best 13B configuration scored 72.17 overall across science, narration and reliability QA.
- **Evaluation:**
  - Human raters (n=22) preferred LLaSA's answers.
  - Data quality was judged by Deepseek-V3, LLaMA-3.1-8B and Gemma-12B, with most samples rated above 4.0 out of 5.
- **Limitations:**
  - All data was generated by a single LLM.
  - Human and benchmark evaluations were small.
  - The model confuses sitting with standing and sometimes hallucinates motion patterns.
  - Current model sizes are too large to run on-device.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

## Institutions

- [Worcester Polytechnic Institute](https://md.research.elamur.ai/free/en/universities/worcester-polytechnic-institute/) — University, Worcester, United States · 8 works · 380 citations

📚 Shelves (5): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 16,667 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 16,194 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 16,039 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 15,891 works · [👁️ Computer Vision and Pattern Recognition](https://md.research.elamur.ai/free/en/shelves/computer-science/computer-vision-and-pattern-recognition/) · 2,869 works

🧭 Topics (2): [🧠 Language models & pretraining](https://md.research.elamur.ai/free/en/topics/language-models/) · 3,320 works · [🧮 Reasoning](https://md.research.elamur.ai/free/en/topics/reasoning/) · 1,426 works

## Cite

Cite as: Sheikh Asif Imran, Mohammad Nur Hossain Khan, Subrata Biswas and Bashima Islam. 2024. LLaSA: A Sensor-Aware LLM for Natural Language Reasoning of Human Activity from IMU Data. arXiv:2406.14498.

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@misc{imran2024llasa,
  title = {{LLaSA: A Sensor-Aware LLM for Natural Language Reasoning of Human Activity from IMU Data}},
  author = {Sheikh Asif Imran, Mohammad Nur Hossain Khan, Subrata Biswas and Bashima Islam},
  year = {2024},
  doi = {10.48550/arxiv.2406.14498},
  eprint = {2406.14498},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.2406.14498}
}
```

- [← previous GAOKAO-MM: A Chinese Human-Level Benchmark for Multimodal Models Evaluation](https://md.research.elamur.ai/free/en/papers/zong-2024-gaokao-mm-chinese-human/)
- [next → Pragmatic Competence Evaluation of Large Language Models for the Korean Language](https://md.research.elamur.ai/free/en/papers/park-2024-pragmatic-competence-evaluation-large-language/)
- [↑ back to papers of 2024](https://md.research.elamur.ai/free/en/papers/2024/page/9/)

## References

38 references, 9 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. 2022. The IoT data deluge in industry and manufacturing. [https://www.newelectronics.co.uk/content/features/the-iot-data-deluge-in-industry-and-manufacturing](https://www.newelectronics.co.uk/content/features/the-iot-data-deluge-in-industry-and-manufacturing). Accessed: 2025-2-1. [↗](https://www.newelectronics.co.uk/content/features/the-iot-data-deluge-in-industry-and-manufacturing)
2. Berk Atil, Sarp Aykent, Alexa Chittams, Lisheng Fu, Rebecca J Passonneau, Evan Radcliffe, Guru Rajan Rajagopal, Adam Sloan, Tomasz Tudrej, Ferhan Ture, et al. 2024. Non-determinism of" deterministic" llm settings. arXiv preprint arXiv:2408.04667 (2024). — [arXiv:2408.04667](https://arxiv.org/abs/2408.04667) [find](https://scholar.google.com/scholar?q=Berk+Atil%2C+Sarp+Aykent%2C+Alexa+Chittams%2C+Lisheng+Fu%2C+Rebecca+J+Passonneau%2C+Evan+Radcliffe%2C+Guru+Rajan+Rajagopal%2C+Adam+Sloan%2C+Tomasz+Tudrej%2C+Ferhan+Ture%2C+et+al%2E+2024%2E+Non%2Ddeterminism+of%22+deterministic%22+)
3. Jinze Bai, Shuai Bai, Shusheng Yang, Shijie Wang, Sinan Tan, Peng Wang, Junyang Lin, Chang Zhou, and Jingren Zhou. 2023. [Qwen-vl: A frontier large vision-language model with versatile abilities](https://md.research.elamur.ai/free/en/papers/bai-2023-qwen-vl-versatile-vision-language/). arXiv preprint arXiv:2308.12966 (2023).
4. Subrata Biswas, Mohammad Nur Hossain Khan, and Bashima Islam. 2025. RAVEN: Representational Alignment of Audio, Video, Embedded Sensors and Natural Language. arXiv preprint (2025). [find](https://scholar.google.com/scholar?q=Subrata+Biswas%2C+Mohammad+Nur+Hossain+Khan%2C+and+Bashima+Islam%2E+2025%2E+RAVEN%3A+Representational+Alignment+of+Audio%2C+Video%2C+Embedded+Sensors+and+Natural+Language%2E+arXiv+preprint+%282025%29%2E)
5. Nimeesha Chan, Felix Parker, William Bennett, Tianyi Wu, Mung Yao Jia, James Fackler, and Kimia Ghobadi. 2024. Medtsllm: Leveraging llms for multimodal medical time series analysis. arXiv preprint arXiv:2408.07773 (2024). — [arXiv:2408.07773](https://arxiv.org/abs/2408.07773) [find](https://scholar.google.com/scholar?q=Nimeesha+Chan%2C+Felix+Parker%2C+William+Bennett%2C+Tianyi+Wu%2C+Mung+Yao+Jia%2C+James+Fackler%2C+and+Kimia+Ghobadi%2E+2024%2E+Medtsllm%3A+Leveraging+llms+for+multimodal+medical+time+series+analysis%2E+arXiv+preprint+arX)
6. Kaixuan Chen, Dalin Zhang, Lina Yao, Bin Guo, Zhiwen Yu, and Yunhao Liu. 2021. Deep learning for sensor-based human activity recognition: Overview, challenges, and opportunities. ACM Computing Surveys (CSUR) 54, 4 (2021), 1–40. [find](https://scholar.google.com/scholar?q=Kaixuan+Chen%2C+Dalin+Zhang%2C+Lina+Yao%2C+Bin+Guo%2C+Zhiwen+Yu%2C+and+Yunhao+Liu%2E+2021%2E+Deep+learning+for+sensor%2Dbased+human+activity+recognition%3A+Overview%2C+challenges%2C+and+opportunities%2E+ACM+Computing+Surveys)
7. Wenqiang Chen, Jiaxuan Cheng, Leyao Wang, Wei Zhao, and Wojciech Matusik. 2024. Sensor2Text: Enabling Natural Language Interactions for Daily Activity Tracking Using Wearable Sensors. Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies 8, 4 (2024), 1–26. [find](https://scholar.google.com/scholar?q=Wenqiang+Chen%2C+Jiaxuan+Cheng%2C+Leyao+Wang%2C+Wei+Zhao%2C+and+Wojciech+Matusik%2E+2024%2E+Sensor2Text%3A+Enabling+Natural+Language+Interactions+for+Daily+Activity+Tracking+Using+Wearable+Sensors%2E+Proceedings+of+t)
8. Kelvin Choi. 2022. Sensor Logger. [https://www.tszheichoi.com/sensorlogger](https://www.tszheichoi.com/sensorlogger). [Accessed 12-11-2024]. [↗](https://www.tszheichoi.com/sensorlogger)
9. Aakanksha Chowdhery, Sharan Narang, Jacob Devlin, Maarten Bosma, Gaurav Mishra, Adam Roberts, Paul Barham, Hyung Won Chung, Charles Sutton, Sebastian Gehrmann, et al. 2023. [Palm: Scaling language modeling with pathways](https://md.research.elamur.ai/free/en/papers/chowdhery-2022-palm-scaling-language-modeling-pathways/). Journal of Machine Learning Research 24, 240 (2023), 1–113.
10. Justin Cosentino, Anastasiya Belyaeva, Xin Liu, Nicholas A Furlotte, Zhun Yang, Chace Lee, Erik Schenck, Yojan Patel, Jian Cui, Logan Douglas Schneider, et al. 2024. Towards a Personal Health Large Language Model. arXiv preprint arXiv:2406.06474 (2024). — [arXiv:2406.06474](https://arxiv.org/abs/2406.06474) [find](https://scholar.google.com/scholar?q=Justin+Cosentino%2C+Anastasiya+Belyaeva%2C+Xin+Liu%2C+Nicholas+A+Furlotte%2C+Zhun+Yang%2C+Chace+Lee%2C+Erik+Schenck%2C+Yojan+Patel%2C+Jian+Cui%2C+Logan+Douglas+Schneider%2C+et+al%2E+2024%2E+Towards+a+Personal+Health+Large+La)
11. Michael Desmond, Zahra Ashktorab, Qian Pan, Casey Dugan, and James M Johnson. 2024. EvaluLLM: LLM assisted evaluation of generative outputs. In Companion Proceedings of the 29th International Conference on Intelligent User Interfaces. 30–32. [find](https://scholar.google.com/scholar?q=Michael+Desmond%2C+Zahra+Ashktorab%2C+Qian+Pan%2C+Casey+Dugan%2C+and+James+M+Johnson%2E+2024%2E+EvaluLLM%3A+LLM+assisted+evaluation+of+generative+outputs%2E+In+Companion+Proceedings+of+the+29th+International+Conferen)
12. Mingqi Gao, Xinyu Hu, Jie Ruan, Xiao Pu, and Xiaojun Wan. 2024. [Llm-based nlg evaluation: Current status and challenges](https://md.research.elamur.ai/free/en/papers/gao-2025-llm-based-nlg-evaluation-current/). arXiv preprint arXiv:2402.01383 (2024).
13. Hristijan Gjoreski, Mathias Ciliberto, Lin Wang, Francisco Javier Ordonez Morales, Sami Mekki, Stefan Valentin, and Daniel Roggen. 2018. The university of sussex-huawei locomotion and transportation dataset for multimodal analytics with mobile devices. IEEE Access 6 (2018), 42592–42604. [find](https://scholar.google.com/scholar?q=Hristijan+Gjoreski%2C+Mathias+Ciliberto%2C+Lin+Wang%2C+Francisco+Javier+Ordonez+Morales%2C+Sami+Mekki%2C+Stefan+Valentin%2C+and+Daniel+Roggen%2E+2018%2E+The+university+of+sussex%2Dhuawei+locomotion+and+transportation+d)
14. Edward J Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen. 2021. [Lora: Low-rank adaptation of large language models](https://md.research.elamur.ai/free/en/papers/hu-2021-lora-low-rank-adaptation-large/). arXiv preprint arXiv:2106.09685 (2021).
15. Yubin Kim, Xuhai Xu, Daniel McDuff, Cynthia Breazeal, and Hae Won Park. 2024. Health-llm: Large language models for health prediction via wearable sensor data. arXiv preprint arXiv:2401.06866 (2024). — [arXiv:2401.06866](https://arxiv.org/abs/2401.06866) [find](https://scholar.google.com/scholar?q=Yubin+Kim%2C+Xuhai+Xu%2C+Daniel+McDuff%2C+Cynthia+Breazeal%2C+and+Hae+Won+Park%2E+2024%2E+Health%2Dllm%3A+Large+language+models+for+health+prediction+via+wearable+sensor+data%2E+arXiv+preprint+arXiv%3A2401%2E06866+%282024%29%2E+)
16. Taku Kudo and John Richardson. 1808. [Sentencepiece: A simple and language independent subword tokenizer and detokenizer for neural text processing](https://md.research.elamur.ai/free/en/papers/kudo-2018-sentencepiece-simple-language-independent-subword/). arXiv 2018. arXiv preprint arXiv:1808.06226 (1808).
17. Ariel N Lee, Cole J Hunter, and Nataniel Ruiz. 2023. Platypus: Quick, cheap, and powerful refinement of llms. arXiv preprint arXiv:2308.07317 (2023). [find](https://scholar.google.com/scholar?q=Ariel+N+Lee%2C+Cole+J+Hunter%2C+and+Nataniel+Ruiz%2E+2023%2E+Platypus%3A+Quick%2C+cheap%2C+and+powerful+refinement+of+llms%2E+arXiv+preprint+arXiv%3A2308%2E07317+%282023%29%2E)
18. Zikang Leng, Amitrajit Bhattacharjee, Hrudhai Rajasekhar, Lizhe Zhang, Elizabeth Bruda, Hyeokhyen Kwon, and Thomas Plötz. 2024. Imugpt 2.0: Language-based cross modality transfer for sensor-based human activity recognition. Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Techn… [find](https://scholar.google.com/scholar?q=Zikang+Leng%2C+Amitrajit+Bhattacharjee%2C+Hrudhai+Rajasekhar%2C+Lizhe+Zhang%2C+Elizabeth+Bruda%2C+Hyeokhyen+Kwon%2C+and+Thomas+Pl%C3%B6tz%2E+2024%2E+Imugpt+2%2E0%3A+Language%2Dbased+cross+modality+transfer+for+sensor%2Dbased+huma)
19. Zechen Li, Shohreh Deldari, Linyao Chen, Hao Xue, and Flora D Salim. 2024. [Sensorllm: Aligning large language models with motion sensors for human activity recognition](https://md.research.elamur.ai/free/en/papers/li-2025-sensorllm-aligning-large-language-models/). arXiv preprint arXiv:2410.10624 (2024).
20. Chin-Yew Lin. 2004. [ROUGE: A Package for Automatic Evaluation of Summaries](https://md.research.elamur.ai/free/en/papers/lin-2004-rouge-package-automatic-evaluation-summaries/). In Text Summarization Branches Out. Association for Computational Linguistics, Barcelona, Spain, 74–81. https://aclanthology.org/W04-1013/ [↗](https://aclanthology.org/W04-1013/)
21. Shilong Liu, Hao Cheng, Haotian Liu, Hao Zhang, Feng Li, Tianhe Ren, Xueyan Zou, Jianwei Yang, Hang Su, Jun Zhu, et al. 2023. Llava-plus: Learning to use tools for creating multimodal agents. arXiv preprint arXiv:2311.05437 (2023). — [arXiv:2311.05437](https://arxiv.org/abs/2311.05437) [find](https://scholar.google.com/scholar?q=Shilong+Liu%2C+Hao+Cheng%2C+Haotian+Liu%2C+Hao+Zhang%2C+Feng+Li%2C+Tianhe+Ren%2C+Xueyan+Zou%2C+Jianwei+Yang%2C+Hang+Su%2C+Jun+Zhu%2C+et+al%2E+2023%2E+Llava%2Dplus%3A+Learning+to+use+tools+for+creating+multimodal+agents%2E+arXiv+pr)
22. Mohammad Malekzadeh, Richard G Clegg, Andrea Cavallaro, and Hamed Haddadi. 2019. Mobile sensor data anonymization. In Proceedings of the international conference on internet of things design and implementation. 49–58. [find](https://scholar.google.com/scholar?q=Mohammad+Malekzadeh%2C+Richard+G+Clegg%2C+Andrea+Cavallaro%2C+and+Hamed+Haddadi%2E+2019%2E+Mobile+sensor+data+anonymization%2E+In+Proceedings+of+the+international+conference+on+internet+of+things+design+and+imple)
23. Mike A Merrill, Akshay Paruchuri, Naghmeh Rezaei, Geza Kovacs, Javier Perez, Yun Liu, Erik Schenck, Nova Hammerquist, Jake Sunshine, Shyam Tailor, et al. 2024. Transforming wearable data into health insights using large language model agents. arXiv preprint arXiv:2406.06464 (2024). — [arXiv:2406.06464](https://arxiv.org/abs/2406.06464) [find](https://scholar.google.com/scholar?q=Mike+A+Merrill%2C+Akshay+Paruchuri%2C+Naghmeh+Rezaei%2C+Geza+Kovacs%2C+Javier+Perez%2C+Yun+Liu%2C+Erik+Schenck%2C+Nova+Hammerquist%2C+Jake+Sunshine%2C+Shyam+Tailor%2C+et+al%2E+2024%2E+Transforming+wearable+data+into+health+i)
24. Xiaomin Ouyang and Mani Srivastava. 2024. LLMSense: Harnessing LLMs for High-level Reasoning Over Spatiotemporal Sensor Traces. arXiv preprint arXiv:2403.19857 (2024). — [arXiv:2403.19857](https://arxiv.org/abs/2403.19857) [find](https://scholar.google.com/scholar?q=Xiaomin+Ouyang+and+Mani+Srivastava%2E+2024%2E+LLMSense%3A+Harnessing+LLMs+for+High%2Dlevel+Reasoning+Over+Spatiotemporal+Sensor+Traces%2E+arXiv+preprint+arXiv%3A2403%2E19857+%282024%29%2E+%E2%80%94+%5BarXiv%3A2403%2E19857%5D%28https%3A%2F%2Farx)
25. Kishore Papineni, Salim Roukos, Todd Ward, and Wei-Jing Zhu. 2002. [Bleu: a method for automatic evaluation of machine translation](https://md.research.elamur.ai/free/en/papers/papineni-2002-bleu-method-automatic-evaluation-machine/). In Proceedings of the 40th annual meeting of the Association for Computational Linguistics. 311–318.
26. Jorge-L Reyes-Ortiz, Luca Oneto, Albert Samà, Xavier Parra, and Davide Anguita. 2016. Transition-aware human activity recognition using smartphones. Neurocomputing 171 (2016), 754–767. [find](https://scholar.google.com/scholar?q=Jorge%2DL+Reyes%2DOrtiz%2C+Luca+Oneto%2C+Albert+Sam%C3%A0%2C+Xavier+Parra%2C+and+Davide+Anguita%2E+2016%2E+Transition%2Daware+human+activity+recognition+using+smartphones%2E+Neurocomputing+171+%282016%29%2C+754%E2%80%93767%2E)
27. Daniel Roggen, Alberto Calatroni, Mirco Rossi, Thomas Holleczek, Kilian Förster, Gerhard Tröster, Paul Lukowicz, David Bannach, Gerald Pirkl, Alois Ferscha, et al. 2010. Collecting complex activity datasets in highly rich networked sensor environments. In 2010 Seventh international conference on net… [find](https://scholar.google.com/scholar?q=Daniel+Roggen%2C+Alberto+Calatroni%2C+Mirco+Rossi%2C+Thomas+Holleczek%2C+Kilian+F%C3%B6rster%2C+Gerhard+Tr%C3%B6ster%2C+Paul+Lukowicz%2C+David+Bannach%2C+Gerald+Pirkl%2C+Alois+Ferscha%2C+et+al%2E+2010%2E+Collecting+complex+activity+da)
28. Thibault Sellam, Dipanjan Das, and Ankur P Parikh. 2020. [BLEURT: Learning robust metrics for text generation](https://md.research.elamur.ai/free/en/papers/sellam-2020-bleurt-learning-robust-metrics-text/). arXiv preprint arXiv:2004.04696 (2020).
29. Muhammad Shoaib, Stephan Bosch, Ozlem Durmaz Incel, Hans Scholten, and Paul JM Havinga. 2014. Fusion of smartphone motion sensors for physical activity recognition. Sensors 14, 6 (2014), 10146–10176. [find](https://scholar.google.com/scholar?q=Muhammad+Shoaib%2C+Stephan+Bosch%2C+Ozlem+Durmaz+Incel%2C+Hans+Scholten%2C+and+Paul+JM+Havinga%2E+2014%2E+Fusion+of+smartphone+motion+sensors+for+physical+activity+recognition%2E+Sensors+14%2C+6+%282014%29%2C+10146%E2%80%9310176%2E)
30. Satyajit Sinha. 2024. State of IoT 2024: Number of connected IoT devices growing 13% to 18.8 billion globally. [https://iot-analytics.com/number-connected-iot-devices/](https://iot-analytics.com/number-connected-iot-devices/). Accessed: 2025-2-1. [↗](https://iot-analytics.com/number-connected-iot-devices/)
31. Allan Stisen, Henrik Blunck, Sourav Bhattacharya, Thor Siiger Prentow, Mikkel Baun Kjærgaard, Anind Dey, Tobias Sonne, and Mads Møller Jensen. 2015. Smart devices are different: Assessing and mitigatingmobile sensing heterogeneities for activity recognition. In Proceedings of the 13th ACM conference… [find](https://scholar.google.com/scholar?q=Allan+Stisen%2C+Henrik+Blunck%2C+Sourav+Bhattacharya%2C+Thor+Siiger+Prentow%2C+Mikkel+Baun+Kj%C3%A6rgaard%2C+Anind+Dey%2C+Tobias+Sonne%2C+and+Mads+M%C3%B8ller+Jensen%2E+2015%2E+Smart+devices+are+different%3A+Assessing+and+mitigati)
32. Junke Wang, Dongdong Chen, Chong Luo, Xiyang Dai, Lu Yuan, Zuxuan Wu, and Yu-Gang Jiang. 2023. Chatvideo: A tracklet-centric multimodal and versatile video understanding system. arXiv preprint arXiv:2304.14407 (2023). — [arXiv:2304.14407](https://arxiv.org/abs/2304.14407) [find](https://scholar.google.com/scholar?q=Junke+Wang%2C+Dongdong+Chen%2C+Chong+Luo%2C+Xiyang+Dai%2C+Lu+Yuan%2C+Zuxuan+Wu%2C+and+Yu%2DGang+Jiang%2E+2023%2E+Chatvideo%3A+A+tracklet%2Dcentric+multimodal+and+versatile+video+understanding+system%2E+arXiv+preprint+arXiv%3A2)
33. Huatao Xu, Liying Han, Qirui Yang, Mo Li, and Mani Srivastava. 2024. Penetrative ai: Making llms comprehend the physical world. In Proceedings of the 25th International Workshop on Mobile Computing Systems and Applications. 1–7. [find](https://scholar.google.com/scholar?q=Huatao+Xu%2C+Liying+Han%2C+Qirui+Yang%2C+Mo+Li%2C+and+Mani+Srivastava%2E+2024%2E+Penetrative+ai%3A+Making+llms+comprehend+the+physical+world%2E+In+Proceedings+of+the+25th+International+Workshop+on+Mobile+Computing+Sy)
34. Huatao Xu, Pengfei Zhou, Rui Tan, Mo Li, and Guobin Shen. 2021. Limu-bert: Unleashing the potential of unlabeled data for imu sensing applications. In Proceedings of the 19th ACM Conference on Embedded Networked Sensor Systems. 220–233. [find](https://scholar.google.com/scholar?q=Huatao+Xu%2C+Pengfei+Zhou%2C+Rui+Tan%2C+Mo+Li%2C+and+Guobin+Shen%2E+2021%2E+Limu%2Dbert%3A+Unleashing+the+potential+of+unlabeled+data+for+imu+sensing+applications%2E+In+Proceedings+of+the+19th+ACM+Conference+on+Embedde)
35. Xuhai Xu, Bingsheng Yao, Yuanzhe Dong, Saadia Gabriel, Hong Yu, James Hendler, Marzyeh Ghassemi, Anind K Dey, and Dakuo Wang. 2024. Mental-llm: Leveraging large language models for mental health prediction via online text data. Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous T… [find](https://scholar.google.com/scholar?q=Xuhai+Xu%2C+Bingsheng+Yao%2C+Yuanzhe+Dong%2C+Saadia+Gabriel%2C+Hong+Yu%2C+James+Hendler%2C+Marzyeh+Ghassemi%2C+Anind+K+Dey%2C+and+Dakuo+Wang%2E+2024%2E+Mental%2Dllm%3A+Leveraging+large+language+models+for+mental+health+predi)
36. Hyungjun Yoon, Biniyam Aschalew Tolera, Taesik Gong, Kimin Lee, and Sung-Ju Lee. 2024. By My Eyes: Grounding Multimodal Large Language Models with Sensor Data via Visual Prompting. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, Yaser Al-Onaizan, Mohit Bans… [find](https://scholar.google.com/scholar?q=Hyungjun+Yoon%2C+Biniyam+Aschalew+Tolera%2C+Taesik+Gong%2C+Kimin+Lee%2C+and+Sung%2DJu+Lee%2E+2024%2E+By+My+Eyes%3A+Grounding+Multimodal+Large+Language+Models+with+Sensor+Data+via+Visual+Prompting%2E+In+Proceedings+of+t)
37. Xiyuan Zhang, Ranak Roy Chowdhury, Rajesh K Gupta, and Jingbo Shang. 2024. Large language models for time series: A survey. arXiv preprint arXiv:2402.01801 (2024). — [arXiv:2402.01801](https://arxiv.org/abs/2402.01801) [find](https://scholar.google.com/scholar?q=Xiyuan+Zhang%2C+Ranak+Roy+Chowdhury%2C+Rajesh+K+Gupta%2C+and+Jingbo+Shang%2E+2024%2E+Large+language+models+for+time+series%3A+A+survey%2E+arXiv+preprint+arXiv%3A2402%2E01801+%282024%29%2E+%E2%80%94+%5BarXiv%3A2402%2E01801%5D%28https%3A%2F%2Farxiv%2Eo)
38. Yuwei Zhang, Kumar Ayush, Siyuan Qiao, A Ali Heydari, Girish Narayanswamy, Maxwell A Xu, Ahmed A Metwally, Shawn Xu, Jake Garrison, Xuhai Xu, et al. 2025. SensorLM: Learning the Language of Wearable Sensors. arXiv preprint arXiv:2506.09108 (2025). — [arXiv:2506.09108](https://arxiv.org/abs/2506.09108) [find](https://scholar.google.com/scholar?q=Yuwei+Zhang%2C+Kumar+Ayush%2C+Siyuan+Qiao%2C+A+Ali+Heydari%2C+Girish+Narayanswamy%2C+Maxwell+A+Xu%2C+Ahmed+A+Metwally%2C+Shawn+Xu%2C+Jake+Garrison%2C+Xuhai+Xu%2C+et+al%2E+2025%2E+SensorLM%3A+Learning+the+Language+of+Wearable+S)

*Updated 30 October 2026*