Mehran Kazemi, Bahare Fatemi, Hritik Bansal, John Palowitch, Chrysovalantis Anastasiou, Sanket Vaibhav Mehta, Lalit K Jain, Virginia Aglietti, Disha Jindal, Peter Chen, Nishanth Dikkala, Gladys Tyen, Xin Liu, Uri Shalit, Silvia Chiappa, Kate Olszewska, Yi Tay, Vinh Q. Tran, Quoc V Le, Orhan Firat. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

While NLP research has made strides in conversational tasks, many approaches focus on single-turn responses with well-defined objectives or evaluation criteria. In contrast, coaching presents unique challenges with initially undefined goals that evolve through multi-turn interactions, subjective evaluation criteria, mixed-initiative dialogue. In this work, we describe and implement five multi-turn coaching agents that exhibit distinct conversational styles, and evaluate them through a user study, collecting first-person feedback on 155 conversations. We find that users highly value core functionality, and that stylistic components in absence of core components are viewed negatively. By comparing user feedback with third-person evaluations from health experts and an LM, we reveal significant misalignment across evaluation approaches. Our findings provide insights into design and evaluation of conversational coaching agents and contribute toward improving human-centered NLP applications.

Remote photoplethysmography (rPPG) enables noncontact, continuous monitoring of physiological signals and offers a practical alternative to traditional health sensing methods. Although rPPG is promising for daily health monitoring, its application in long-term personal care scenarios-such as mirrorfacing routines in high-altitude environments-remains challenging due to ambient lighting variations, frequent occlusions from hand movements, and dynamic facial postures. To address these challenges, we present the Long-term Altitude Daily Health (LADH) dataset, the first long-term rPPG dataset containing 240 synchronized RGB and infrared (IR) facial videos from 21 participants across five common personal care scenarios, along with ground-truth PPG, respiration, and blood oxygen signals. Our experiments demonstrate that combining RGB and IR video inputs improves the accuracy and robustness of non-contact physiological monitoring, achieving a mean absolute error (MAE) of 4.99 BPM in heart rate estimation. Furthermore, we find that multi-task learning enhances performance across multiple physiological indicators simultaneously. Dataset and code are open at https://github.com/McJackTang/FusionVitals.

Prosociality has been well-documented to positively impact mental, social, and physical well-being.However, existing studies of interventions for promoting prosociality have limitations such as

Wearable sensors have become ubiquitous thanks to a variety of health tracking features. The resulting continuous and longitudinal measurements from everyday life generate large volumes of data. However, making sense of these observations for scientific and actionable insights is non-trivial. Inspired by the empirical success of generative modeling, where large neural networks learn powerful representations from vast amounts of text, image, video, or audio data, we investigate the scaling properties of wearable sensor foundation models across compute, data, and model size. Using a dataset of up to 40 million hours of in-situ heart rate, heart rate variability, accelerometer, electrodermal activity, skin temperature, and altimeter per-minute data from over 165,000 people, we create LSM, a multimodal foundation model built on the largest wearable-signals dataset with the most extensive range of sensor modalities to date. Our results establish the scaling laws of LSM for tasks such as imputation, interpolation and extrapolation across both time and sensor modalities. Moreover, we highlight how LSM enables sample-efficient downstream learning for tasks including exercise and activity recognition.

Vocal health plays a crucial role in peoples' lives, significantly impacting their communicative abilities and interactions. However, despite the global prevalence of voice disorders, many lack access to convenient diagnosis and treatment. This paper introduces VocalAgent, an audio large language model (LLM) to address these challenges through vocal health diagnosis. We leverage Qwen-Audio-Chat fine-tuned on three datasets collected in-situ from hospital patients, and present a multifaceted evaluation framework encompassing a safety assessment to mitigate diagnostic biases, cross-lingual performance analysis, and modality ablation studies. VocalAgent demonstrates superior accuracy on voice disorder classification compared to state-of-the-art baselines. Its LLM-based method offers a scalable solution for broader adoption of health diagnostics, while underscoring the importance of ethical and technical validation.

Language models (LMs) are increasingly being deployed to perform autonomous data analyses. However, their data awareness—the ability to recognize, reason over, and appropriately handle data artifacts such as missing values, outliers, and logical inconsistencies—remains underexplored. These artifacts are especially common in real-world tabular data and, if mishandled, can significantly compromise the validity of analytical conclusions. To address this gap, we present RADAR, a benchmark for systematically evaluating data-aware reasoning on tabular data. We develop a framework to simulate data artifacts via programmatic perturbations to enable targeted evaluation of model behavior. RADAR comprises 2,980 table-query pairs, grounded in real-world data spanning 9 domains and 5 data artifact types. In addition to evaluating artifact handling, RADAR systematically varies table size to study how reasoning performance holds when increasing table size. Our evaluation reveals that, despite decent performance on tables without data artifacts, frontier models degrade significantly when data artifacts are introduced, exposing critical gaps in their capacity for robust, data-aware analysis. Designed to be flexible and extensible, RADAR supports diverse perturbation types and controllable table sizes, offering a valuable resource for advancing tabular reasoning.

We present SensorLM, a family of sensor-language foundation models that enable wearable sensor data understanding with natural language. Despite its pervasive nature, aligning and interpreting sensor data with language remains challenging due to the lack of paired, richly annotated sensor-text descriptions in uncurated, real-world wearable data. We introduce a hierarchical caption generation pipeline designed to capture statistical, structural, and semantic information from sensor data. This approach enabled the curation of the largest sensor-language dataset to date, comprising over 59.7 million hours of data from more than 103,000 people. Furthermore, SensorLM extends prominent multimodal pretraining architectures (e.g., CLIP, CoCa) and recovers them as specific variants within a generic architecture. Extensive experiments on real-world tasks in human activity analysis and healthcare verify the superior performance of SensorLM over state-of-the-art in zero-shot recognition, few-shot learning, and cross-modal retrieval. SensorLM also demonstrates intriguing capabilities including scaling behaviors, label efficiency, sensor captioning, and zero-shot generalization to unseen tasks. Code is available at https://github.com/Google-Health/consumer-health-research/tree/main/sensorlm.

All-day smart glasses are likely to emerge as platforms capable of continuous contextual sensing, uniquely positioning them for unprecedented assistance in our daily lives. Integrating the multi-modal AI agents required for human memory enhancement while performing continuous sensing, however, presents a major energy efficiency challenge for all-day usage. Achieving this balance requires intelligent, context-aware sensor management. Our approach, EgoTrigger, leverages audio cues from the microphone to selectively activate power-intensive cameras, enabling efficient sensing while preserving substantial utility for human memory enhancement. EgoTrigger uses a lightweight audio model (YAMNet) and a custom classification head to trigger image capture from hand-object interaction (HOI) audio cues, such as the sound of a drawer opening or a medication bottle being opened. In addition to evaluating on the QA-Ego4D dataset, we introduce and evaluate on the Human Memory Enhancement Question-Answer (HME-QA) dataset. Our dataset contains 340 human-annotated first-person QA pairs from full-length Ego4D videos that were curated to ensure that they contained audio, focusing on HOI moments critical for contextual understanding and memory. Our results show EgoTrigger can use 54% fewer frames on average, significantly saving energy in both power-hungry sensing components (e.g., cameras) and downstream operations (e.g., wireless transmission), while achieving comparable performance on datasets for an episodic memory task. We believe this context-aware triggering strategy represents a promising direction for enabling energy-efficient, functional smart glasses capable of all-day use - supporting applications like helping users recall where they placed their keys or information about their routine activities (e.g., taking medications).

Akshay Paruchuri, Jake Garrison, Shun Liao, John B Hernandez, Jacob Sunshine, Tim Althoff, Xin Liu, Daniel McDuff. Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing. 2024.

Understanding of human visual perception has historically inspired the design of computer vision architectures. As an example, perception occurs at different scales both spatially and temporally, suggesting that the extraction of salient visual information may be made more effective by attending to specific features at varying scales. Visual changes in the body, due to physiological processes, also occur at varying scales and with modality-specific characteristic properties. Inspired by this, we present BigSmall, an efficient architecture for physiological and behavioral measurement. We present the first joint camera-based facial action, cardiac, and pulmonary measurement model. We propose a multi-branch network with wrapping temporal shift modules that yields efficiency gains and accuracy on par with task-optimized methods. We observe that fusing low-level features leads to suboptimal performance, but that fusing high level features enables efficiency gains with negligible losses in accuracy. We experimentally validate that BigSmall significantly reduces computational cost while achieving comparable results on multiple physiological measurement tasks simultaneously with a unified model.

Machine learning models for camera-based physiological measurement can have weak generalization due to a lack of representative training data. Body motion is one of the most significant sources of noise when attempting to recover the subtle cardiac pulse from a video. We explore motion transfer as a form of data augmentation to introduce motion variation while preserving physiological changes of interest. We adapt a neural video synthesis approach to augment videos for the task of remote photoplethysmography (rPPG) and study the effects of motion augmentation with respect to 1) the magnitude and 2) the type of motion. After training on motion-augmented versions of publicly available datasets, we demonstrate a 47% improvement over existing inter-dataset results using various state-of-the-art methods on the PURE dataset. We also present inter-dataset results on five benchmark datasets to show improvements of up to 79% using TS-CAN, a neural rPPG estimation method. Our findings illustrate the usefulness of motion transfer as a data augmentation technique for improving the generalization of models for camera-based physiological sensing. We release our code for using motion transfer as a data augmentation technique on three publicly available datasets, UBFC-rPPG, PURE, and SCAMPS, and models pre-trained on motion-augmented data here: https://motion-matters.github.io/

Camera-based physiological measurement is a fast growing field of computer vision. Remote photoplethysmography (rPPG) utilizes imaging devices (e.g., cameras) to measure the peripheral blood volume pulse (BVP) via photoplethysmography, and enables cardiac measurement via webcams and smartphones. However, the task is non-trivial with important pre-processing, modeling and post-processing steps required to obtain state-of-the-art results. Replication of results and benchmarking of new models is critical for scientific progress; however, as with many other applications of deep learning, reliable codebases are not easy to find or use. We present a comprehensive toolbox, rPPG-Toolbox, unsupervised and supervised rPPG models with support for public benchmark datasets, data augmentation and systematic evaluation: https://github.com/ubicomplab/rPPG-Toolbox.

Camera-based physiological measurement is a growing field with neural models providing state-of-the-art performance. Prior research has explored various "end-to-end" architectures; however these methods still require several preprocessing steps and are not able to run directly on mobile and edge devices. The operations are often non-trivial to implement, making replication and deployment difficult and can even have a higher computational budget than the "core" network itself. In this paper, we propose two novel and efficient neural models for camera-based physiological measurement called EfficientPhys that remove the need for face detection, segmentation, normalization, color space transformation or any other preprocessing steps. Using an input of raw video frames, our models achieve strong accuracy on three public datasets. We show that this is the case whether using a transformer or convolutional backbone. We further evaluate the latency of the proposed networks and show that our most lightweight network also achieves a 33% improvement in efficiency.

The use of cameras and computational algorithms for noninvasive, low-cost and scalable measurement of physiological (e.g., cardiac and pulmonary) vital signs is very attractive. However, diverse data representing a range of environments, body motions, illumination conditions and physiological states is laborious, time consuming and expensive to obtain. Synthetic data have proven a valuable tool in several areas of machine learning, yet are not widely available for camera measurement of physiological states. Synthetic data offer "perfect" labels (e.g., without noise and with precise synchronization), labels that may not be possible to obtain otherwise (e.g., precise pixel level segmentation maps) and provide a high degree of control over variation and diversity in the dataset. We present SCAMPS, a dataset of synthetics containing 2,800 videos (1.68M frames) with aligned cardiac and respiratory signals and facial action intensities. The RGB frames are provided alongside segmentation maps and precise descriptive statistics about the underlying waveforms, including inter-beat interval, heart rate variability, and pulse arrival time. Finally, we present baseline results training on these synthetic data and testing on real-world datasets to illustrate generalizability.

Recent research has demonstrated the capability of behavior signals captured by smartphones and wearables for longitudinal behavior modeling. However, there is a lack of a comprehensive public dataset that serves as an open testbed for fair comparison among algorithms. Moreover, prior studies mainly evaluate algorithms using data from a single population within a short period, without measuring the cross-dataset generalizability of these algorithms. We present the first multi-year passive sensing datasets, containing over 700 user-years and 497 unique users’ data collected from mobile and wearable sensors, together with a wide range of well-being metrics. Our datasets can support multiple cross-dataset evaluations of behavior modeling algorithms’ generalizability across different users and years. As a starting point, we provide the benchmark results of 18 algorithms on the task of depression detection. Our results indicate that both prior depression detection algorithms and domain generalization techniques show potential but need further research to achieve adequate cross-dataset generalizability. We envision our multi-year datasets can support the ML community in developing generalizable longitudinal behavior modeling algorithms.

We present HulaMove, a novel interaction technique that leverages the movement of the waist as a new eyes-free and hands-free input method for both the physical world and the virtual world. We first conducted a user study (N=12) to understand users’ ability to control their waist. We found that users could easily discriminate eight shifting directions and two rotating orientations, and quickly confirm actions by returning to the original position (quick return). We developed a design space with eight gestures for waist interaction based on the results and implemented an IMU-based real-time system. Using a hierarchical machine learning model, our system could recognize waist gestures at an accuracy of 97.5%. Finally, we conducted a second user study (N=12) for usability testing in both real-world scenarios and virtual reality settings. Our usability study indicated that HulaMove significantly reduced interaction time by 41.8% compared to a touch screen method, and greatly improved users’ sense of presence in the virtual world. This novel technique provides an additional input method when users’ eyes or hands are busy, accelerates users’ daily operations, and augments their immersive experience in the virtual world.

Sleep is critical to human function, mediating factors like memory, mood, energy, and alertness; therefore, it is commonly conjectured that a good night’s sleep is important for job performance. However, both real-world sleep behavior and job performance are difficult to measure at scale. In this work, we demonstrate that people’s everyday interactions with online mobile apps can reveal insights into their job performance in real-world contexts. We present an observational study in which we objectively tracked the sleep behavior and job performance of salespeople (N = 15) and athletes (N = 19) for 18 months, leveraging a mattress sensor and online mobile app to conduct the largest study of this kind to date. We first demonstrate that cumulative sleep measures are significantly correlated with job performance metrics, showing that an hour of daily sleep loss for a week was associated with a 9.0% average reduction in contracts established for salespeople and a 9.5% average reduction in game grade for the athletes. We then investigate the utility of online app interaction time as a passively collectible and scalable performance indicator. We show that app interaction time is correlated with the job performance of the athletes, but not the salespeople. To support that our app-based performance indicator truly captures meaningful variation in psychomotor function as it relates to sleep and is robust against potential confounds, we conducted a second study to evaluate the relationship between sleep behavior and app interaction time in a cohort of 274 participants. Using a generalized additive model to control for per-participant random effects, we demonstrate that participants who lost one hour of daily sleep for a week exhibited average app interaction times that were 5.0% slower. We also find that app interaction time exhibits meaningful chronobiologically consistent correlations with sleep history, time awake, and circadian rhythms. The findings from this work reveal an opportunity for online app developers to generate new insights regarding cognition and productivity.

Telehealth and remote health monitoring have become increasingly important during the SARS-CoV-2 pandemic and it is widely expected that this will have a lasting impact on healthcare practices. These tools can help reduce the risk of exposing patients and medical staff to infection, make healthcare services more accessible, and allow providers to see more patients. However, objective measurement of vital signs is challenging without direct contact with a patient. We present a video-based and on-device optical cardiopulmonary vital sign measurement approach. It leverages a novel multi-task temporal shift convolutional attention network (MTTS-CAN) and enables real-time cardiovascular and respiratory measurements on mobile platforms. We evaluate our system on an Advanced RISC Machine (ARM) CPU and achieve state-of-the-art accuracy while running at over 150 frames per second which enables real-time applications. Systematic experimentation on large benchmark datasets reveals that our approach leads to substantial (20%-50%) reductions in error and generalizes well across datasets.

Continuous monitoring of hand function in individuals with upper limb paresis, such as stroke survivors, could provide a quantitative assessment of their real-world functional performance, which has great potential to enhance the clinical guidance of rehabilitation interventions. In this paper, we explore a novel wearable approach to quantify the amount of hand use by leveraging Radio Frequency Identification (RFID) technologies. We introduce a prototype implementation of our wearable RFID system composed of a wrist-worn reader (antenna) and a small passive tag placed on a fingernail. Then, we discuss a machine learning-based data analytic pipeline that analyzes the backscattered RF signal to estimate the amount of hand use. The accuracy of the system is validated against an optoelectronic motion capture system - the gold standard for human movement analyses - using a dataset collected from five neurologically intact individuals. The proposed wearable RFID system could accurately estimate the amount of hand use with R2of 0.67 and Normalized Root Mean Square Error of 7.3%, and shows great potential for clinical applications.

Objective assessment of stroke survivors' upper limb movements in ambulatory settings can provide clinicians with important information regarding the real impact of rehabilitation outside the clinic and help to establish individually-tailored therapeutic programs. This paper explores a novel approach to monitor the amount of hand use, which is relevant to the purposeful, goal-directed use of the limbs, based on a body networked sensor system composed of miniaturized finger- and wrist-worn accelerometers. The main contributions of this paper are twofold. First, this paper introduces and validates a new benchmark measurement of the amount of hand use based on data recorded by a motion capture system, the gold standard for human movement analysis. Second, this paper introduces a machine learning-based analytic pipeline that estimates the amount of hand use using data obtained from the wearable sensors and validates its estimation performance against the aforementioned benchmark measurement. Based on data collected from 18 neurologically intact individuals performing 11 motor tasks resembling various activities of daily living, the analytic results presented herein show that our new benchmark measure is reliable and responsive, and that the proposed wearable system can yield an accurate estimation of the amount of hand use (normalized root mean square error of 0.11 and average Pearson correlation of 0.78). This study has the potential to open up new research and clinical opportunities for monitoring hand function in ambulatory settings, ultimately enabling evidence-based, patient-centered rehabilitation and healthcare.