Fengmin Wu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Fengmin Wu has a total of 21 co-authors. Here are the top co-authors by number of shared publications:
- Sicong Liu
Northwestern Polytechnical University
4 shared publications - Zhiwen Yu
Harbin Engineering University
4 shared publications - Bin Guo
Northwestern Polytechnical University
3 shared publications - Hongkai Wen
University of Warwick, Coventry, United Kingdom
2 shared publications - Lehao Wang
Northwestern Polytechnical University, Xi'an, China
2 shared publications - Zimu Zhou
Department of Data Science, City University of Hong Kong, Hong Kong, China
2 shared publications
Latest Publications
- Open Access
- 26 October 2026
- Published by Association for Computing Machinery (ACM)
- p. 297-311
AppFlow: Memory Scheduling for Cold Launch of Large Apps on Mobile and Vehicle Systems
GB-scale large apps like on-device LLMs and rich media editors are becoming the next-generation trend, but their heavy memory and I/O demands, especially during multitasking, cause devices to reclaim or kill processes, turning warm apps into cold launches. The challenge lies not in storing them, but in fast, accurate launching. For users, 1s is the usability cliff, yet our measurements show 86.6% of GB-scale cold launches exceed it. Also, Android Vitals flags only ≥5s as slow, exposing a large satisfaction gap. Existing optimizations are designed in isolation and conflict. For example, preloading reduces I/O stalls but consumes scarce memory and is undone by reclamation, while reclamation and killing free memory but sacrifice background survivability, leading to repeated cold relaunches. Our key insight is that, although multitasking makes runtime behavior complex, each app's file access pattern remains predictable. The challenge lies in exploiting this predictability, i.e., preloading without exhausting memory, reclaiming without undoing gains, and killing selectively to preserve background survivability. We introduce AppFlow, a prediction-based system-wide scheduler that integrates a Selective File Preloader, an Adaptive Memory Reclaimer, and a Context-Aware Process Killer. Implemented across the Android framework and Linux kernel without app changes, AppFlow cuts GB-scale cold-launch latency by 66.5% (e.g., 2s→690ms) and sustains 95% of launches within 1s over a 100-day test, significantly improving responsiveness and multitasking experience.
- 26 October 2026
- Published by Association for Computing Machinery (ACM)
- p. 297-311
Open AccessAppFlow: Memory Scheduling for Cold Launch of Large Apps on Mobile and Vehicle Systems
GB-scale large apps like on-device LLMs and rich media editors are becoming the next-generation trend, but their heavy memory and I/O demands, especially during multitasking, cause devices to reclaim or kill processes, turning warm apps into cold launches. The challenge lies not in storing them, but in fast, accurate launching. For users, 1s is the usability cliff, yet our measurements show 86.6% of GB-scale cold launches exceed it. Also, Android Vitals flags only ≥5s as slow, exposing a large satisfaction gap. Existing optimizations are designed in isolation and conflict. For example, preloading reduces I/O stalls but consumes scarce memory and is undone by reclamation, while reclamation and killing free memory but sacrifice background survivability, leading to repeated cold relaunches. Our key insight is that, although multitasking makes runtime behavior complex, each app's file access pattern remains predictable. The challenge lies in exploiting this predictability, i.e., preloading without exhausting memory, reclaiming without undoing gains, and killing selectively to preserve background survivability. We introduce AppFlow, a prediction-based system-wide scheduler that integrates a Selective File Preloader, an Adaptive Memory Reclaimer, and a Context-Aware Process Killer. Implemented across the Android framework and Linux kernel without app changes, AppFlow cuts GB-scale cold-launch latency by 66.5% (e.g., 2s→690ms) and sustains 95% of launches within 1s over a 100-day test, significantly improving responsiveness and multitasking experience.
- Open Access
- 30 September 2026
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies
- Vol. 10 (3) , 1-27
UbiHearo: Bringing Scenario-Aware Sound Guidance to DHH Users with Mobile Agents
For Deaf and Hard-of-Hearing (DHH) users, effective sound awareness goes beyond detecting what sound occurs; it requires understanding what it means, how urgent it is , and what to do. However, existing mobile assistants largely reduce acoustic scenes to isolated labels or generic alerts, failing to provide contextual guidance when the same sound implies different risks across scenarios. We present UbiHearo, a mobile hearing assistant agent that closes the sensing-reasoning-action loop for scenario-aware and personalized DHH assistance. UbiHearo adopts a neuro-symbolic design that transforms continuous acoustic and physical context streams into structured intermediate representations (IRs) for controllable reasoning and guidance. It introduces three key mechanisms. First , a perception-thresholded sensing mechanism enables energy-efficient always-on awareness by activating processing only for perceptually meaningful events while preserving safety-critical cues through deterministic overrides. Second , a staged neuro-symbolic reasoning framework performs promptless scenario understanding by conditioning an on-device audio-language model on structured IRs instead of user-authored context descriptions. Third , a bounded human-in-the-loop personalization mechanism adapts user-specific guidance through lightweight on-device decision-boundary updates while preserving population-aligned reasoning. We evaluate UbiHearo on smartphones and in-vehicle systems across diverse scenarios. It achieves favorable accuracy-latency-energy tradeoffs and personalized guidance.
- 30 September 2026
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies
- Vol. 10 (3) , 1-27
Open AccessUbiHearo: Bringing Scenario-Aware Sound Guidance to DHH Users with Mobile Agents
For Deaf and Hard-of-Hearing (DHH) users, effective sound awareness goes beyond detecting what sound occurs; it requires understanding what it means, how urgent it is , and what to do. However, existing mobile assistants largely reduce acoustic scenes to isolated labels or generic alerts, failing to provide contextual guidance when the same sound implies different risks across scenarios. We present UbiHearo, a mobile hearing assistant agent that closes the sensing-reasoning-action loop for scenario-aware and personalized DHH assistance. UbiHearo adopts a neuro-symbolic design that transforms continuous acoustic and physical context streams into structured intermediate representations (IRs) for controllable reasoning and guidance. It introduces three key mechanisms. First , a perception-thresholded sensing mechanism enables energy-efficient always-on awareness by activating processing only for perceptually meaningful events while preserving safety-critical cues through deterministic overrides. Second , a staged neuro-symbolic reasoning framework performs promptless scenario understanding by conditioning an on-device audio-language model on structured IRs instead of user-authored context descriptions. Third , a bounded human-in-the-loop personalization mechanism adapts user-specific guidance through lightweight on-device decision-boundary updates while preserving population-aligned reasoning. We evaluate UbiHearo on smartphones and in-vehicle systems across diverse scenarios. It achieves favorable accuracy-latency-energy tradeoffs and personalized guidance.
- Open Access
- 20 June 2026
- Published by Association for Computing Machinery (ACM)
- p. 392-404
AdaSprite: Resource-efficient Online Co-Adaptation for V2I Systems Under Large-scale Data Drifts
- 20 June 2026
- Published by Association for Computing Machinery (ACM)
- p. 392-404
Open AccessAdaSprite: Resource-efficient Online Co-Adaptation for V2I Systems Under Large-scale Data Drifts
- 24 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. PP (15361233) , 1-14
AdaFlowLite: Scalable and Non-Blocking Inference on Asynchronous Mobile Data
The rise of mobile devices equipped with numerous sensors, such as LiDAR and cameras, has driven the adoption of multi-modal deep intelligence for distributed sensing tasks, such as smart cabins and driving assistance. However, the arrival time of mobile sensory data vary due to modality size and network dynamics, which can lead to delays (if waiting for slow data) or accuracy decline (if inference proceeds without waiting). Moreover, the diversity and dynamic nature of mobile systems exacerbate this challenge. In response, we present a shift to opportunistic inference for asynchronous distributed multi-modal data, enabling inference as soon as partial data arrives. While existing methods focus on optimizing modality consistency and complementarity, known as modal affinity, they lack a computational approach to control this affinity in open-world mobile environments. ${\sf AdaFlowLite}$ pioneers the formulation of structured cross-modality affinity in mobile contexts using a hierarchical analysis-based normalized matrix. This approach accommodates the diversity and dynamics of modalities, generalizing across different types and numbers of inputs. Employing an multi-modal lightweight Swin Transformer (MMLST), ${\sf AdaFlowLite}$ facilitates real-time and flexible data imputation, adapting to various modalities and downstream tasks without retraining. Experiments show that ${\sf AdaFlowLite}$ significantly reduces inference latency by up to 80.4% and enhances accuracy by up to 62.1%, while achieving nearly a 50% reduction in energy consumption, outperforming status quo approaches. Also, this method can enhance LLM performance to preprocess asynchronous data.
- 24 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. PP (15361233) , 1-14
AdaFlowLite: Scalable and Non-Blocking Inference on Asynchronous Mobile Data
The rise of mobile devices equipped with numerous sensors, such as LiDAR and cameras, has driven the adoption of multi-modal deep intelligence for distributed sensing tasks, such as smart cabins and driving assistance. However, the arrival time of mobile sensory data vary due to modality size and network dynamics, which can lead to delays (if waiting for slow data) or accuracy decline (if inference proceeds without waiting). Moreover, the diversity and dynamic nature of mobile systems exacerbate this challenge. In response, we present a shift to opportunistic inference for asynchronous distributed multi-modal data, enabling inference as soon as partial data arrives. While existing methods focus on optimizing modality consistency and complementarity, known as modal affinity, they lack a computational approach to control this affinity in open-world mobile environments. ${\sf AdaFlowLite}$ pioneers the formulation of structured cross-modality affinity in mobile contexts using a hierarchical analysis-based normalized matrix. This approach accommodates the diversity and dynamics of modalities, generalizing across different types and numbers of inputs. Employing an multi-modal lightweight Swin Transformer (MMLST), ${\sf AdaFlowLite}$ facilitates real-time and flexible data imputation, adapting to various modalities and downstream tasks without retraining. Experiments show that ${\sf AdaFlowLite}$ significantly reduces inference latency by up to 80.4% and enhances accuracy by up to 62.1%, while achieving nearly a 50% reduction in energy consumption, outperforming status quo approaches. Also, this method can enhance LLM performance to preprocess asynchronous data.
- Open Access
- 4 November 2024
- Published by Association for Computing Machinery (ACM)
- p. 928-929
Non-blocking Inference on Asynchronous Mobile Sensor Data with Affinity Control
The rise of sensor-rich mobile devices has led to the adoption of multi-modal deep intelligence for distributed sensing tasks. However, varying arrival times of mobile sensory data can cause delays or accuracy decline. The diversity and dynamic nature of mobile systems further exacerbate this challenge. To address this, we present an opportunistic inference approach for asynchronous distributed multi-modal data, enabling inference upon partial data arrival. AdaFlow pioneers the formulation of structured cross-modality affinity using a hierarchical analysis-based normalized matrix, accommodating the diversity and dynamics of modalities. Employing an affinity attention-based conditional GAN (ACGAN), AdaFlow facilitates flexible data imputation, adapting to various modalities and tasks without retraining. Experiments show significant reductions in inference latency and enhanced accuracy compared to status quo approaches.
- 4 November 2024
- Published by Association for Computing Machinery (ACM)
- p. 928-929
Open AccessNon-blocking Inference on Asynchronous Mobile Sensor Data with Affinity Control
The rise of sensor-rich mobile devices has led to the adoption of multi-modal deep intelligence for distributed sensing tasks. However, varying arrival times of mobile sensory data can cause delays or accuracy decline. The diversity and dynamic nature of mobile systems further exacerbate this challenge. To address this, we present an opportunistic inference approach for asynchronous distributed multi-modal data, enabling inference upon partial data arrival. AdaFlow pioneers the formulation of structured cross-modality affinity using a hierarchical analysis-based normalized matrix, accommodating the diversity and dynamics of modalities. Employing an affinity attention-based conditional GAN (ACGAN), AdaFlow facilitates flexible data imputation, adapting to various modalities and tasks without retraining. Experiments show significant reductions in inference latency and enhanced accuracy compared to status quo approaches.
- Open Access
- 4 November 2024
- Published by Association for Computing Machinery (ACM)
- p. 606-618
AdaFlow: Opportunistic Inference on Asynchronous Mobile Data with Generalized Affinity Control
The rise of mobile devices equipped with numerous sensors, such as LiDAR and cameras, has spurred the adoption of multi-modal deep intelligence for distributed sensing tasks, such as smart cabins and driving assistance. However, the arrival times of mobile sensory data vary due to modality size and network dynamics, which can lead to delays (if waiting for slower data) or accuracy decline (if inference proceeds without waiting). Moreover, the diversity and dynamic nature of mobile systems exacerbate this challenge. In response, we present a shift to opportunistic inference for asynchronous distributed multi-modal data, enabling inference as soon as partial data arrives. While existing methods focus on optimizing modality consistency and complementarity, known as modal affinity, they lack a computational approach to control this affinity in open-world mobile environments. AdaFlow pioneers the formulation of structured cross-modality affinity in mobile contexts using a hierarchical analysis-based normalized matrix. This approach accommodates the diversity and dynamics of modalities, generalizing across different types and numbers of inputs. Employing an affinity attention-based conditional GAN (ACGAN), AdaFlow facilitates flexible data imputation, adapting to various modalities and downstream tasks without retraining. Experiments show that AdaFlow significantly reduces inference latency by up to 79.9% and enhances accuracy by up to 61.9%, outperforming status quo approaches. Also, this method can enhance LLM performance to preprocess asynchronous data.
- 4 November 2024
- Published by Association for Computing Machinery (ACM)
- p. 606-618
Open AccessAdaFlow: Opportunistic Inference on Asynchronous Mobile Data with Generalized Affinity Control
The rise of mobile devices equipped with numerous sensors, such as LiDAR and cameras, has spurred the adoption of multi-modal deep intelligence for distributed sensing tasks, such as smart cabins and driving assistance. However, the arrival times of mobile sensory data vary due to modality size and network dynamics, which can lead to delays (if waiting for slower data) or accuracy decline (if inference proceeds without waiting). Moreover, the diversity and dynamic nature of mobile systems exacerbate this challenge. In response, we present a shift to opportunistic inference for asynchronous distributed multi-modal data, enabling inference as soon as partial data arrives. While existing methods focus on optimizing modality consistency and complementarity, known as modal affinity, they lack a computational approach to control this affinity in open-world mobile environments. AdaFlow pioneers the formulation of structured cross-modality affinity in mobile contexts using a hierarchical analysis-based normalized matrix. This approach accommodates the diversity and dynamics of modalities, generalizing across different types and numbers of inputs. Employing an affinity attention-based conditional GAN (ACGAN), AdaFlow facilitates flexible data imputation, adapting to various modalities and downstream tasks without retraining. Experiments show that AdaFlow significantly reduces inference latency by up to 79.9% and enhances accuracy by up to 61.9%, outperforming status quo approaches. Also, this method can enhance LLM performance to preprocess asynchronous data.