Xiaofeng Hou
Associate Professor
Works on energy-efficient AI computing architectures and system software.
Xiaofeng Hou's lab at SJTU works on accelerators, chips & architecture and efficient training & inference. 42 papers since 2023, 7 at top venues, mostly ISCA and AAAI. Applications close .
At a glance
- papers, 2023–2026
- 11/ year
- at T1 venues
- 17%
- PhD students
- —
- MS students
- —
- alumni listed
- —
Papers by yearfrom the lab homepage, 2023–2026
- 12 papers2026 · T1 3 · collecting
- 6 papers2025 · T1 0
- 14 papers2024 · T1 3
- 10 papers2023 · T1 1
T1 venuesOtherVenues: ISCA, AAAI, ICDE, ICME, MICRO
Research focuspaper titles that use each topic's words
- Efficient training & inference16 papers
- Accelerators, chips & architecture4 papers
- energy-efficient1 paper
- system software0 papers
- computing architectures0 papers
Paper list
From the lab homepage.
202612 papers
- AAAI (T1)AdaReason: Progressive Training of Multi-LoRA Adapters for Budget-Adaptive Language Reasoning Models
- International Symposium on High-Performance Computer ArchitectureAUM: Unleashing the Efficiency Potential of Shared Processors with Accelerator Units for LLM Serving
- AAAI (T1)DesireKV: Decoupling Sensitivity and Importance for Reasoning-Aware KV Cache Compression
- MICRO (T1)Enhancing the Efficiency of Embodied AI System with Squeezed Analog Perception
- IEEE International Conference on Computer DesignForwarding Vision Encoding to Computational Sensors for Efficient Edge VLM Inference
- IEEE/ACM International Symposium on Quality of Service, full paperLocMore: Locating More Bursty Latency-critical Jobs on Resource-constrained Nodes
- International Conference on Architectural Support for Programming Languages and Operating SystemsMoE-APEX: An Efficient MoE Inference System with Adaptive Precision Expert Offloading
- Annual Technical ConferenceMoE-Lego: A Plane-Grained System for Dynamic-Precision MoE Offloading
- IEEE/ACM International Symposium on Quality of Service, short paperOpScope: Exploiting Operation-Driven Visual Scope for QoS-Stable Cloud Gaming
- ACM Transactions on Architecture and Code OptimizationPCR: A Prefetch-Enhanced Cache Reuse System for Low-Latency RAG Serving
- IEEE International Conference on Computer DesignSensE: From Sampled-Input Execution to Stream-Aware Edge Multimodal MoE Inference
- IEEE/ACM International Symposium on Quality of Service, short paperStarkServe: A Framework for Elastic Serverless LLM Inference at the Extreme Edge
20256 papers
- Transactions on Mobile ComputingBAT: A Versatile Bipartite Attention-based Approach for Comprehensive Truth Inference in Mobile Crowdsourcing
- ACM Transactions on Architecture and Code OptimizationEnhancing High-Throughput GPU Random Walks Through Multi-Task Concurrency Orchestration
- International Conference on Architectural Support for Programming Languages and Operating SystemsEXIST: Enabling Extremely Efficient Intra-Service Tracing Observability in Datacenters
- Frontiers of Computer ScienceFLAPS: fluctuation-aware power auction strategy for reducing the power overload probability
- Journal of Parallel and Distributed ComputingMMBypass: Towards Efficient Multi-modal AI Computing with Adaptive Bypass Network
- USENIX Annual Technical ConferenceSpaceExit: Enabling Efficient Adaptive Computing in Space with Early Exits
202414 papers
- ISCA (T1)A Tale of Two Domains: Exploring Efficient Architecture Design for Truly Autonomous Things
- ACM Transactions on Architecture and Code OptimizationA2: Towards Accelerator Level Parallelism for Autonomous Micromobility Systems
- IEEE International Conference on Computer DesignAutoVCoder: A Systematic Framework for Automated Verilog Code Generation using LLMs
- International Conference for High Performance Computing, Networking, Storage, and AnalysisBoosting Data Center Performance via Intelligently Managed Multi-backend Disaggregated Memory
- ISCA (T1)CoCG: Fine-grained Cloud Game Co-location on Heterogeneous Platform
- IEEE International Conference on Computer DesignContinuous Energy Efficiency Optimization for Autonomous Embedded Systems Using Shadow Cycles
- IEEE/ACM International Symposium on Quality of ServiceCPM: A Cross-layer Power Management Facility to Enable Highly-efficient Real-time AIoT Systems
- Proceedings of the International Conference on Very Large Data BasesFlowWalker: A Memory-efficient and High-performance GPU-based Dynamic Graph Random Walk Framework
- ICDE (T1)Graph Contrastive Learning for Truth Inferencec
- IEEE Transactions on ComputersImproving Efficiency in Multi-modal Autonomous Embedded Systems through Adaptive Gating
- EEE Real-Time Systems SymposiumJigsaw: Taming BEV-centric Perception on Dual-SoC for Autonomous Driving
- ICMEM2SN : Adaptive and Dynamic Multi-modal Shortcut Network Architecture for Latency-aware Applications
202310 papers
- ISCA (T1)Architecting Efficient Multi-modal AIoT Systems
- ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel ProgrammingCoWalker: High-Throughput GPU Random Walk with Fine-tuned Concurrent Query Processing (Poster)
- European Conference on Artificial IntelligenceLabel Aggregation with Self-Supervision Enhanced Graph Transformer
- IEEE International Symposium on Workload CharacterizationMMBench: Benchmarking End-to-End Multi-modal DNNs and Understanding Their Hardware-Software Implications
- European Conference on Parallel ProcessingMMExit: Enabling Fast and Efficient Multi-modal DNN Inference with Adaptive Network Exits
- ACM Symposium on Cloud ComputingNot All Resources are Visible: Exploiting Fragmented Shadow Resources in Shared-State Scheduler Architecture
- IEEE Transactions on ComputersOptimizing GPU-based Graph Sampling and Random Walk for Efficiency and Scalability
- Science China Information SciencesPower Synchronization: Taming Massive Diversified Serverless Functions under Power Constraints
- IEEE Journal on Selected Areas in CommunicationsPractical Network Modeling Using Weak Supervision Signals for Human-Centric Networking in Metaverse
- IEEE Real-Time Systems SymposiumSMG: A System-level Modality Gating Facility for Fast and Energy-Efficient Multimodal Computing
Where alumni went
Alumni are not listed on the lab homepage.
Apply to SJTUFall 2027
Computer Science PhD (international applicants)
- Round
- Self-funded
- Opens
- —
- GRE
- See program page
- English
- See program page
- Fee
- See program page
Not yet confirmed for this intake.
Similar labsnearest by research summary
- CHETH Zurich
Torsten Hoefler
- Distributed, cloud & edge systems
- CNSJTU
Guohao Dai
- Efficient training & inference
- GBCambridge
Robert Mullins
- Accelerators, chips & architecture
- CNSJTU
Yu Feng
- Efficient training & inference
- USUCSD
Tajana Šimunić Rosing
- Accelerators, chips & architecture
- CNSJTU
Jieru Zhao
22 T1
Your notes
Save this lab to keep notes.
Data
Some numbers could not be verified:
- The member list could not be found on the lab site.
- The alumni list could not be found on the lab site.
- Collected
- Verified
- Papers
- The lab homepage
- Members
- Not collected
- Alumni
- Not collected
Tiers are venue classifications, not lab rankings.
SJTUcloses