Fangcheng Fu
副教授
Works on machine learning systems and data management, including the Hetu distributed deep learning system.
Fangcheng Fu's lab at SJTU works on distributed, cloud & edge systems and data management & big data. 56 papers since 2023, 45 at top venues, mostly ICML and NeurIPS. Applications close .
At a glance
- papers, 2023–2026
- 15/ year
- at T1 venues
- 80%
- PhD students
- —
- MS students
- —
- alumni listed
- —
Papers by yearfrom the lab homepage, 2023–2026
- 22 papers2026 · T1 19 · collecting
- 21 papers2025 · T1 17
- 8 papers2024 · T1 6
- 5 papers2023 · T1 3
T1 venuesOtherVenues: ICML, NeurIPS, SIGMOD, ASPLOS, ICDE
Research focuspaper titles that use each topic's words
- Data management & big data12 papers
- Distributed, cloud & edge systems5 papers
- system5 papers
- data management0 papers
- learning systems0 papers
Paper list
From the lab homepage.
202622 papers
- MLSys (T1)BOute: Cost-Efficient LLM Serving with Heterogeneous LLMs and GPUs via Multi-Objective Bayesian Optimization
- ICLR (T1)Cascadia: An Efficient Cascade Serving System for Large Language Models
- TKDE (T1)Collaborative-Guided Diffusion for Sequential Recommendation
- ICML (T1)DARTS: Distribution-Aware Active Rollout Trajectory Shaping for Accelerating LLM Reinforcement Learning
- Retrieval-Augmented Generation for AI-Generated Content: A SurveyData Science and Engineering 11:1-29 (2026)
- NeurIPS (T1)DataFlex: A Unified Benchmark and Evaluation Platform for Data-Centric Training of Large Language Models
- ICML (T1)EchoAttention: Exploiting Token-Pair Redundancy and Frame-Block Similarity for Efficient Long Video Generation
- ICML (T1)Efficient Multi-round LLM Inference over Disaggregated Serving
- NeurIPS (T1)Efficient Serving for Dynamic Agent Workflows with Prediction-based KV-Cache Management
- PPoPPElastor: Elastic and Efficient Model Partitioning and Checkpointing for Fault-tolerant Distributed Training
- NeurIPS (T1)FELPS: Fair and Efficient Scheduling for Multi-LoRA Serving System
- OSDI (T1)Hetu v2: A General and Scalable Deep Learning System with Hierarchical and Heterogeneous Single Program Multiple Data Annotations
202521 papers
- SIGCOMM (T1)ByteScale: Communication-Efficient Scaling of LLM Training with a 2048K Context Length on 16384 GPUs
- ICML (T1)Demystifying Cost-Efficiency in LLM Serving over Heterogeneous GPUs
- ACL (T1)Enhancing Unsupervised Sentence Embeddings via Knowledge-Driven Data Augmentation and Gaussian-Decayed Contrastive Learning
- ASPLOS (T1)FlexSP: Accelerating Large Language Model Training via Flexible Sequence Parallelism
- AAAI (T1)HaCore: Efficient Coreset Construction with Locality Sensitive Hashing for Vertical Federated Learning
- ICDE (T1)Hounding Data Diversity: Towards Participant Selection in Vertical Federated Learning
- Towards Scalable and Efficient Graph Structure LearningICDE 2025
- EMNLP (T1)Improving Low-Resource Sequence Labeling with Knowledge Fusion and Contextual Label Explanations
- VLDB (T1)LobRA: Multi-tenant Fine-tuning over Heterogeneous Data
- SIGMOD (T1)Malleus: Straggler-Resilient Hybrid Parallel Training of Large-scale Models via Malleable Data and Model Parallelization
- SIGMOD (T1)Memo: Fine-grained Tensor Management For Ultra-long Context LLM Training
- IJCAI (T1)Model Rake: A Defense Against Stealing Attacks in Split Learning
20248 papers
- AAAI (T1)Accelerating Text-to-image Editing via Cache-enabled Sparse Diffusion Inference
- NeurIPS (T1)Efficient Multi-task LLM Quantization and Serving for Multiple LoRA Adapters
- SOSP (T1)Enabling Parallelism Hot Switching for Efficient Training of Large Language Models
- ICDE (T1)Generative and Contrastive Paradigms Are Complementary for Graph Self-Supervised Learning
- NeurIPS (T1)LSH-MoE: Communication-efficient MoE Training via Locality-Sensitive Hashing
- ProjPert: Projection-based Perturbation for Label Protection in Split Learning based Vertical Federated LearningTKDE 36(7):3417-3428 (2024)
- Improving Automatic Parallel Training via Balanced Memory Workload OptimizationTKDE 36(8):3906-3920 (2024)
- IJCAI (T1)X-former Elucidator: Reviving Efficient Attention for Long Context Language Modeling
20235 papers
- VLDB (T1)Angel-PTM: A Scalable and Economical Large-scale Pre-training System in Tencent
- ICDE (T1)KVSAgg: Secure Aggregation of Distributed Key-Value Sets
- Accelerating Text-to-image Editing via Cache-enabled Sparse Diffusion InferenceMLSys Workshop NeurIPS 2023
- IJCAI (T1)OSDP: Optimal Sharded Data Parallel for Distributed Deep Learning
- P2CG: A Privacy Preserving Collaborative Graph Neural Network Training FrameworkVLDB Journal 32(4):717-736 (2023)
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
- USCMU
Shreya Shankar
- Efficient training & inference
- GBUCL
Ran Ben Basat
- Efficient training & inference
- CNSJTU
Tao Song
- Distributed, cloud & edge systems
- USUMD
Abhinav Bhatele
- Efficient training & inference
- CNSJTU
Xingda Wei
- Efficient training & inference
- CNTsinghua
Wenlai Zhao
- Distributed, cloud & edge systems
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