Pascal Frossard
Full Professor
Leads LTS4, working on graph signal processing, robustness of deep networks and machine learning for structured data.
Pascal Frossard's lab at EPFL works on trustworthy & safe ML, signal processing & information theory and speech enhancement & signal processing. 27 papers since 2023, 10 at top venues, mostly NeurIPS and AAAI. 16 PhD students. 34 alumni, most often to Sony and Google. Applications close .
At a glance
- papers, 2023–2026
- 7/ year
- at T1 venues
- 37%
- PhD students
- 16
- MS students
- 0
- alumni listed
- 34
Papers by yearfrom the lab homepage, 2023–2026
- 11 papers2026 · T1 4 · collecting
- 16 papers2025 · T1 6
- 0 papers2024 · T1 0
- 0 papers2023 · T1 0
T1 venuesOtherVenues: NeurIPS, AAAI, ECCV, ICASSP, ISBI
Research focuspaper titles that use each topic's words
- Signal processing & information theory2 papers
- Speech enhancement & signal processing2 papers
- graph signal1 paper
- Trustworthy & safe ML0 papers
- deep networks0 papers
Paper list
From the lab homepage.
202611 papers
- Medical Image Analysis (T1)Future cardiovascular events prediction from invasive coronary angiography: A graph representation learning perspective (opens in a new tab)
- IEEE Signal Processing LettersGraph Signal Separation With Learnable Spectral Filters (opens in a new tab)
- IEEE Transactions on Signal and Information Processing over NetworksGraph-Dictionary Signal Model for Sparse Representation of Multivariate Data (opens in a new tab)
- IEEE Transactions on Intelligent Transportation SystemsLane-Changing Maneuver Prediction With Urban Arterial Traffic Data From Drones (opens in a new tab)
- —Multi-scale modeling of human tissues from spatial transcriptomics with TERRA (opens in a new tab)
- Nature Communications (T1)Prognostic RNA-splicing archetypes in breast cancer identified by extended pre-training of histopathology foundation models (opens in a new tab)
- —Resolving context-specific protein-protein interactomes for biological discovery and therapeutic target prioritisation (opens in a new tab)
- NEJM AIResponse: Machine Learning Should Enrich, Not Replace, Human Deliberation in End-of-Life Decisions (opens in a new tab)
- MICCAI (T1)Revisiting Automatic Data Curation for Vision Foundation Models in Digital Pathology (opens in a new tab)
- AAAI (T1)Semantic Document Derendering: SVG Reconstruction via Vision-Language Modeling (opens in a new tab)
- —Topology-aware reconstruction of cellular state landscapes from microscopy using self-supervised learning (opens in a new tab)
202516 papers
- ACM SIGGRAPH European Conference on Visual Media ProductionA Fast Volumetric Capture and Reconstruction Pipeline for Dynamic Point Clouds and Gaussian Splats (opens in a new tab)
- The International Journal of Cardiovascular ImagingAI-Driven multi-view learning from CCTA for myocardial infarction diagnosis (opens in a new tab)
- EMBCCM-UNet: A Self-Supervised Learning-Based Model for Coronary Artery Segmentation in X-Ray Angiography (opens in a new tab)
- NeurIPS (T1)Continuous Simplicial Neural Networks (opens in a new tab)
- NATURE REVIEWS MOLECULAR CELL BIOLOGYDecoding the Interactions and Functions of Non-coding Rna with Artificial Intelligence (opens in a new tab)
- —Extended pre-training of histopathology foundation models uncovers co-existing breast cancer archetypes characterized by RNA splicing or TGF-β dysregulation (opens in a new tab)
- NeurIPS (T1)Flow based approach for Dynamic Temporal Causal models with non-Gaussian or Heteroscedastic Noises (opens in a new tab)
- IEEE Transactions on Artificial IntelligenceGraphTeacher: Transductive Fine-Tuning of Encoders through Graph Neural Networks (opens in a new tab)
- NeurIPS (T1)Inductive Domain Transfer In Misspecified Simulation-Based Inference (opens in a new tab)
- NEJM AIMachine Learning–Based Patient Preference Prediction: A Proof of Concept (opens in a new tab)
- NeurIPS (T1)MEMOIR: Lifelong Model Editing with Minimal Overwrite and Informed Retention for LLMs (opens in a new tab)
- ICASSP (T1)OSLO-IC: On-the-Sphere Learned Omnidirectional Image Compression with Attention Modules and Spatial Context (opens in a new tab)
Where alumni went34 listed, 76% to industry
Industry 16Academia 5
Most often to
- Sony
- Google DeepMind
Also to
- EPFL
- Imperial College London
- UCL
Apply to EPFLFall 2027
EDIC Computer and Communication Sciences PhD
- Round
- Round 1
- Opens
- —
- GRE
- optional
- English
- TOEFL 100 / IELTS 7.0
- Fee
- none
Confirmed on the program page.
Similar labsnearest by research summary
- USMichigan
Gabriel Poesia
- Reasoning & neuro-symbolic AI
- CNSJTU
Wenrui Dai
- Learning theory & optimization
- CNZhejiang
Jiachi Chen
- Trustworthy & safe ML
- USUIUC
Bo Li
- Trustworthy & safe ML
- USCMU
Hoda Heidari
- Trustworthy & safe ML
- USMichigan
Samet Oymak
- Learning theory & optimization
Your notes
Save this lab to keep notes.
Data
Checked against the lab homepage.
- Collected
- Verified
- Papers
- The lab homepage
- Members
- Lab homepage
- Alumni
- Lab homepage
Tiers are venue classifications, not lab rankings.
EPFLcloses