Scholar
Philippe Mulhem
Google Scholar ID: _NlUevwAAAAJ
LIG-CNRS
Citations & Impact
Citations
424
H-index
11
i10-index
14
Publications
20
Co-authors
0
Contact
Publications
Resume
Academic Achievements
- Co-organizer of CLEF 2024 conference (Grenoble, September 9–12, 2024)
- Co-organizer of LongEval CLEF 2024 campaign
- Published ECIR 2024 paper: 'LongEval: Longitudinal Evaluation of Model Performance at CLEF 2024'
- Program co-chair of CORIA 2024 (French IR conference, La Rochelle)
- Presented two papers at CORIA 2024 on collection impact and visual exploration tools for longitudinal IR evaluation
- Publications available on HAL, LIG website, DBLP, and Google Scholar
Research Experience
- CNRS Researcher at Laboratoire d’Informatique de Grenoble (LIG)
- Head of the MRIM research team
- Worked until October 2016 on integrating language models on graphs for photograph description, in collaboration with Globe VIP
- Participated in multimedia indexing and retrieval evaluation campaigns: NTCIR Lifelogging, CLEF Image Annotation
- Co-managed (with Lorraine Goeuriot) a CLEF Tweet Contextualization Lab task on evaluating tweet indexing/retrieval quality from festivals
- Co-proposed (with Yves Chiaramella and Franck Fourel) the “Fetch&Browse” framework for structured multimedia document retrieval, influential in INEX campaigns (2003–2010)
Background
- Information Retrieval Researcher
- Works on modeling and experimenting with indexing and retrieval of text and visual data
- Currently focuses on continuous evaluation of Web search engines in the Kodicare project
- Also works on explainable video retrieval
- Collaborates with Schneider Electric on indexing and retrieval of normative documents
- Previously focused on theoretical IR models (e.g., using Jensen-Shannon divergences) and hypotheses like the Document-Document Matching Constraint
- Investigates whether such constraints apply to continuous representations (e.g., LDA, LSA, deep learning)