Scholar
Anna Rohrbach
Google Scholar ID: GHpxNQIAAAAJ
Professor, TU Darmstadt, Germany
Vision and Language
Artificial Intelligence
Multimodal Grounded Learning
Follow
Homepage
↗
Google Scholar
↗
Citations & Impact
All-time
Citations
9,382
H-index
37
i10-index
53
Publications
20
Co-authors
28
list available
Contact
No contact links provided.
Publications
11 items
HaloProbe: Bayesian Detection and Mitigation of Object Hallucinations in Vision-Language Models
2026
Cited
0
Tuning Just Enough: Lightweight Backdoor Attacks on Multi-Encoder Diffusion Models
2026
Cited
0
VeriTaS: The First Dynamic Benchmark for Multimodal Automated Fact-Checking
2026
Cited
0
Spurious-Aware Prototype Refinement for Reliable Out-of-Distribution Detection
2025
Cited
0
DeepFake Doctor: Diagnosing and Treating Audio-Video Fake Detection
2025
Cited
0
Diffusion Classifiers Understand Compositionality, but Conditions Apply
2025
Cited
0
Erased but Not Forgotten: How Backdoors Compromise Concept Erasure
2025
Cited
0
When To Solve, When To Verify: Compute-Optimal Problem Solving and Generative Verification for LLM Reasoning
2025
Cited
0
Load more
Resume (English only)
Academic Achievements
2025: 'Spurious-Aware Prototype Refinement for Reliable Out-of-Distribution Detection' accepted at NeurIPS 2025
2025: 'Classifiers Understand Compositionality, but Conditions Apply' accepted at NeurIPS 2025 D&B
2025: 'When To Solve, When To Verify...' accepted at COLM 2025
2025: Multimodal fact-checking method DEFAME accepted at ICML 2025
2024: Team won 1st place in AVERITEC Shared Task (FEVER Workshop @ EMNLP 2024)
2023: Recipient of DAGM German Pattern Recognition Award
2022: Team won Ego4D PNR Temporal Localization Challenge
Served as Area Chair for ICCV, CVPR, NeurIPS D&B across multiple years
Recognized as Outstanding Reviewer at CVPR and NeurIPS (2021–2022)
2024: Honored as Outstanding Area Chair at ECCV
Lead of the 'Reasonable Artificial Intelligence (RAI)' project, funded as a Cluster of Excellence
Background
Research at the intersection of vision and language
Focus on building explainable and compositional models
Dedicated to diagnosing and mitigating bias in AI systems
Developing multimodal models that can learn from language advice
Work spans image/video captioning, visual grounding, visual question answering, text-to-image synthesis, and multimodal forensics
Co-authors
28 total
Trevor Darrell
Professor of Computer Science, U.C. Berkeley
Marcus Rohrbach
Professor for Multimodal Reliable AI, TU Darmstadt, Germany
Bernt Schiele
Professor, Max Planck Institute for Informatics, Saarland University, Saarland Informatics Campus
Co-author 4
Kate Saenko
Boston University
Lisa Anne M Hendricks
DeepMind
Zeynep Akata
Professor at Technical University of Munich and Director at Helmholtz Munich
Ronghang Hu
Research Scientist, AI at Meta
×
Welcome back
Sign in to Agora
Welcome back! Please sign in to continue.
Email address
Password
Forgot password?
Continue
Do not have an account?
Sign up