Zeynep Akata
Professor
Develops explainable machine learning that combines vision and language so models can justify their decisions, including zero-shot and few-shot learning with textual explanations.
Zeynep Akata's lab at TUM works on trustworthy & safe ML and human-AI collaboration & trust. 69 papers since 2023, 56 at top venues, mostly ICLR and ICML. 12 PhD students and 2 master's students. 18 alumni, most often to Apple and Bosch.
At a glance
- papers, 2023–2026
- 18/ year
- at T1 venues
- 81%
- PhD students
- 12
- MS students
- 2
- alumni listed
- 18
Papers by yearfrom the lab homepage, 2023–2026
- 23 papers2026 · T1 21 · collecting
- 14 papers2025 · T1 14
- 13 papers2024 · T1 11
- 19 papers2023 · T1 10
T1 venuesOtherVenues: ICLR, ICML, NeurIPS, CVPR, ICCV
Research focuspaper titles that use each topic's words
- zero-shot4 papers
- explanations3 papers
- few-shot learning1 paper
- Trustworthy & safe ML0 papers
- Human-AI collaboration & trust0 papers
Paper list
From the lab homepage.
202623 papers
- TMLR (T1)A Systematic Study of In-the-Wild Model Merging for Large Language Models
- NeurIPS (T1)APM: Evaluating Style Personalization in LLMs with Arbitrary Preference Mappings
- ICLR (T1)Are Reasoning LLMs Robust to Interventions on Their Chain-of-Thought?
- ICML (T1)Attentive Multi-Layer Fusion for Vision Transformers
- NeurIPS (T1)CapTrack: Multifaceted Evaluation of Forgetting in LLM Post-Training
- ICML (T1)Diamond Maps: Efficient Reward Alignment via Stochastic Flow Maps
- ICML (T1)Dissecting Multimodal In-Context Learning: Modality Asymmetries and Circuit Dynamics in modern Transformers
- Nature Communications MedicineEvaluating large language models for diagnostic reasoning from unstructured clinical narratives in epilepsy
- CVPR (T1)Explaining CLIP Zero-shot Predictions Through Concepts
- CVPR (T1)FINER: MLLMs Hallucinate under Fine-grained Negative Queries
- ECCV (T1)From Drop-off to Recovery: A Mechanistic Analysis of Segmentation in MLLMs
- ICML (T1)Mantis: Lightweight Foundation Model for Time Series Classification
202514 papers
- ICLR (T1)Building, Reusing, and Generalizing Abstract Representations from Concrete Sequences
- NeurIPS (T1)Concept-Guided Interpretability via Neural Chunking
- CVPR (T1)Context-Aware Multimodal Pretraining
- CVPR (T1)COSMOS: Cross-Modality Self-Distillation for Vision Language Pre-training
- ICLR (T1)Decoupling Angles and Strength in Low-rank Adaptation
- ICLR (T1)Disentangled Representation Learning with the Gromov-Monge Gap
- CVPR (T1)FLAIR: VLM with Fine-grained Language-informed Image Representations
- CVPR (T1)How to Merge Your Multimodal Models Over Time?
- NeurIPS (T1)Noise Hypernetworks: Amortizing Test-Time Compute in Diffusion Models
- ICLR (T1)Revealing and Reducing Gender Biases in Vision and Language Assistants (VLAs)
- ICCV (T1)Scalable Ranked Preference Optimization for Text-to-Image Generation
- NeurIPS (T1)Sparse Autoencoders Learn Monosemantic Features in Vision-Language Models
202413 papers
- NeurIPS (T1)A Practitioner's Guide to Continual Multimodal Pretraining
- ECCV (T1)DataDream: Few-shot Guided Dataset Generation
- ECCV (T1)EgoCVR: An Egocentric Benchmark for Fine-Grained Composed Video Retrieval
- ICML (T1)ETHER: Efficient Finetuning of Large-Scale Models with Hyperplane Reflections
- ICLR (T1)Fantastic Gains and Where to Find Them - On the Existence and Prospect of General Knowledge Transfer
- ECCV (T1)Geometry Fidelity for Spherical Images
- PNAS (T1)How should the advent of large language models affect the practice of science?
- ECCV (T1)Improving Intervention Efficacy via Concept Realignment in Concept Bottleneck Models
- Conference on Lifelong Learning Agents, CoLLAsReflecting on the State of Rehearsal-Free Continual Learning with Pretrained Models
- NeurIPS (T1)ReNO: Enhancing One-step Text-to-Image Models through Reward-based Noise Optimization
- ICML WorkshopSemioLLM: Assessing Large Language Models for Semiological Analysis in Epilepsy Research
- ICLR (T1)Unbalancedness in Neural Monge Maps Improves Unpaired Domain Translation
202319 papers
- TMLR (T1)Addressing caveats of neural persistence with deep graph persistence
- CVPR (T1)Bridging the Gap between Model Explanations in Partially Annotated Multi-label Classification
- 'Methods in Ecology and Evolution' published by John Wiley & Sons Ltd on behalf of British Ecological SocietyClassifying the unknown: Insect identification with deep hierarchical Bayesian learning
- DAGM German Conference on Pattern Recognition, GCPRDeViL: Decoding Vision Features into Language
- ICLR (T1)Disentanglement of Correlated Factors via Hausdorff Factorized Support
- ICCV (T1)Image-free Classifier Injection for Zero-Shot Classification
- Neural Information Processing Systems, NeurIPSIn-Context Impersonation Reveals Large Language Models' Strengths and Biases
- ICCV (T1)Iterative Superquadric Recomposition of 3D Objects from Multiple Views
- NeurIPS (T1)Meta-in-context learning in large language models
- ICCV (T1)PDiscoNet: Semantically consistent part discovery for fine-grained recognition
- ICCV (T1)ProbVLM: Probabilistic Adapter for Frozen Vision-Language Models
- DAGM German Conference on Pattern Recognition, GCPRText-to-feature diffusion for audio-visual few-shot learning
Where alumni went18 listed, 91% to industry
Industry 10Academia 1
Most often to
- Apple
- Bosch
- Google DeepMind
Also to
- ETH Zurich
Apply to TUMRolling
Not announced yet
Computer Science Doctorate (CIT)
- Round
- Rolling (position-based)
- Opens
- —
- GRE
- See program page
- English
- Very good written and spoken English; no minimum test score
- Fee
- See program page
Confirmed on the program page.
Similar labsnearest by research summary
- USMichigan
Atul Prakash
- Trustworthy & safe ML
- USMIT
Ashia Wilson
- Learning theory & optimization
- GBCambridge
Ferenc Huszár
7 T1 - CHETH Zurich
Martin Vechev
- Trustworthy & safe ML
- CNZhejiang
Yueting Zhuang
- Knowledge, QA & retrieval
- USNYU
Allen Liu
- 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.