PaperHub
影响力指数
87.24/100
前 0.8%
全站排名 #501
发表论文34 篇
平均评分5.1
年均产出11.3 篇/年

Zeynep Akata

Full Professor@Technische Universität München·德国·OpenReview
研究方向

Explainable Machine Learning · Generative AI · Vision and Language · Zero-Shot Learning

6.0
19

Post-hoc Probabilistic Vision-Language Models

ICLR 2026Poster
5.5
18

Person-Centric Annotations of LAION-400M: Auditing Bias and Its Transfer to Models

ICLR 2026Poster
通讯
5.0
22

DD-Ranking: Rethinking the Evaluation of Dataset Distillation

ICLR 2026Rejected
4.5
15

Are Reasoning LLMs Robust to Interventions on their Chain-of-Thought?

ICLR 2026Poster
三作
4.5
19

TimeSAE: Sparse Decoding for Faithful Explanations of Black-Box Time Series Models

ICLR 2026Withdrawn
通讯
4.5
16

Reference-Free Rating of LLM Responses via Latent Information

ICLR 2026Rejected
通讯
4.0
22

When We Don’t See the Same Picture: Aligning Agents with Divergent Visual Spaces

ICLR 2026Rejected
通讯
4.0
18

Stitch: Training-Free Position Control in Multimodal Diffusion Transformers

ICLR 2026Rejected
通讯
4.0
26

Towards understanding multimodal in-context learning

ICLR 2026Rejected
通讯
4.0
19

Beyond the Final Layer: Attentive Multi-Layer Fusion for Vision Transformers

ICLR 2026Rejected
3.5
7

Aligning Signal Leakage Matters for Synthetic Data Generation of Satellite Imagery

ICLR 2026Withdrawn
通讯
3.5
19

Training-free Uncertainty Guidance for Complex Visual Tasks with MLLMs

ICLR 2026Rejected
通讯
3.0
17

What is the Color of RED? Vision–Language Models Prefer to Read Rather Than See

ICLR 2026Rejected
三作
2.7
7

A Large Scale Analysis of Gender Biases in Text-to-Image Generative Models

ICLR 2026Withdrawn
通讯
2.7
21

TiViT: Time Series Representations Lie Hidden in Pretrained Vision Transformers

ICLR 2026Rejected
通讯
2.5
5

Towards Automatic Discovery and Explanation of Differences Between Vision Models

ICLR 2026Withdrawn
三作
7.8
18

Noise Hypernetworks: Amortizing Test-Time Compute in Diffusion Models

NeurIPS 2025Poster
通讯
7.5
27

Concept-Guided Interpretability via Neural Chunking

NeurIPS 2025Poster
通讯
7.3
21

A Large Scale Analysis of Gender Biases in Text-to-Image Generative Models

NeurIPS 2025Rejected
通讯
7.0
9

WikiBigEdit: Understanding the Limits of Lifelong Knowledge Editing in LLMs

ICML 2025Poster
7.0
17

Sparse Autoencoders Learn Monosemantic Features in Vision-Language Models

NeurIPS 2025Poster
通讯
7.0
21

Revealing and Reducing Gender Biases in Vision and Language Assistants (VLAs)

ICLR 2025Poster
通讯
6.5
24

Building, Reusing, and Generalizing Abstract Representations from Concrete Sequences

ICLR 2025Poster
6.0
20

Scalable Ranked Preference Optimization for Text-to-Image Generation

ICLR 2025Rejected
三作
5.5
27

Disentangled Representation Learning with the Gromov-Monge Gap

ICLR 2025Poster
5.3
24

Decoupling Angles and Strength in Low-rank Adaptation

ICLR 2025Poster
三作
5.3
22

Adapting Communicating MLLMs on the Fly in Referring Expression Tasks

ICLR 2025Rejected
通讯
4.7
27

From Bulk to Budget: Best Practices To Compress Multimodal Large Language Models

ICLR 2025Rejected
通讯
4.7
13

Dual-Stream Adapters for Anomaly Segmentation

ICLR 2025Rejected
合作者 (20)