David T. Hoffmann

18 papers A* 5Journal 10Unranked 3
YearRankTypeTitle / Venue / Authors
2026 J jnl
No Hard Negatives Required: Concept Centric Learning Leads to Compositionality without Degrading Zero-shot Capabilities of Contrastive Models.
CoRR
Hai X. Pham, David T. Hoffmann, Ricardo Guerrero, Brais Martínez
2025 J jnl
CLIP Won't Learn Object-Attribute Binding from Natural Data and Here is Why.
CoRR
Bijay Gurung, David T. Hoffmann, Thomas Brox
2025 — conf
Common Data Properties Limit Object-Attribute Binding in CLIP.
DAGM GCPR
Bijay Gurung, David T. Hoffmann, Thomas Brox
2025 A* conf
Floxels: Fast Unsupervised Voxel Based Scene Flow Estimation.
CVPR
David T. Hoffmann, Syed Haseeb Raza, Hanqiu Jiang, Denis Tananaev, Steffen Klingenhoefer, Martin Meinke
2025 J jnl
Floxels: Fast Unsupervised Voxel Based Scene Flow Estimation.
CoRR
David T. Hoffmann, Syed Haseeb Raza, Hanqiu Jiang, Denis Tananaev, Steffen Klingenhoefer, Martin Meinke
2025 A* conf
Two Effects, One Trigger: On the Modality Gap, Object Bias, and Information Imbalance in Contrastive Vision-Language Models.
ICLR
Simon Schrodi, David T. Hoffmann, Max Argus, Volker Fischer, Thomas Brox
2025 — conf
Unlocking In-Context Learning for Natural Datasets Beyond Language Modelling.
DAGM GCPR
Jelena Bratulic, Sudhanshu Mittal, David T. Hoffmann, Samuel Böhm, Robin Tibor Schirrmeister, Tonio Ball, Christian Rupprecht, Thomas Brox
2024 A* conf
Eureka-Moments in Transformers: Multi-Step Tasks Reveal Softmax Induced Optimization Problems.
ICML
David T. Hoffmann, Simon Schrodi, Jelena Bratulic, Nadine Behrmann, Volker Fischer, Thomas Brox
2024 J jnl
Two Effects, One Trigger: On the Modality Gap, Object Bias, and Information Imbalance in Contrastive Vision-Language Representation Learning.
CoRR
Simon Schrodi, David T. Hoffmann, Max Argus, Volker Fischer, Thomas Brox
2023 J jnl
Eureka-Moments in Transformers: Multi-Step Tasks Reveal Softmax Induced Optimization Problems.
CoRR
David T. Hoffmann, Simon Schrodi, Nadine Behrmann, Volker Fischer, Thomas Brox
2022 A* conf
Ranking Info Noise Contrastive Estimation: Boosting Contrastive Learning via Ranked Positives.
AAAI
David T. Hoffmann, Nadine Behrmann, Juergen Gall, Thomas Brox, Mehdi Noroozi
2022 J jnl
Ranking Info Noise Contrastive Estimation: Boosting Contrastive Learning via Ranked Positives.
CoRR
David T. Hoffmann, Nadine Behrmann, Juergen Gall, Thomas Brox, Mehdi Noroozi
2021 A* conf
AGORA: Avatars in Geography Optimized for Regression Analysis.
CVPR
Priyanka Patel, Chun-Hao P. Huang, Joachim Tesch, David T. Hoffmann, Shashank Tripathi, Michael J. Black
2021 J jnl
AGORA: Avatars in Geography Optimized for Regression Analysis.
CoRR
Priyanka Patel, Chun-Hao P. Huang, Joachim Tesch, David T. Hoffmann, Shashank Tripathi, Michael J. Black
2020 J jnl
Learning Multi-human Optical Flow.
Int. J. Comput. Vis.
Anurag Ranjan, David T. Hoffmann, Dimitrios Tzionas, Siyu Tang, Javier Romero, Michael J. Black
2019 J jnl
Learning Multi-Human Optical Flow.
CoRR
Anurag Ranjan, David T. Hoffmann, Dimitrios Tzionas, Siyu Tang, Javier Romero, Michael J. Black
2019 — conf
Learning to Train with Synthetic Humans.
GCPR
David T. Hoffmann, Dimitrios Tzionas, Michael J. Black, Siyu Tang
2019 J jnl
Learning to Train with Synthetic Humans.
CoRR
David T. Hoffmann, Dimitrios Tzionas, Michael J. Black, Siyu Tang