A. Sophia Koepke
@askoepke.bsky.social
Junior research group leader at TUM | University of Tübingen. Currently at BAIR (Berkeley). Previously at VGG (Oxford). Interested in multi-modal learning.
🔗 https://akoepke.github.io/
🔗 https://akoepke.github.io/
Thanks to Daniil Zverev*, @thwiedemer.bsky.social*, @bayesiankitten.bsky.social, Matthias Bethge (@bethgelab.bsky.social), and @wielandbrendel.bsky.social for making VGGSound sounder! 🙌 🎉 🐗
October 21, 2025 at 6:08 PM
Thanks to Daniil Zverev*, @thwiedemer.bsky.social*, @bayesiankitten.bsky.social, Matthias Bethge (@bethgelab.bsky.social), and @wielandbrendel.bsky.social for making VGGSound sounder! 🙌 🎉 🐗
📊 With VGGSounder, we show that existing models don’t always benefit from multimodal input and sometimes performance even degrades.
Code and data: vggsounder.github.io
Code and data: vggsounder.github.io
VGGSounder: Audio-Visual Evaluations for Foundation Models
VGGSounder, a multi-label audio-visual classification dataset with modality annotations.
vggsounder.github.io
October 21, 2025 at 6:07 PM
📊 With VGGSounder, we show that existing models don’t always benefit from multimodal input and sometimes performance even degrades.
Code and data: vggsounder.github.io
Code and data: vggsounder.github.io
VGGSounder is a new video classification benchmark for audio-visual foundation models:
We provide:
📢 Re-annotated VGGSound test set
📢 Modality-specific manual labels
📢 A modality confusion metric to diagnose when models misuse modalities
Paper: arxiv.org/pdf/2508.08237
We provide:
📢 Re-annotated VGGSound test set
📢 Modality-specific manual labels
📢 A modality confusion metric to diagnose when models misuse modalities
Paper: arxiv.org/pdf/2508.08237
October 21, 2025 at 6:06 PM
VGGSounder is a new video classification benchmark for audio-visual foundation models:
We provide:
📢 Re-annotated VGGSound test set
📢 Modality-specific manual labels
📢 A modality confusion metric to diagnose when models misuse modalities
Paper: arxiv.org/pdf/2508.08237
We provide:
📢 Re-annotated VGGSound test set
📢 Modality-specific manual labels
📢 A modality confusion metric to diagnose when models misuse modalities
Paper: arxiv.org/pdf/2508.08237
Thanks to @munichcenterml.bsky.social for supporting the workshop with a best paper award (announced at 2.50pm CDT)!
June 11, 2025 at 5:59 PM
Thanks to @munichcenterml.bsky.social for supporting the workshop with a best paper award (announced at 2.50pm CDT)!
We have fantastic speakers, including @saining.bsky.social, @aidanematzadeh.bsky.social, @ranjaykrishna.bsky.social, Ludwig Schmidt, @lisadunlap.bsky.social, and Ishan Misra.
June 11, 2025 at 5:57 PM
We have fantastic speakers, including @saining.bsky.social, @aidanematzadeh.bsky.social, @ranjaykrishna.bsky.social, Ludwig Schmidt, @lisadunlap.bsky.social, and Ishan Misra.