Accepted Papers

Our workshop includes a proceedings/archival track and a community/non-archival track. We are excited to host poster and lightning talk presentations of all contributed papers in both tracks.

• For the poster session each paper has a number in 182–193.

• For the lightning talks each paper has one slide in one minute in the ⚡️ lightning talk deck.

We are thrilled to announce two best papers, once for each track, which were rewarded with Theo’s chocolate in-person in Seattle:

• 🏆 Proceedings Best Paper: Test-time Specialization of Dynamic Neural Networks by Leroux et al.

• 🏆 Community Best Paper: Persistent Test-time Adaptation in Recurring Testing Scenarios by Hoang et al.

Proceedings

• P1. (#182). MixStyle-Based Contrastive Test-Time Adaptation: Pathway to Domain Generalization.

  Kota Yamashita & Kazuhiro Hotta.

• P2. (#183). Fully Test-time Adaptation for Object Detection.

  Xiaoqian Ruan & Wei Tang.

• P3. (#184). Test-time Specialization of Dynamic Neural Networks. (poster)

  Sam Leroux, Dewant Katare, Aaron Ding, Pieter Simoens.

• P4. (#185) ST2ST: Self-Supervised Test-time Adaptation for Video Action Recognition.

  Masud An-Nur Islam Fahim, Mohammed Innat, Jani Boutellier.

• P5. (#186) Unknown Sample Discovery for Source Free Open Set Domain Adaptation. (poster)

  Chowdhury Sadman Jahan, Andreas Savakis.

Community

• C1. (#187) Test-time Adaptation for Regression by Subspace Alignment. (poster)

  Kazuki Adachi, Shin’ya Yamaguchi, Atsutoshi Kumagai.

• C2. (#188) Just Shift It: Test-Time Prototype Shifting for Zero-Shot Generalization with Vision-Language Models. (poster)

  Elaine Sui, Xiaohan Wang, Serena Yeung-Levy.

• C3. (#189) Persistent Test-time Adaptation in Recurring Testing Scenarios. (poster)

  Trung-Hieu Hoang, MinhDuc Vo, Minh N. Do.

• C4. (#190) OT-VP: Optimal Transport-guided Visual Prompting for Test-Time Adaptation.

  Yunbei Zhang, Akshay Mehra, Jihun Hamm.

• C5. (#191) BECoTTA: Input-dependent Online Blending of Experts for Continual Test-time Adaptation.

  Daeun Lee, Jaehong Yoon, Sung Ju Hwang.

• C6. (#192) Adaptive Randomized Smoothing for Certified Multi-Step Defence.

  Shadab Shaikh, Saiyue Lyu, Frederick Shpilevskiy, Evan Shelhamer, Mathias Lécuyer.

• C7. (#193) A Lost Opportunity for Vision-Language Models: A Comparative Study of Online Test-time Adaptation for Vision-Language Models.

  Mario Döbler, Robert A. Marsden, Tobias Raichle, Bin Yang.