3 papers
cs.LG2026
Audio-Visual Continual Test-Time Adaptation without Forgetting
Sarthak Kumar Maharana, Akshay Mehra, Bhavya Ramakrishna +2
Audio-visual continual test-time adaptation involves continually adapting a source audio-visual model at test-time, to unlabeled non-stationary domains, where either or both modali…
cs.CV2025
DPAR: Dynamic Patchification for Efficient Autoregressive Visual Generation
Divyansh Srivastava, Akshay Mehra, Pranav Maneriker +5
Decoder-only autoregressive image generation typically relies on fixed-length tokenization schemes whose token counts grow quadratically with resolution, substantially increasing t…
cs.LG2025
Coreset Selection via LLM-based Concept Bottlenecks
Akshay Mehra, Trisha Mittal, Subhadra Gopalakrishnan +1
Coreset Selection (CS) aims to identify a subset of the training dataset that achieves model performance comparable to using the entire dataset. Many state-of-the-art CS methods se…