Abstract: Speech emotion recognition (SER) is crucial for human–computer interaction (HCI), yet remains a challenge in cross-domain scenarios. Emotional expressions vary significantly across speakers, ...
Code for "Robust Table Retrieval Under Serialization Shift via Centroid-Aligned Adapters". The project measures how serialization format (CSV, TSV, HTML, Markdown, JSON, etc.) shifts table embeddings ...
Abstract: Image-text retrieval aims to align image regions with textual words for semantic matching, facilitating bidirectional retrieval between images and texts. While significant progress has been ...
The research introduces a novel memory architecture called MSA (Memory Sparse Attention). Through a combination of the Memory Sparse Attention mechanism, Document-wise RoPE for extreme context ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results