Search papers, labs, and topics across Lattice.
3
0
4
4
A striking task-dependent robustness gap reveals that while ASR thrives on direct audio retrieval, AQA falters due to bottlenecks in mediated information access.
Performance improvements in speech model fine-tuning are often an illusion, heavily reliant on the specific pretrained instance rather than true methodological advancements.
Genre biases in music evaluation models can lead to significant misalignments with human preferences, but a new training approach can correct this skew.
