Speech-LLMs often exhibit prompt overfitting, where models solely trained on automatic speech recognition (ASR) instruction fail to generalize to new instructions such as speech translation and continue to behave primarily as ASR system. We propose DirectSpeech2LLM, a simple end-to-end framework that preserves the inst...
Hemant Yadav, Sunayana Sitaram, R. Zimmermann et al.· 0 citations
A wrong verdict does not mean the judge lacks the information, and a simple diagnostic shows when it is worth recovering, and on rubric tasks that score one answer at a time, reading the internals gives no advantage.
This work introduces interactional cultural markers, measurable patterns of doctor-patient interaction grounded in cross-cultural clinical communication, and uses them to compare real, simulated, and synthetic consultations from Indian and US clinical contexts to find distinct patterns of participation and control.
Krithi Shailya, Siddharth D. Jaiswal, A. Makani et al.· 0 citations
This work introduces Pluralis v0.1, a novel multimodal, multi-regional, and multilingual dataset built from a culture-first perspective and calls upon the research community to utilize this foundation to advance the science of multilingual, multicultural evaluation to better support AI cultural alignment globally.
Alicia Parrish, Rajat C. Shinde, Sanket Badhe et al.· arXiv.org· 0 citations
A single trace-extraction regex, not the model, manufactured a multilingual failure: two worked examples raise one model's measured accuracy twenty-sixfold while its accuracy on readable outputs barely moves.
