TL;DR
It is concluded that genuine multi-speaker dialogue inference remains an unsolved problem for current LLMs under standard prompting strategies.
Experiments with six LLAMA and MISTRAL models show that LoRA substantially improves dialogue generation while requiring only 25-50% of the training data to recover over 90% of the final performance gains, and zero-shot evaluation with frontier LLMs shows that TalkFa remains a challenging benchmark.
Cross-lingual probe transfer and direct comparison of probe weights show that NLI representations are most alignable in the middle layers: probes transfer best there, and probes trained independently on different languages converge to similar weight vectors, peaking mid-network.
This work proposes Spoken Function Calling (SFC), a novel semantic understanding perspective that optimizes semantic understanding with structured rule definitions, to evolve beyond traditional closed-set SLU.
TRILOGUE (TRIlingual spoken diaLOGUE fact-checking) is introduced, a large-scale trilingual benchmark of source-grounded spoken dialogues in English, Russian, and Kazakh that supports claim check-worthiness detection, source-article evidence retrieval, and claim verification with claim-only, gold-evidence, and retrieve...
EmoLASP's LLM pipeline demonstrates the potential advantages of using a reasoning approach to ensure emotion prediction consistency and to reduce both the cost of fine-tuning and the cost of prompting with long dialogue histories.
MEUSLI is introduced, the first open-science multilingual projector family that links a Whisper encoder with open-source multilingual LLMs, enabling fully open-source end-to-end ASR in 28 European languages.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
