Semantic-level and verification-based uncertainty methods outperform token-level baselines for audio reasoning in ALLMs, but their relative performance on hallucination and unanswerable-question benchmarks is model- and task-dependent.
Building a taiwanese mandarin spoken language model: A first attempt
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
citation-role summary
background 2
citation-polarity summary
verdicts
UNVERDICTED 6roles
background 2polarities
background 2representative citing papers
AQUA-Bench evaluates audio QA models on three unanswerability scenarios: missing correct answers, mismatched choice sets, and questions irrelevant to the audio.
Mamba-based HuBERT models match or exceed Transformer versions on speech tasks while using far less compute for long sequences and streaming ASR.
Audio-language models retain 60-72% of benchmark scores without audio, and most audio-dependent items can be solved from short fragments rather than full clips.
ASPIRin decouples speaking timing from token content via binary action space projection and applies GRPO with rule-based rewards to optimize interactivity in SLMs without semantic collapse or repetition.
TW-Sound580K dataset plus Tai-LALM model with dynamic Dual-ASR arbitration lifts localized Taiwanese audio-language accuracy to 49.1% on the TAU benchmark.
citing papers explorer
-
Walking Through Uncertainty: An Empirical Study of Uncertainty Estimation for Audio-Aware Large Language Models
Semantic-level and verification-based uncertainty methods outperform token-level baselines for audio reasoning in ALLMs, but their relative performance on hallucination and unanswerable-question benchmarks is model- and task-dependent.
-
AQUA-Bench: Beyond Finding Answers to Knowing When There Are None in Audio Question Answering
AQUA-Bench evaluates audio QA models on three unanswerability scenarios: missing correct answers, mismatched choice sets, and questions irrelevant to the audio.
-
An Exploration of Mamba for Speech Self-Supervised Models
Mamba-based HuBERT models match or exceed Transformer versions on speech tasks while using far less compute for long sequences and streaming ASR.
-
All That Glitters Is Not Audio: Rethinking Text Priors and Audio Reliance in Audio-Language Evaluation
Audio-language models retain 60-72% of benchmark scores without audio, and most audio-dependent items can be solved from short fragments rather than full clips.
-
ASPIRin: Action Space Projection for Interactivity-Optimized Reinforcement Learning in Full-Duplex Speech Language Models
ASPIRin decouples speaking timing from token content via binary action space projection and applies GRPO with rule-based rewards to optimize interactivity in SLMs without semantic collapse or repetition.
-
TW-Sound580K: A Regional Audio-Text Dataset with Verification-Guided Curation for Localized Audio-Language Modeling
TW-Sound580K dataset plus Tai-LALM model with dynamic Dual-ASR arbitration lifts localized Taiwanese audio-language accuracy to 49.1% on the TAU benchmark.