Showing 2026Show all
2 papers · 1 filter
cs.SD2026
ParaPairAudioBench: Paralinguistic Pairwise Audio Benchmark for LALM-as-a-Judge
Jisu Jeon, Seungyeon Jwa, Joosung Lee +6
Large Audio-Language Models (LALMs) have been widely used as judge models for the automatic evaluation of generated speech. However, prior approaches predominantly focus on holisti…
cs.CL2026
The Lexical Imprint of Pre-training Data
Woojin Chung, Jeonghoon Kim, Woomin Song +3
How much of zero-shot benchmark accuracy is captured by word-level alignment between a pre-training corpus and the benchmark target? We study this question through word-level unigr…