发表机构
Centific AI Research（Centific 人工智能研究院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对现有音频推理基准偏重英语和通用领域的问题，提出SEABED，一个基于真实东南亚语音语料构建的六任务问答基准，含5,404个问答对，评估显示最先进模型Gemini 3.5 Flash仅达50.3%加权平均准确率，并额外检验推理是否基于音频证据。
AI 中文摘要
现代音频语言模型的评判标准不再仅限于其转录文字的能力，而在于其能否对所听到的内容进行推理：从语气和韵律中恢复意义，区分方言和地区语言，并解决书面形式留下的歧义。这一能力现由不断增长的音频推理基准系列来衡量，但几乎全部集中在英语和通用领域音频上。东南亚（SEA）地区则使用继承自英语任务分类法（包括识别、翻译和副语言分类）的基准，这些基准测试的是模型能否听到东南亚语音，而非能否从中进行推理。我们推出了SEABED，一个以音频优先的问答数据集，旨在对语言和音频推理模型在东南亚语音上的表现进行基准测试。SEABED包含一套完全基于真实、公开可用的东南亚语音语料库构建的六项音频推理任务，共产生5,404个问答对。我们评估了六个前沿及区域特定的音频大语言模型：即使是最先进的模型Gemini 3.5 Flash也仅达到50.3%的加权平均准确率。SEABED不仅评估答案的准确性，还评估模型所述推理是否基于音频证据。基准数据的样本可在此处获取：此https URL。
英文摘要
Modern audio-language models are no longer judged only on what words they can transcribe, but on whether they can reason over what they hear: recovering meaning that lives in tone and prosody, telling dialects and regional languages apart, and resolving ambiguity that the written form leaves open. This capability is now measured by a growing family of audio-reasoning benchmarks, but almost entirely in English and on general-domain audio. Southeast Asia (SEA) is served instead by benchmarks that inherit an English task taxonomy of recognition, translation, and paralinguistic classification, and therefore test whether a model hears SEA speech rather than whether it can reason from it. We introduce SEABED, an audio-first question answering dataset designed to benchmark language and audio reasoning models on SEA speech. SEABED comprises a suite of six audio-reasoning tasks built entirely from real, openly available SEA speech corpora, yielding 5,404 question-answer pairs. We evaluate six frontier and region-specific audio LLMs: even the state-of-the-art model Gemini 3.5 Flash achieves only 50.3% weighted average accuracy. SEABED evaluates not only answer accuracy, but also whether models' stated reasoning is grounded in the audio evidence. A sample of the benchmark data is available here: https://huggingface.co/datasets/CentificAIResearch/SEABED.
CommentsAccepted at SALMA[EMNLP 2026] Workshop
