The paper investigates how audio-language models react to local sound‑event edits by comparing matched same‑class and different‑class replacements. Using consistent editing parameters across 600 synthetic soundscapes and three models, the study finds that different‑class replacements cause 49.7 to 60.9 percentage points more answer changes, while same‑class replacements keep accuracy within 5.2 percentage points of the original. Soundscape‑weighted selective accuracy for the inserted class ranges from 73.7 to 81.9%.
The paper introduces TRACE, a method for auditing large audio‑language models (LALMs) by deleting the queried source and each rival sound in turn to test whether views containing the target sound consistently score higher than those without it. Across 8 LALMs and 3,456 controlled mixtures, 6.3% of correct answers fail this ordering, a failure that a simple target‑only score check misses 42.5% of the time. The failure rate rises from 4.9% to 8.4% as the target signal‑to‑noise ratio drops from +5 to –5 dB, and the issue persists even under loudness and re‑pairing controls.
By Yi Zhang, Yi Li, Hongwei Du
arXiv:2608.27817v4 Announce Type: replace-cross
Abstract: Speech and audio LLMs are evaluated by comparing waveform predictions with predictions from an automatic speech recognition (ASR) transcript....
By Mengzhe Geng
The paper introduces SAC‑Bench, a benchmark of 616,031 audio clips that test audio‑language models (ALMs) on how they handle conflicts between spoken claims and acoustic evidence. It covers 188 sound‑event classes across seven semantic, temporal, and spatial tasks, pairing environmental recordings with either compatible or contradictory synthesized spoken claims and including audio‑only controls. Evaluation of five open‑weight ALMs shows that high audio‑only accuracy often coexists with significant drops in performance under conflicting claims, revealing gaps in conflict robustness and the mixed effects of prompt interventions.
By Wenbin Wang, Yuheng Li, Koichiro Yoshino
The paper investigates how to evaluate audio‑language models by separating the use of acoustic evidence from the need to invoke a generative audio model. Using a controlled call‑decision framework, the authors compare policies that rely on transcript labels, encoder outputs from CLAP, AST, or WavLM, and optional calls to generative models such as Qwen2‑Audio, Qwen2.5‑Omni, or MOSS‑Audio. Results on the VocalSound dataset show that while transcript‑only accuracy is low (0.296), encoder‑only controls achieve high accuracy (≈0.85) without any generative calls, and adding generative calls yields only a marginal improvement (0.925 vs. 0.921).
By Mengzhe Geng
AnchorPrompt is an adaptation technique for large audio‑language models that keeps the base model frozen and learns a single block of prompt vectors inserted at the decoder input. By training these prompts through self‑distillation on diverse audio and text perturbations, the method improves answer consistency and reduces hallucinations across multiple benchmarks. The approach is perturbation‑agnostic at inference, enabling zero‑shot transfer to unseen distortions such as reverberation and choice permutations.
By Pooneh Mousavi, Amir Ivry, Mirco Ravanelli, Cem Subakan
A large audio-language model (LALM) can answer "is a cat present?" about a mixture correctly, and its confidence can drop when the cat is removed, yet this does not show that it can tell the cat's pre...
We benchmark eleven audio classification methods: five task-aware closed-set LLMs (four Gemini models plus open-weight Kimi-Audio-7B-Instruct), four fixed-vocabulary taggers (YAMNet, PANNs, Whisper-AT, and SSLAM), a zero-shot audio-text model (CLAP), and an audio-grounded LLM (BAT). We evaluate them on a closed-set sound-source identification task over 2,242 clips spanning 23 fine-grained classes and 11 categories.
The paper investigates how to evaluate generative audio large language models (Audio‑LLMs) on known closed‑set tasks by separating the decision to call a generative model from the use of acoustic evidence. It introduces a controlled call‑decision framework where a policy can choose between a transcript label, encoder evidence from CLAP, AST, or WavLM, or a generative call to Qwen2‑Audio, Qwen2.5‑Omni, or MOSS‑Audio, and measures the impact of generative calls on accuracy. Results on the VocalSound dataset show that while transcript‑only accuracy is low (0.296), encoder‑based controls achieve high accuracy (≈0.85) without any generative calls, and adding generative calls yields only a marginal improvement (0.925 vs. 0.921).
By Mengzhe Geng
The paper investigates how large audio‑language models (LALMs) hallucinate objects by affirming sounds that are not present. Using a black‑box causal test, the authors isolate and remove the claimed evidence region via spectral gating and compare the effect against a random‑window control across a 2×2×2 grid on three LALMs. The study finds that localization of evidence improves detection over random removal, that raw detection values are model‑dependent, and that contrastive decoding combining blank and language‑prior contrasts can significantly boost accuracy, but a unified causal‑grounding profile does not transfer between models.
By Mohammad Khalooei, Mohammad Sabokrou
arXiv:2610.16942v1 Announce Type: cross
Abstract: Large audio-language models (LALMs) are fast becoming a general interface to sound, yet how far their broad pretraining carries into specialized doma...
By Zhe Sun, Liang An, Jiayu Yao, Ronglai Qian, Jianing Cai, Ziwei Huang, Qing Fan, Zhiyuan Yang, Yujun Cai, Hongli Cao, Yiwei Wang, Xiaoyan Wang
The paper investigates how Audio Large Language Models (Audio LLMs) actually use audio input to determine answers, rather than relying on textual cues. It finds that replacing audio with silence or unrelated audio degrades performance more after training than before, that acoustic information shapes representations in early-to-middle layers and influences final predictions in middle-to-late layers, and that training impacts specific layer bands most strongly. These observations offer a mechanistic view of how training enhances the use of acoustic evidence in Audio LLMs.
By Hyebin Cho, Suho Yoo, Jihoo Jung, Joon Son Chung