Audio QA

Audio decisions often depend on evidence that a transcript does not preserve, while their probability estimates can depend on how answer options are ordered. AudioJev maps a waveform, question and supplied alternatives directly to a candidate distribution through one shared full-parameter model. We define order calibration as preserving an answer's probability under meaning-preserving option permutations. Random-derangement SKL training pairs each question with a reordered view in which every alternative changes position, supervises both answers, and aligns the two distributions before applying a symmetric KL penalty. Inference retains a single candidate-scoring forward, with no calibration head or order ensemble. Across three training seeds, AudioJev reaches 68.88%/55.33% mean accuracy on complete MMAU/MMAR and reduces random-order SKL by 43.8%/60.5% relative to the single-view removal ablation. The same model handles intent, environmental sound, note properties, speech activity and conversational transitions. Paired removal ablations and multi-order evaluation measure predictive accuracy and probability stability together, establishing a direct audio interface whose calibration objective acts on candidate meaning rather than presentation position. Inference code and model weights are available at https://github.com/SihanLv/AudioJev-Inference and https://huggingface.co/shlv/AudioJev.

Embodied assistants in domestic environments must infer what happened, where and when it occurred, and how to respond. To address this, we introduce RMS-AQA, a spatial audio question answering (SAQA) benchmark for real-world domestic environments. The benchmark features a two-stage question-answering (QA) format to comprehensively assess the ability of audio-language models (ALMs) to first ground audible sound events and subsequently perform complex spatio-temporal reasoning based on that grounding. To maximize acoustic realism, our dataset combines authentic real-world first-order Ambisonics (FOA) recordings with high-fidelity synthetic data generated using measured room impulse responses (RIRs). Furthermore, we provide a lightweight spatial plug-in that injects FOA-format data into frozen audio-language backbones. Experimental results reveal that the primary challenges stem from concurrent sources, far distance, and sim-to-real domain gap between RIR-synthesized and authentic recordings.

Large audio-language models (LALMs) perform strongly on individual audio tasks, but whether these capabilities can be reliably composed remains underexplored. We conduct a controlled diagnostic study of capability composition in LALMs, requiring models to integrate audio-attribute recognition, cue-conditioned segment selection, and downstream ASR or question answering. We construct two-utterance inputs with distinct acoustic cues to evaluate composition across environmental sound, gender, and emotion cues, with ASR, Math QA, and Factual QA as downstream tasks. Across four open-source LALMs, compositional QA accuracy decreases in 39 of 40 model-task-cue settings, by an average of 26.7 percentage points. ASR exhibits a similarly consistent degradation, with WER increasing in 39 of 40 settings by an average of 28.5 percentage points, while the magnitude of degradation varies across models, cue types, and cue salience. We further probe these failures through output format, positional preference, and chain-of-thought (CoT) analyses. Our study reveals a systematic gap between possessing individual audio capabilities and reliably composing them.

Omni-modal large language models (LLMs) are expected to answer a question using the modality it explicitly refers to. However, existing training paradigms rarely verify whether models actually follow this modality, because multimodal inputs from the same sample often provide redundant evidence for the same answer. In this work, we uncover a pervasive cross-modal shortcut in omni-modal LLMs: when asked an audio-related question, models rely on the image as much as on the audio, and sometimes even more. To systematically diagnose this behavior, we introduce the Factorized Modality Diagnostic, which independently swaps audio and images between samples to isolate each modality's causal contribution. Across two model families in different settings, we find that this shortcut persists throughout supervised fine-tuning and reinforcement learning post-training, while judge-based RL may further amplify such reliance on irrelevant visual information. Based on this finding, we propose DMC-Repair, which trains models on the same kind of cross-modal swapped samples while assigning supervision according to the modality specified by the question. This prevents models from exploiting the spurious correspondence between modalities within the same clip. Experiments demonstrate that DMC-Repair reduces the image-induced share of the answer effect by 59.9%, effectively suppressing the cross-modal shortcut without compromising audio-question answering performance. The reduction in shortcut reliance generalizes across two model families and zero-shot to an unseen dataset and an unseen benchmark, and persists through subsequent post-training. Code is available at https://anonymous.4open.science/r/DMC-Repair.

Multi-turn, multi-audio temporal question answering requires models to track target events across follow-up questions, recording switches, and historical references, recovering complete instances and their boundaries for temporal calculation and comparison. We propose TEMA, which connects event perception with evidence-based answering through Route, specifying the audio scope, and Span, describing all relevant intervals as conditional audio captions. We construct TEMA-Dialog with 40,704 dialogs and per-turn evidence and answer supervision, and TEMA-Bench for joint evaluation of evidence and final answers. Training combines temporal grounding initialization, full-dialog supervised fine-tuning, and completeness-first Span-only GRPO. Experiments on Qwen2.5-Omni and AF-Next show improved temporal question answering, particularly event localization and cross-audio comparison. Reinforcement learning applied solely to evidence further improves interval recovery and answer accuracy.

Audio-language models can produce confident answers unsupported by the audio, motivating uncertainty estimates that identify unreliable responses. We compare probability-based, sampling-based, self-verification, evidential, and contrastive measures across four open-weight models and five audio QA benchmarks. In multiple-choice evaluation, first-token measures are strongest overall, with top-1 probability achieving a mean AUROC of .740, compared with .708 for ten-sample discrete semantic entropy, while requiring no additional model calls. Across four benchmarks, shifting from multiple-choice to open-ended evaluation lowers mean accuracy from 57.6% to 36.6%, yet uncertainty remains predictive of errors: semantic entropy, maximum token entropy, and semantic agreement achieve mean AUROCs of .697, .694, and .693, respectively. To test whether uncertainty reflects the evidence available to answer the question, we perform input ablations that remove either the audio or the question. Across top-1 confidence, entropy, and sampling-based measures, removing audio reduces error-detection AUROC by .101 on average, compared with .010 when removing the question. Together, these results establish efficient uncertainty baselines and show that uncertainty in audio-language models depends substantially more on available audio evidence than on question text.

Audio-language models (ALMs) integrate acoustic perception with the knowledge encoded in language models, enabling contextual understanding of auditory events. Making these capabilities practical on devices with limited memory and computation motivates our focus on small ALMs with fewer than 200M parameters. We introduce a recipe that brings together architecture, data, and three-stage training to build Mizar, a 159.3M-parameter ALM. Its architecture connects a compact CED-Small audio encoder to SmolLM2-135M through a frequency-merging mapper. With supervision drawn from ReasonAQA, AudioMCQ, and AVQA, the model undergoes three training stages: audio-language alignment (Stage 1), audio-dependent fine-tuning (Stage 2), and post-training (Stage 3) aimed at strengthening weak skills while retaining learned capabilities. Across five random seeds, Mizar achieves mean accuracies of 52.92% on MMAU, 42.42% on MMAR, and 36.02% on ADQA-clean, surpassing the previous best-performing ALM below 200M parameters on all three benchmarks. It also supports local inference on a single CPU: on questions from the MMAU benchmark, the mean latency from opening the audio file to generating a complete answer is 1.09 seconds. Code and checkpoints are available at https://github.com/KaiyangLi1992/Mizar_159M.

Question answering has advanced rapidly with large language models, but predominantly for high-resource languages, in both text and spoken settings. Spoken question answering (SQA) benchmark for Telugu remains unexplored, and the reliability of automatic evaluation in this setting remains unquantified. We introduce VākQA, a Telugu SQA benchmark of 2,001 factoid question-answer pairs across six domains, with 2.53 hours of speech audio, bilingual transcriptions, and human-verified reference answers. We first validate evaluation methods against human judgements: Gemini-as-a-judge best approximates human ratings but is non-uniformly strict, while open-weight judges systematically penalize correct Telugu answers that differ in surface form from the reference. Using this validated setup, we benchmark proprietary and open-weight models across input modality, language, and domain. We observe that Telugu phrasing retains cultural specificity that is lost in translation, speech input introduces phonetic confusions that alter question meaning, and cascaded ASR-MT errors compound progressively. VākQA is publicly released.

As Speech Large Language Models (speech-LLMs) become increasingly multilingual, Arabic remains significantly underrepresented, highlighting the need for dedicated infrastructure to train and evaluate Arabic speech-LLMs. To address this gap, we introduce Nuha-Speech, a comprehensive initiative to develop general-purpose Arabic speech-LLMs spanning dataset construction, model training, and systematic evaluation. Specifically, we constructed a large-scale Arabic Speech Question-Answering (SQA) corpus comprising over 1.5 million training samples to allow instruction tuning over a broad range of core speech tasks. Then, the corpus was used for supervised fine-tuning based on Qwen-Omni model variants at different scales. Finally, we designed an evaluation framework featuring diverse tasks and tailored metrics. Through this work, we aim to establish foundational infrastructures for Arabic Speech-LLMs under constraints imposed by limited Arabic speech resources.

The advance of multimodal large language models (MLLMs) has fundamentally reshaped the paradigm of human-computer interaction, especially speech interaction models capable of seamless conversations. Despite remarkable performance as general voice assistants, their performance in specialized domains remains underexplored, particularly in scientific areas. Scientific interactions introduce formidable challenges, involving rare technical terminology, spoken norms of abbreviations, and the natural verbalization of symbolic special expressions. In this paper, we introduce S$^3$-Bench, a systematic evaluation framework covering 10 major disciplines, consisting of a Knowledge set for speech question-answering and a Dialogue set for multi-turn progressive interactions with simulated user agents. By decomposing a complete atomic turn into stages of speech recognition, perception, knowledge utilization with reasoning, and response pronunciation, we systematically characterize the common challenges and performance tradeoffs of existing approaches. Furthermore, experiments on multi-turn interactions reveal persistent limitations in user adaptation and the generation of accurate, comprehensive, and efficient responses.

Large audio-language models (LALMs) can hallucinate audio objects, answering "yes" to absent sound events, thus undermining reliability in audio question answering. We propose Token-Adaptive Decoding (TAD), a training-free strategy for hallucination mitigation that grounds the initial yes/no decision by contrasting logits under real audio with a matched silent reference. TAD introduces a token-adaptive, confidence-guided gate that is decision-critical at the first decoding step and class-conditional on affirmative tokens, using the audio-silent margin to avoid overcorrection when evidence is weak or already sufficient. Experiments on AudioCaps-Hallucination show that, relative to Audio-Aware Decoding (AAD), a contrastive baseline with fixed contrast strength, TAD improves F1 for Qwen2 by 0.059 to 0.117 across Popular, Adversarial, and Random splits, and for Gemma by 0.025 to 0.064, while on Clotho-AQA it raises F1 from 0.810 to 0.816 on Qwen2 and remains comparable to AAD on Gemma.

As large audio language models (LALMs) advance, robust evaluation frameworks have become essential. In this context, Spanish speech understanding under realistic acoustic conditions has received particularly little attention. We introduce ESCUCHA, the first Spanish speech understanding benchmark designed to evaluate LALMs across heterogeneous acoustic conditions and reasoning abilities. ESCUCHA comprises 1,000 human-curated questions paired with audio, totaling 162.9 hours sourced directly ``from the wild'' rather than drawn from existing datasets, with durations ranging from a few seconds to over 80 minutes. The benchmark emphasizes reasoning, spanning 9 perceptual and 10 reasoning categories, and it captures linguistic diversity through multiple Spanish accents and non-normative speech. ESCUCHA further includes multi-audio questions, spoken questions, and audio instructions, and it flags which questions support open-ended evaluation. Benchmarking several state-of-the-art multimodal and speech models reveals substantial performance gaps relative to trained humans.

Temporal grounding in long recordings remains challenging for audio-conditioned LLMs. We present a time-aware audio LLM that answers questions with explicit timestamps over up to 120 minutes of input. Our approach interleaves periodic time markers with continuous audio tokens using large-scale synthetic supervision from a cascaded pipeline. Our model achieves strong temporal-grounding accuracy on short and long benchmarks and supports time-anchored fragment descriptions and summaries. Extensive ablations examine how time representation, marker frequency, tokenization, and duration-mixture design affect accuracy and computational cost. We release model weights and datasets to support further research on time-aware audio understanding, available at https://huggingface.co/ai-sage/GigaChat3.1-Audio-10B-A1.8B.

Recent music audio-language models achieve high accuracy on instrument question-answering benchmarks, but it remains unclear whether this reflects robust audio grounding or benchmark-specific shortcuts. In this paper, we introduce an OpenMIC-derived diagnostic benchmark sequence for instrument grounding in music audio-language models, extending binary instrument-presence QA to genre-prior-reduced examples, confusable instrument discrimination, longer audio context, and temporal localization. Across these settings, high binary QA accuracy often fails to predict model behavior: models can exhibit option-position bias, confusable-instrument errors, and temporal response bias. These results suggest that instrument grounding should be evaluated with multi-axis diagnostic benchmarks rather than a single aggregate accuracy.

We frame the system as diagnostic data curation for a large audio-language model: before fine-tuning, we probe Qwen3-Omni-30B-A3B-Instruct under normal, empty-audio, and shuffled-audio conditions to identify how the model's answers change when audio evidence is removed or mismatched. These model confusion patterns are used to bucket training samples into text-prior, shuffle-leak, strong audio-dependent, and hard or misleading cases. Our strongest train-only system fine-tunes only on strong-audio items, where the normal audio-question pair is correct but both counterfactual variants fail, plus a small number of empty-audio negatives and a text-only response normalizer for parse-failed generations. On the official development set, the best train-only system reaches 67.27% accuracy after response normalization, compared with 65.90% for our local Qwen3-Omni baseline. Final submissions additionally include models trained using train+development splits and a three-model ensemble.

While LALMs show promise on audio question answering, they fail to focus on question-relevant segments of audio and provide a clear, checkable reasoning process when dealing with complex audio reasoning. Reinforcement learning and tool-augmented prompting can help models better relate questions to audio but lack a reliable way to understand, integrate, and self-verify audio segments. To address this gap, we present EChO-Agent, a modular agent framework that reformulates complex audio QA as a planning, tool execution, evidence integration, and answer verification workflow. Experiments on MMAR benchmark show EChO-Agent improves both accuracy and rubric scores over baseline and ablation studies show evidence integration is the key factor.

Audio agents extend large audio-language models (LALMs) by decomposing audio questions into tool calls, intermediate evidence, and iterative reasoning steps. However, as LALMs become stronger, the key challenge shifts from enabling tool use to determining when agentic evidence acquisition genuinely benefits audio understanding. We propose Audio-Mind, an auditable and pluggable framework for conditional evidence acquisition in audio understanding. Audio-Mind dynamically combines a strong frontend with planner-guided tool use, preserving frontend judgment when initial evidence is sufficient while acquiring bounded external evidence for questions with unresolved evidence gaps. Experiments on MMAR and MSU-Bench show that Audio-Mind outperforms prior audio-agent baselines, reaching 80.4% accuracy on MMAR and 82.8% accuracy on MSU-Bench. A matched-backbone comparison highlights why this design matters: under strong audio frontends, agentic decomposition can become an orchestration bottleneck when the workflow does not preserve the frontend's holistic audio-grounded judgment. Beyond accuracy, Audio-Mind produces higher-quality, auditable reasoning traces that expose uncertainty, tool evidence, and answer rationales, offering a potential basis for more reliable audio-QA annotation and error analysis.

Recent advances in Large Audio-Language Models (LALMs) have made real-time, streaming spoken interaction increasingly practical. In this setting, reasoning quality and responsiveness are tightly coupled: delaying reasoning until the speech endpoint can improve answer quality but moves deliberation into user-visible response delay, while answering too early risks committing before decisive evidence arrives. We introduce a learnable wait-think-answer control formulation for LALMs. Motivated by the incremental nature of human conversation, the controller decides under partial audio evidence when to wait, when to externalize a compact reasoning update, and when to answer. Using Qwen2.5-Omni-7B as the base model, we construct aligned wait-think-answer traces from spoken reasoning data, train the controller with supervised fine-tuning (SFT), and then apply Decoupled Clip and Dynamic Sampling Policy Optimization (DAPO). The reward combines answer correctness, action validity, update timing, latency synchronization, reasoning quality, and chain consistency, optimizing the complete wait-think-answer trajectory and not the final answer alone. On a six-task synthetic spoken reasoning question answering (SRQA) benchmark, the six-reward DAPO controller improves the row-weighted accuracy from 67.6% to 70.3% while reducing post-endpoint final-think length by 14% under the same Qwen deployment harness. On a 186-item human-recorded Real Audio Bench, a transfer check beyond text-to-speech (TTS)-rendered speech, the controller family remains functional: SFT achieves the strongest accuracy, while the six-reward DAPO controller is the only learned variant whose final-think length falls below the base. These results suggest that a streaming model should learn when to make intermediate reasoning explicit during the audio stream.

Training general-purpose Audio Large Language Models (ALLMs) across diverse datasets is essential for holistic audio understanding, yet it faces significant challenges due to dataset heterogeneity, which often leads to conflicting gradients and slow convergence. Despite its impact, how to explicitly manage this heterogeneity during training remains underexplored, with current practices relying primarily on uniform mixture. In this work, we analyze multi-dataset AudioQA training from a convergence perspective and propose Grouped Sequential Training (GST). GST strategically organizes datasets into affinity-aware groups and introduces them via a progressive scheduling protocol, effectively balancing the stability of parallel training with the efficiency of sequential optimization. To ensure scalability, we develop gradient-based affinity metrics that capture inter-dataset relationships without the prohibitive cost of empirical transferability estimation. Extensive evaluations on 14 AudioQA datasets spanning speech, music, and environmental sounds demonstrate that GST achieves 30--40% faster convergence than standard parallel training while maintaining or even surpassing the performance of mix-all training. Our results provide both theoretical insights and a practical, model-agnostic framework for efficient large-scale ALLM optimization.

Medical audio data is difficult to collect due to privacy regulations and high annotation costs arising from domain expertise. Thus, existing benchmarks tend to underrepresent complex medical audio scenarios. To address this challenge, we present MedMosaic, a medical audio question-answering dataset designed to benchmark language and audio reasoning models under realistic clinical constraints. MedMosaic features a diverse range of medical audio types, including condition-related physiological sounds, carefully constructed synthetic voices to mimic speech with artifacts as well as real short and long length clinical conversations to model varying context lengths. The dataset also features a total of 46,701 question-answer pairs, spanning categories such as multiple-choice, sequential multi-turn, and open-ended question-answers, enabling systematic evaluation of multi-hop reasoning and answer generation capabilities. Benchmarking 13 audio and multimodal reasoning models reveals that reasoning remains challenging for all evaluated systems, with substantial performance variation across question types. In particular, even state-of-the-art model like Gemini-2.5-pro can only achieve 68.1% accuracy approximately. These findings underscore persistent limitations in medical reasoning and highlight the need for more robust, domain-specific multimodal reasoning models. A sample of benchmark data is available here: https://shorturl.at/Lyp33

Large Audio-Language Models show consistent performance gains across speech and audio benchmarks, yet high scores may not reflect true auditory perception. If a model can answer questions without processing the acoustic signal, the benchmark fails as a measure of auditory understanding. We present a diagnostic framework using two axes: text prior, which measures answerability from text and general knowledge alone, and audio reliance, which assesses actual dependency on the acoustic signal. Evaluating eight LALMs across three benchmarks, we find that models retain 60-72% of their full audio scores even without any audio input. Moreover, among items that require audio, only 3.0-4.2% need the complete audio clip; the majority can be resolved using localized fragments. These findings challenge the assumption that benchmark performance equals robust audio understanding, and we conclude with practical guidelines for improving evaluation reliability and benchmark design.

Existing audio question answering benchmarks largely emphasize sound event classification or caption-grounded queries, often enabling models to succeed through shortcut strategies, short-duration cues, lexical priors, dataset-specific biases, or even bypassing audio via metadata and captions rather than genuine reasoning Thus, we present AUDITA (Audio Understanding from Diverse Internet Trivia Authors), a large-scale, real-world benchmark to rigorously evaluate audio reasoning beyond surface-level acoustic recognition. AUDITA comprises carefully curated, human-authored trivia questions grounded in real-world audio, designed to stress robust auditory reasoning through challenging distractors and long-range temporal dependencies, using probing queries that cannot be answered from isolated text or sound cues alone. Human average accuracy of 32.13% shows both the challenge of the task while demonstrating meaningful comprehension of the audio. In stark contrast, state of-the-art audio question answering models perform poorly, with average accuracy below 8.86%. Beyond raw accuracy, we apply Item Response Theory (IRT) to estimate latent proficiency, question difficulty, and expose systematic deficiencies of the models and data.

Recent Large Audio Language Models have demonstrated impressive capabilities in audio understanding. However, they often suffer from perceptual errors, while reliable audio reasoning is unattainable without first grounding the model's perception in structured auditory scenes. Inspired by Auditory Scene Analysis, we first introduce a Perception-Aware Question Answering (PAQA) dataset. PAQA implements a hierarchical decoupling strategy that separates speech from environmental sound and distinguishes multiple speakers, providing explicit perceptual reasoning for training. Building on this, we propose HyPeR, a two-stage Hybrid Perception-Reasoning framework. In Stage I, we finetune the model on PAQA to perceive acoustic attributes in complex audio. In Stage II, we leverage GRPO to refine the model's internal deliberation. We also introduce PAUSE tokens to facilitate latent computation during acoustically ambiguous phases and design perceptual consistency reward to align reasoning rationales with raw audio. Experiments across benchmarks demonstrate that HyPeR achieves absolute improvements over the base model, with performance comparable to large-scale models, stressing the effectiveness of hybrid perception-grounded reasoning for robust and multi-speaker audio understanding.

Answering natural-language questions over multi-hour audio requires both event recognition and temporal grounding. Current large audio-language models perform well on short clips, but are limited by context length, query-time cost, and weak temporal localization. We present LA-RAG (Long Audio-Retrieval Augmented Generation), a structured framework that converts continuous audio into timestamped event records using an open-vocabulary Audio Grounding Model (AGM), stores them in a SQL event database, and answers queries through intent-aware retrieval followed by LLM-based generation. LA-RAG supports offline grounding mode, where long recordings are pre-indexed for low-latency QA, and inference-time grounding mode, where query-conditioned grounding is performed for shorter open-ended clips. We create 24-hour Home-IoT and Industrial-IoT audio benchmarks and augment CASTELLA, a real-world audio moment retrieval dataset with QA pairs. In offline grounding mode, LA-RAG achieves 76.88% overall accuracy on Home-IoT and 71.10% on Industrial-IoT, with average query latencies below 0.6 seconds. In inference-time grounding mode, state-of-the-art LALMs achieve competitive event-detection accuracy on CASTELLA-QA but low temporal detection F1. We further show that LALMs augmented with our structured retrieval metadata achieve consistent temporal detection improvements, with F1 gains of 11-17% across baseline models with improved latency. These results show that explicit timestamped grounding and structured retrieval provide a practical complement to generative audio-language models for deployment-oriented long-audio QA.

As conversational multimodal AI tools are increasingly adopted to process patient data for health assessment, robust benchmarks are needed to measure progress and expose failure modes under realistic conditions. Despite the importance of respiratory audio for mobile health screening, respiratory audio question answering remains underexplored, with existing studies evaluated narrowly and lacking real-world heterogeneity across modalities, devices, and question types. We hence introduce the \textbf{Respiratory-Audio Question-Answering (RA-QA) benchmark}, including a standardized data generation pipeline, a comprehensive multimodal QA collection, and a unified evaluation protocol. RA-QA harmonizes public RA datasets into a collection of 9 million format-diverse QA pairs covering diagnostic and contextual attributes. We benchmark general audio-language models as well as domain-specific architectures, establishing reproducible reference points and showing how current approaches fail under heterogeneity.

Reliable assessment of the abilities of large audio language models (LALMs) is essential to advancing the state of the art. As benchmarks rapidly evolve to incorporate complex reasoning and subjective tasks, they increasingly necessitate open-ended responses from LALMs. We present Open-ended Response Correctness Assessment (ORCA) -- a reliable and lightweight model-based approach for answer correctness and disagreement modeling. We employ a three-stage annotation pipeline combining human judgment, structured feedback, and human-AI correction, yielding 9,663 annotations across 3,699 question-answer pairs from 15 LALMs on three audio understanding and reasoning benchmarks (achieving a Krippendorff's alpha of 0.82). Our experiments employing curriculum learning show that ORCA models achieve a Spearman correlation of 0.91 with average human correctness ratings on seen benchmarks and generalize to unseen benchmarks with a score of 0.85, outperforming several LLM judge baselines including Gemini 2.5 Flash. Furthermore, we demonstrate that ORCA's predicted variance correlates strongly with human disagreement, allowing it to effectively identify problematic benchmark items.

Large Audio-Language Models (LALMs) excel in Audio QA but often suffer from hallucinations ungrounded in the audio. To our knowledge, we are the first to propose applying vector steering to the audio domain to mitigate this. Unlike text-based steering, our silence-anchored contrastive approach steers the model away from hallucinations by contrasting active audio against a silent baseline. Probing internal states reveals a strong correlation between specific layer representations and output correctness. Leveraging this, we introduce Layer-Weighted Vector Steering (LWVS), a training-free intervention that increases steering strength at influential layers. On the Audio Hallucination QA dataset, LWVS significantly outperforms baselines, boosting Recall on the Gemma model by 15.6% (53.4% to 69.0%). Crucially, MMAU benchmark tests confirm LWVS preserves and even enhances general audio understanding, achieving an 8% relative accuracy increase on the Qwen model (54.8% to 59.2%).

Recent advances in large audio language models (LALMs) have primarily been assessed using a multiple-choice question answering (MCQA) framework. However, subtle changes, such as shifting the order of choices, result in substantially different results. Existing MCQA frameworks do not account for this variability and report a single accuracy number per benchmark or category. We dive into the MCQA evaluation framework and conduct a systematic study spanning three benchmarks (MMAU, MMAR and MMSU) and four models: Audio Flamingo 2, Audio Flamingo 3, Qwen2.5-Omni-7B-Instruct, and Kimi-Audio-7B-Instruct. Our findings indicate that models are sensitive not only to the ordering of choices, but also to the paraphrasing of the question and the choices. Finally, we propose a simpler evaluation protocol and metric that account for subtle variations and provide a more detailed evaluation report of LALMs within the MCQA framework.

Masked token prediction has emerged as a powerful pre-training objective across language, vision, and speech, offering the potential to unify these diverse modalities through a single pre-training task. However, its application for general audio understanding remains underexplored, with BEATs being the only notable example. BEATs has seen limited modifications due to the absence of open-source pre-training code. Furthermore, BEATs was trained only on AudioSet, restricting its broader downstream applicability. To address these gaps, we present OpenBEATs, an open-source framework that extends BEATs via multi-domain audio pre-training. We conduct comprehensive evaluations across six types of tasks, twenty five datasets, and three audio domains, including audio reasoning tasks such as audio question answering, entailment, and captioning. OpenBEATs achieves state-of-the-art performance on six bioacoustics datasets, two environmental sound datasets and five reasoning datasets, performing better than models exceeding a billion parameters at one-fourth their parameter size. These results demonstrate the effectiveness of multi-domain datasets and masked token prediction task to learn general-purpose audio representations. To promote further research and reproducibility, we release all pre-training and evaluation code, pretrained and fine-tuned checkpoints, and training logs at https://github.com/Shikhar-S/OpenBEATs