VoxSafeBench reveals that speech language models recognize social norms from text but fail to apply them when acoustic cues like speaker or scene determine the appropriate response.
hub Baseline reference
Mmau-pro: A challenging and comprehensive benchmark for holistic evaluation of audio general intelligence
Baseline reference. 57% of citing Pith papers use this work as a benchmark or comparison.
18
Pith papers citing it
Baseline
57% of classified citations
hub tools
citation-role summary
dataset 4
background 2
other 1
citation-polarity summary
years
2026 18representative citing papers
Introduces CASU benchmark with four tasks to evaluate context-aware auditory scene understanding in LALMs via semi-synthetic audio compositions of speech, events, and environments.
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.
A reasoning-guided ordinal SER framework conditions LALMs on paired speech, trains on semantic and GeMAPS-derived reasoning traces, and applies direct preference optimization to improve comparative emotion prediction with only 5% of conventional training data.
Introduces a benchmark for mechanistic analysis of temporal failures in LALMs and shows attention scaling at bottleneck layers improves accuracy from 55.9% to 59.1%.
MedMosaic is a new large-scale medical audio question-answering benchmark showing that even Gemini-2.5-pro reaches only ~68% accuracy across diverse clinical audio scenarios.
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.
HeadRouter prunes audio tokens more effectively by dynamically routing based on per-head importance for semantic versus acoustic tasks, exceeding baseline performance at 70% token retention on Qwen2.5-Omni models.
Self-distillation of Qwen3-Omni-Thinking on 545k Cogito-Pipe audio reasoning traces yields the best open-source MMAR CoT scores and top-tier challenge ranking.
A three-stage synthetic data pipeline generates 8800 doctor-patient conversations totaling 1.3k hours of audio and LLM-produced SOAP notes, with evaluation showing cascaded transcription-then-summarization models outperform end-to-end audio models.
MAPO is a dual-branch RL framework using modality relevance masks from cross-modal differential entropy and auxiliary attention losses to reduce late-stage modality collapse in audio reasoning models and improve benchmark results.
A survey of Large Audio Language Models that establishes a taxonomy of trustworthiness vulnerabilities and proposes a Defense-in-Depth roadmap for audio intelligence.
OmniJigsaw is a self-supervised proxy task that reconstructs shuffled audio-visual clips via joint integration, sample-level selection, and clip-level masking strategies, yielding gains on 15 video, audio, and reasoning benchmarks.
MOSS-Audio is an audio-language model using a 12.5 Hz encoder, DeepStack cross-layer injection, time markers, and an event-preserving annotation pipeline for unified audio understanding.
Audio-Mind introduces a conditional, auditable agentic framework for audio understanding that preserves frontend judgment and acquires bounded external evidence only when needed, reporting 80.4% on MMAR and 82.8% on MSU-Bench.
VISA ranks 2nd in the Interspeech 2026 ARC Agent Track by adding multi-modal feature extraction, consistency-checked model voting, and rubric-aligned routing to large audio language models, reaching 66.23% Rubrics score and 77.40% accuracy.
LLMs exhibit a persistent modality gap versus specialized audio encoders on MSEB tasks, with no conclusive evidence favoring audio-native over cascaded architectures.
A survey that provides a unified formulation of audio reasoning and reviews advances across Audio-to-Text, Audio-to-Speech, Audio-Visual, and Agentic paradigms while discussing challenges and future directions.
citing papers explorer
-
VoxSafeBench: Not Just What Is Said, but Who, How, and Where
VoxSafeBench reveals that speech language models recognize social norms from text but fail to apply them when acoustic cues like speaker or scene determine the appropriate response.
-
From Sounds to Scenes: A Benchmark for Evaluating Context-Aware Auditory Scene Understanding in Large Audio Language Models
Introduces CASU benchmark with four tasks to evaluate context-aware auditory scene understanding in LALMs via semi-synthetic audio compositions of speech, events, and environments.
-
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.
-
Comparative Reasoning: Making an Audio Language Model Better at Comparing Emotions
A reasoning-guided ordinal SER framework conditions LALMs on paired speech, trains on semantic and GeMAPS-derived reasoning traces, and applies direct preference optimization to improve comparative emotion prediction with only 5% of conventional training data.
-
A Closer Look at Failure Modes in Temporal Understanding of Large Audio-Language Models
Introduces a benchmark for mechanistic analysis of temporal failures in LALMs and shows attention scaling at bottleneck layers improves accuracy from 55.9% to 59.1%.
-
MedMosaic: A Challenging Large Scale Benchmark of Diverse Medical Audio
MedMosaic is a new large-scale medical audio question-answering benchmark showing that even Gemini-2.5-pro reaches only ~68% accuracy across diverse clinical audio scenarios.
-
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.
-
HeadRouter: Dynamic Head-Weight Routing for Task-Adaptive Audio Token Pruning in Large Audio Language Models
HeadRouter prunes audio tokens more effectively by dynamically routing based on per-head importance for semantic versus acoustic tasks, exceeding baseline performance at 70% token retention on Qwen2.5-Omni models.
-
Audio-Cogito: Towards Deep Audio Reasoning in Large Audio Language Models
Self-distillation of Qwen3-Omni-Thinking on 545k Cogito-Pipe audio reasoning traces yields the best open-source MMAR CoT scores and top-tier challenge ranking.
-
Generating Synthetic Doctor-Patient Conversations for Long-form Audio Summarization
A three-stage synthetic data pipeline generates 8800 doctor-patient conversations totaling 1.3k hours of audio and LLM-produced SOAP notes, with evaluation showing cascaded transcription-then-summarization models outperform end-to-end audio models.
-
Escape the Language Prior: Mitigating Late-Stage Modality Collapse in Audio Reasoning via Modality-Aware Policy Optimization
MAPO is a dual-branch RL framework using modality relevance masks from cross-modal differential entropy and auxiliary attention losses to reduce late-stage modality collapse in audio reasoning models and improve benchmark results.
-
A Survey of Large Audio Language Models: Generalization, Trustworthiness, and Outlook
A survey of Large Audio Language Models that establishes a taxonomy of trustworthiness vulnerabilities and proposes a Defense-in-Depth roadmap for audio intelligence.
-
OmniJigsaw: Enhancing Omni-Modal Reasoning via Modality-Orchestrated Reordering
OmniJigsaw is a self-supervised proxy task that reconstructs shuffled audio-visual clips via joint integration, sample-level selection, and clip-level masking strategies, yielding gains on 15 video, audio, and reasoning benchmarks.
-
MOSS-Audio Technical Report
MOSS-Audio is an audio-language model using a 12.5 Hz encoder, DeepStack cross-layer injection, time markers, and an event-preserving annotation pipeline for unified audio understanding.
-
Audio-Mind: An Auditable Agentic Framework for Audio Understanding
Audio-Mind introduces a conditional, auditable agentic framework for audio understanding that preserves frontend judgment and acquires bounded external evidence only when needed, reporting 80.4% on MMAR and 82.8% on MSU-Bench.
-
VISA: A Visual Information Strengthened Audio-Reasoning System for the Interspeech 2026 ARC Agent Track
VISA ranks 2nd in the Interspeech 2026 ARC Agent Track by adding multi-modal feature extraction, consistency-checked model voting, and rubric-aligned routing to large audio language models, reaching 66.23% Rubrics score and 77.40% accuracy.
-
Benchmarking LLMs on the Massive Sound Embedding Benchmark (MSEB)
LLMs exhibit a persistent modality gap versus specialized audio encoders on MSEB tasks, with no conclusive evidence favoring audio-native over cascaded architectures.
-
A Survey of Audio Reasoning in Multimodal Foundation Models
A survey that provides a unified formulation of audio reasoning and reviews advances across Audio-to-Text, Audio-to-Speech, Audio-Visual, and Agentic paradigms while discussing challenges and future directions.