# Tracing Audio Grounding and Answer Selection in Audio LLMs

arXiv: https://arxiv.org/abs/2609.04637 · HTML: https://mlsanity.com/paper/2609.04637/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 24 of its 24 references that are in the map.
- First posted on arXiv 4 Sep 2026.

## Builds on (most cited first, 24 of 24)

- [Qwen2.5-Omni Technical Report](https://mlsanity.com/paper/2503.20215/) (2025)
- [Qwen2-Audio Technical Report](https://mlsanity.com/paper/2407.10759/) (2024)
- [Dissecting Recall of Factual Associations in Auto-Regressive Language Models](https://mlsanity.com/paper/2304.14767/) (2023)
- [MMAU: A Massive Multi-Task Audio Understanding and Reasoning Benchmark](https://mlsanity.com/paper/2410.19168/) (2024)
- [MMAR: A Challenging Benchmark for Deep Reasoning in Speech, Audio, Music, and Their Mix](https://mlsanity.com/paper/2505.13032/) (2025)
- [MMSU: A Massive Multi-task Spoken Language Understanding and Reasoning Benchmark](https://mlsanity.com/paper/2506.04779/) (2025)
- [Understanding Sounds, Missing the Questions: The Challenge of Object Hallucination in Large Audio-Language Models](https://mlsanity.com/paper/2406.08402/) (2024)
- [Can Large Audio-Language Models Truly Hear? Tackling Hallucinations with Multi-Task Assessment and Stepwise Audio Reasoning](https://mlsanity.com/paper/2410.16130/) (2024)
- [AudioLens: A Closer Look at Auditory Attribute Perception of Large Audio-Language Models](https://mlsanity.com/paper/2506.05140/) (2025)
- [Measuring Audio's Impact on Correctness: Audio-Contribution-Aware Post-Training of Large Audio Language Models](https://mlsanity.com/paper/2509.21060/) (2025)
- [Map the Flow: Revealing Hidden Pathways of Information in VideoLLMs](https://mlsanity.com/paper/2510.13251/) (2025)
- [What You See is What You Ask: Evaluating Audio Descriptions](https://mlsanity.com/paper/2510.00808/) (2025)
- [All That Glitters Is Not Audio: Rethinking Text Priors and Audio Reliance in Audio-Language Evaluation](https://mlsanity.com/paper/2604.24401/) (2026)
- [Discovering and Causally Validating Emotion-Sensitive Neurons in Large Audio-Language Models](https://mlsanity.com/paper/2601.03115/) (2026)
- [AQUA-Bench: Beyond Finding Answers to Knowing When There Are None in Audio Question Answering](https://mlsanity.com/paper/2601.12248/) (2026)
- [Multi-Domain Audio Question Answering Benchmark Toward Acoustic Content Reasoning](https://mlsanity.com/paper/2505.07365/) (2025)
- [Can Large Audio Language Models Understand Audio Well? Speech, Scene and Events Understanding Benchmark for LALMs](https://mlsanity.com/paper/2509.13148/) (2025)
- [Causal Tracing of Audio-Text Fusion in Large Audio Language Models](https://mlsanity.com/paper/2603.13768/) (2026)
- [When Silence Matters: The Impact of Irrelevant Audio on Text Reasoning in Large Audio-Language Models](https://mlsanity.com/paper/2510.00626/) (2025)
- [Are Audio-Language Models Listening? Audio-Specialist Heads for Adaptive Audio Steering](https://mlsanity.com/paper/2603.06854/) (2026)
- [Who Wins the Conflict? Mechanistic Interpretability of Text Bias in Audio LLMs](https://mlsanity.com/paper/2606.18924/) (2026)
- [Summary of DCASE 2026 Task 5: Audio-Dependent Question Answering](https://mlsanity.com/paper/2607.18718/) (2026)
- [From Semantics to Readout: Mechanistic Understanding of Audio Tokens after Fine-Tuning for Temporal Audio Grounding](https://mlsanity.com/paper/2607.25355/) (2026)
- [Exploring Audio Hallucination in Egocentric Video Understanding](https://mlsanity.com/paper/2604.23860/) (2026)

## Built on by (most cited first, 1 of 1)

- [Does Location Matter? A Black-Box Causal Audit Reveals Model-Dependent Audio Grounding in LALM Hallucination](https://mlsanity.com/paper/2610.13112/) (2026)

## Abstract (arXiv)

Audio Large Language Models (Audio LLMs) have advanced in audio understanding, yet they can still predict the answer by reasoning from textual cues or linguistic priors rather than the provided audio. A common remedy is to train models on data whose answers cannot be inferred from text alone. This approach can improve performance, but what changes within the model remains unclear. In this paper, we ask what must happen inside the model for the audio to actually determine the answer. Our findings are threefold. (1) Replacing the audio with silence or unrelated audio causes substantially larger performance degradation in the trained model than in the pretrained model. (2) Acoustic information most strongly shapes the model's representations of the answer choices in early-to-middle layers, while training mainly increases the influence of audio information on the final prediction in middle-to-late layers. (3) The weights learned during training have their largest impact in specific layer bands. Together, these results provide a mechanistic account of how training strengthens the use of acoustic evidence in Audio LLMs.