# RefTAG: Reflection-aware Temporal Audio Grounding in Large Audio Language Models

arXiv: https://arxiv.org/abs/2610.16217 · HTML: https://mlsanity.com/paper/2610.16217/

## Key facts

- No citing papers in the map yet (data through 19 Oct 2026).
- Placed in Audio language models and spoken dialogue, one of 11 directions in Speech and audio language models.
- Builds on 23 of its 24 references that are in the map.
- First posted on arXiv 12 Oct 2026.

## Placed in

- Speech and audio language models / [Audio language models and spoken dialogue](https://mlsanity.com/speech/audio-language-models-and-spoken-dialogue/)

## Builds on (most cited first, 23 of 23)

- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Qwen3-Omni Technical Report](https://mlsanity.com/paper/2509.17765/) (2025)
- [Clotho: An Audio Captioning Dataset](https://mlsanity.com/paper/1910.09387/) (2019)
- [MMAU: A Massive Multi-Task Audio Understanding and Reasoning Benchmark](https://mlsanity.com/paper/2410.19168/) (2024)
- [Audio Flamingo 3: Advancing Audio Intelligence with Fully Open Large Audio Language Models](https://mlsanity.com/paper/2507.08128/) (2025)
- [MMAR: A Challenging Benchmark for Deep Reasoning in Speech, Audio, Music, and Their Mix](https://mlsanity.com/paper/2505.13032/) (2025)
- [The Benefit Of Temporally-Strong Labels In Audio Event Classification](https://mlsanity.com/paper/2105.07031/) (2021)
- [TimeMarker: A Versatile Video-LLM for Long and Short Video Understanding with Superior Temporal Localization Ability](https://mlsanity.com/paper/2411.18211/) (2024)
- [SRPO: Enhancing Multimodal LLM Reasoning via Reflection-Aware Reinforcement Learning](https://mlsanity.com/paper/2506.01713/) (2025)
- [AudioTime: A Temporally-aligned Audio-text Benchmark Dataset](https://mlsanity.com/paper/2407.02857/) (2024)
- [Text-to-Audio Grounding: Building Correspondence Between Captions and Sound Events](https://mlsanity.com/paper/2102.11474/) (2021)
- [MOSS-Audio Technical Report](https://mlsanity.com/paper/2606.01802/) (2026)
- [TACOS: Temporally-aligned Audio CaptiOnS for Language-Audio Pretraining](https://mlsanity.com/paper/2505.07609/) (2025)
- [Towards Weakly Supervised Text-to-Audio Grounding](https://mlsanity.com/paper/2401.02584/) (2024)
- [Listening Between the Frames: Bridging Temporal Gaps in Large Audio-Language Models](https://mlsanity.com/paper/2511.11039/) (2025)
- [DisTime: Distribution-based Time Representation for Video Large Language Models](https://mlsanity.com/paper/2505.24329/) (2025)
- [Audio Flamingo Sound-CoT Technical Report: Improving Chain-of-Thought Reasoning in Sound Understanding](https://mlsanity.com/paper/2508.11818/) (2025)
- [Event-Grounded Question Answering over Long Audio via Structured Retrieval](https://mlsanity.com/paper/2602.14612/) (2026)
- [Listening with Time: Precise Temporal Awareness for Long-Form Audio Understanding](https://mlsanity.com/paper/2604.22245/) (2026)
- [FireRedAudio: A General-Purpose Audio Language Model with Decoupled Continuous Representations for Understanding and Generation](https://mlsanity.com/paper/2608.24168/) (2026)
- [Towards Fine-grained Temporal Perception: Post-Training Large Audio-Language Models with Audio-Side Time Prompt](https://mlsanity.com/paper/2604.13715/) (2026)
- [A Sensitivity Analysis of Multi-Event Audio Grounding in Audio LLMs](https://mlsanity.com/paper/2603.03855/) (2026)
- [TAG-Bench: Benchmarking Temporal Audio Grounding in Large Audio Language Models](https://mlsanity.com/paper/2609.01542/) (2026)

## Abstract (arXiv)

Temporal audio grounding requires locating every occurrence of a queried target, making both occurrence counts and interval boundaries essential. We propose RefTAG, a post-training framework for verifiable temporal self-correction. Its Reflection Chain-of-Thought (Re-CoT) protocol combines draft intervals, discrete count and boundary actions, and revised intervals in one response. Supervised fine-tuning on these trajectories (Re-CoT-SFT) teaches the protocol; reinforcement learning (RL) jointly rewards final localization and revision outcomes. Across 8,228 queries, Re-CoT-SFT improves query-weighted mIoU over final-answer SFT by 1.74 percentage points. RL further improves TAG-Bench multi-target mIoU by 0.40 points and R@0.5 by 0.86 points. The final model achieves 38.50% TAG-Bench mIoU and 51.01% mean mIoU across three datasets, outperforming the evaluated open-source MOSS-Audio and FireRedAudio models and the commercial Gemini-3.1-Pro on both measures. These results demonstrate that explicitly coupling count- and boundary-aware revision with joint reward optimization provides an effective and verifiable path to temporal audio grounding.