# TORUS: A Test of Rendering-Understanding Self-Coherence for Unified Audio Models

arXiv: https://arxiv.org/abs/2607.28896 · HTML: https://mlsanity.com/paper/2607.28896/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 25 of its 42 references that are in the map.
- First posted on arXiv 30 Jul 2026.

## Builds on (most cited first, 25 of 25)

- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [Robust Speech Recognition via Large-Scale Weak Supervision](https://mlsanity.com/paper/2212.04356/) (2022)
- [Large-scale Contrastive Language-Audio Pretraining with Feature Fusion and Keyword-to-Caption Augmentation](https://mlsanity.com/paper/2211.06687/) (2022)
- [CNN Architectures for Large-Scale Audio Classification](https://mlsanity.com/paper/1609.09430/) (2016)
- [Unified-IO 2: Scaling Autoregressive Multimodal Models with Vision, Language, Audio, and Action](https://mlsanity.com/paper/2312.17172/) (2023)
- [MMAU: A Massive Multi-Task Audio Understanding and Reasoning Benchmark](https://mlsanity.com/paper/2410.19168/) (2024)
- [Efficient Training of Audio Transformers with Patchout](https://mlsanity.com/paper/2110.05069/) (2021)
- [MMSU: A Massive Multi-task Spoken Language Understanding and Reasoning Benchmark](https://mlsanity.com/paper/2506.04779/) (2025)
- [Codec Does Matter: Exploring the Semantic Shortcoming of Codec for Audio Language Model](https://mlsanity.com/paper/2408.17175/) (2024)
- [TangoFlux: Super Fast and Faithful Text to Audio Generation with Flow Matching and Clap-Ranked Preference Optimization](https://mlsanity.com/paper/2412.21037/) (2024)
- [Zero-Shot Unsupervised and Text-Based Audio Editing Using DDPM Inversion](https://mlsanity.com/paper/2402.10009/) (2024)
- [ACE-Step 1.5: Pushing the Boundaries of Open-Source Music Generation](https://mlsanity.com/paper/2602.00744/) (2026)
- [MME-Unify: A Comprehensive Benchmark for Unified Multimodal Understanding and Generation Models](https://mlsanity.com/paper/2504.03641/) (2025)
- [Quantifying the Gap between Understanding and Generation within Unified Multimodal Models](https://mlsanity.com/paper/2602.02140/) (2026)
- [Audio-Omni: Extending Multi-modal Understanding to Versatile Audio Generation and Editing](https://mlsanity.com/paper/2604.10708/) (2026)
- [UALM: Unified Audio Language Model for Understanding, Generation and Reasoning](https://mlsanity.com/paper/2510.12000/) (2025)
- [MMEDIT: A Unified Framework for Multi-Type Audio Editing via Audio Language Model](https://mlsanity.com/paper/2512.20339/) (2025)
- [AQAScore: Evaluating Semantic Alignment in Text-to-Audio Generation via Audio Question Answering](https://mlsanity.com/paper/2601.14728/) (2026)
- [All That Glitters Is Not Audio: Rethinking Text Priors and Audio Reliance in Audio-Language Evaluation](https://mlsanity.com/paper/2604.24401/) (2026)
- [UniAudio 2.0: A Unified Audio Language Model with Text-Aligned Factorized Audio Tokenization](https://mlsanity.com/paper/2602.04683/) (2026)
- [Beyond Accuracy: Benchmarking Cross-Task Consistency in Unified Multimodal Models](https://mlsanity.com/paper/2604.25072/) (2026)
- [MMAE: A Massive Multitask Audio Editing Benchmark](https://mlsanity.com/paper/2606.07229/) (2026)
- [SpeechEditBench: A Bilingual Multi-Attribute Benchmark for Instruction-Guided Speech Editing](https://mlsanity.com/paper/2606.01804/) (2026)
- [RiTTA: Modeling Event Relations in Text-to-Audio Generation](https://mlsanity.com/paper/2412.15922/) (2024)
- [UmniBench: Unified Understand and Generation Model Oriented Omni-dimensional Benchmark](https://mlsanity.com/paper/2512.17196/) (2025)

## Abstract (arXiv)

Unified audio models capable of audio understanding, audio generation and, increasingly, audio editing are proliferating rapidly. Yet a basic question about them remains unanswered: do the two heads of a unified model agree about the same audio? Current practice evaluates each capability in isolation on specialized benchmarks, and never asks whether a model can make sense of its own generations. We present TORUS, the first self-coherence test for audio-native unified models. TORUS comprises 48 three-stage self-coherence tests carrying 432 six-option questions spanning speech, sound and music across five task families. We holistically evaluate five open unified models alongside a Cascaded Baseline that combines state-of-the-art specialized generation, editing and understanding models. The best unified model answers 50.5% of questions against the Cascaded Baseline's 63.2% and a 16.7% chance floor. Models struggle on audio editing. Among the evaluated audio models (specialized and unified), we observe limited self-coherence, and thus position self-coherence as an essential test for future audio systems.