mlsanityData to 14 Oct 2026

SonicWorld: Source-Centric Latent Dynamics for Spatial Audio Reasoning

arXiv 2610.17131 ↗First posted 12 Oct 2026cs.SD cs.AI eess.AScited by 0

On the map: LLM reasoning → Audio-language model reasoning also fits: Spatial and embodied reasoning

Abstract (arXiv)

Dynamic spatial audio reasoning requires a model to associate changing acoustic observations with the correct source and recover its spatial state at a queried time or interval, a challenge intensified by overlapping speech and source motion. To study this problem, we introduce Sonic-Bench, comprising 2.8K first-order Ambisonics clips and 28.8K target-specific question-answer pairs across 480 simulated indoor scenes. The benchmark covers localization, distance estimation, and motion reasoning in single- and dual-source settings. We further propose SonicWorld, which integrates physical, learned acoustic, and semantic features into persistent source states through learned transitions and recurrent updates, and projects them into an Omni LLM for spatial reasoning. Compared with Spatial-Omni on the Sonic-Bench test split, SonicWorld improves direction accuracy from 31.98% to 34.69% and reduces azimuth error from 57.22deg to 48.01deg. In zero-shot transfer to Spatial LibriSpeech, SonicWorld retains strong localization performance, achieving 73.98% direction accuracy and 19.29deg azimuth error, compared with 27.14% and 70.74deg for Spatial-Omni.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Audio-language model reasoning, one of 17 directions in LLM reasoning. Builds on 43 of its 53 references that are in the map. First posted on arXiv 12 Oct 2026.

Builds on

43 of 53 references.

Closest ancestors · 6

Same direction, newest first

  1. 2026JAEGER: Joint 3D Audio-Visual Grounding and Reasoning in Simulated Physical Environments20 Feb
  2. 2026Spatial Audio Question Answering and Reasoning on Dynamic Source Movements18 Feb
  3. 2025STAR-Bench: Probing Deep Spatio-Temporal Reasoning as Audio 4D Intelligence28 Oct
  4. 2025Audio Flamingo 3: Advancing Audio Intelligence with Fully Open Large Audio Language Models10 Jul
  5. 2025SAVVY: Spatial Awareness via Audio-Visual LLMs through Seeing and Hearing4 Jun
  6. 2024BAT: Learning to Reason about Spatial Sounds with Large Language Models2 Feb

Other references · 37

  1. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,380
  2. 2014Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translationcited by 4,057
  3. 2023InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Taskscited by 2,565
  4. 2017What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?cited by 2,352
  5. 2018Learning Latent Dynamics for Planning from Pixelscited by 1,264
  6. 2019Habitat: A Platform for Embodied AI Researchcited by 1,191
  7. 2024How Far Are We to GPT-4V? Closing the Gap to Commercial Multimodal Models with Open-Source Suitescited by 1,050
  8. 2020Object-Centric Learning with Slot Attentioncited by 876
  9. 2021Habitat-Matterport 3D Dataset (HM3D): 1000 Large-scale 3D Environments for Embodied AIcited by 592
  10. 2024Qwen2-Audio Technical Reportcited by 560
  11. 2025Phi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAscited by 521
  12. 2023SALMONN: Towards Generic Hearing Abilities for Large Language Modelscited by 506
25 more
  1. 2025Qwen3-Omni Technical Reportcited by 497
  2. 2022BEATs: Audio Pre-Training with Acoustic Tokenizerscited by 388
  3. 2021Conditional Object-Centric Learning from Videocited by 248
  4. 2025Kimi-Audio Technical Reportcited by 222
  5. 2023Listen, Think, and Understandcited by 209
  6. 2024Audio Flamingo: A Novel Audio Language Model with Few-Shot Learning and Dialogue Abilitiescited by 186
  7. 2022SAVi++: Towards End-to-End Object-Centric Learning from Real-World Videoscited by 166
  8. 2022SlotFormer: Unsupervised Visual Dynamics Simulation with Object-Centric Modelscited by 147
  9. 2022SoundSpaces 2.0: A Simulation Platform for Visual-Acoustic Learningcited by 95
  10. 2018Sound Event Localization and Detection of Overlapping Sources Using Convolutional Recurrent Neural Networkscited by 85
  11. 2021A Survey of Sound Source Localization with Deep Learning Methodscited by 46
  12. 2023STARSS23: An Audio-Visual Dataset of Spatial Recordings of Real Scenes with Spatiotemporal Annotations of Sound Eventscited by 41
  13. 2020ACCDOA: Activity-Coupled Cartesian Direction of Arrival Representation for Sound Event Localization and Detectioncited by 26
  14. 2021Multi-ACCDOA: Localizing and Detecting Overlapping Sounds from the Same Class with Auxiliary Duplicating Permutation Invariant Trainingcited by 25
  15. 2021A Dataset of Dynamic Reverberant Sound Scenes with Directional Interferers for Sound Event Localization and Detectioncited by 21
  16. 2019The LOCATA Challenge: Acoustic Source Localization and Trackingcited by 17
  17. 2024Learning Spatially-Aware Language and Audio Embeddingscited by 17
  18. 2021SALSA: Spatial Cue-Augmented Log-Spectrogram Features for Polyphonic Sound Event Localization and Detectioncited by 15
  19. 2024Can Large Language Models Understand Spatial Audio?cited by 15
  20. 2019Localization, Detection and Tracking of Multiple Moving Sound Sources with a Convolutional Recurrent Neural Networkcited by 10
  21. 2023Spatial LibriSpeech: An Augmented Dataset for Spatial Audio Learningcited by 7
  22. 2025SALM: Spatial Audio Language Model with Structured Embeddings for Understanding and Editingcited by 7
  23. 2026Spatial-Omni: Spatial Audio Understanding Integration in Multimodal LLMs via FOA Encodingcited by 7
  24. 2025ASAudio: A Survey of Advanced Spatial Audio Researchcited by 5
  25. 2026A Survey of Audio Reasoning in Multimodal Foundation Modelscited by 3

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026The World is Not Mono: Enabling Spatial Understanding in Large Audio-Language Models
  2. 2025SPUR: A Plug-and-Play Framework for Integrating Spatial Audio Understanding and Reasoning into Large Audio-Language Models
  3. 2026FloorSAV: Elucidating Spatial Audio-Visual Context with 2D Floormap for AV-LLMs
  4. 2025Spatial Audio Motion Understanding and Reasoning
  5. 2026Listen, See and Track: Spatio-Temporal Audio-Visual Sound Event Reasoning for Omni-Modal Language Models
  6. 2025OWL: Geometry-Aware Spatial Reasoning for Audio Large Language Models
  7. 2026SAVU-BENCH: A Real-World Benchmark for Spatial Audio-Visual Understanding
  8. 2026SONIC-O1: A Real-World Benchmark for Evaluating Multimodal Large Language Models on Audio-Video Understanding