Saved in:
| Main Authors: | , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2605.09152 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910205757882368 |
|---|---|
| author | Hu, Jucheng Chen, Zhangquan Chen, Yulin Hong, Chengjie Zhou, Liang Wang, Tairan Li, Sifei Zhu, Giulio Zhou, Feng Zeng, Yiheng Yang, Suorong Zhou, Dongzhan |
| author_facet | Hu, Jucheng Chen, Zhangquan Chen, Yulin Hong, Chengjie Zhou, Liang Wang, Tairan Li, Sifei Zhu, Giulio Zhou, Feng Zeng, Yiheng Yang, Suorong Zhou, Dongzhan |
| contents | Deciphering animal intent is a fundamental challenge in computational ethology, largely because of semantic aliasing, the phenomenon where identical external signals (e.g., a cat's purr) correspond to radically different internal states depending on physiological context. Existing Multimodal Large Language Models (MLLMs) are blind to high-frequency biological time-series data, restricting them to superficial behavioural pattern matching rather than genuine latent-state reasoning. To bridge this gap, we introduce Meow-Omni 1, the first open-source, quad-modal MLLM purpose-built for computational ethology. It natively fuses video, audio, and physiological time-series streams with textual reasoning. Through targeted architectural adaptation, we integrate specialized scientific encoders into a unified backbone and formalize intent inference via physiologically grounded cross-modal alignment. Evaluated on MeowBench, a novel, expert-verified quad-modal benchmark, Meow-Omni 1 achieves state-of-the-art intent-recognition accuracy (71.16%), substantially outperforming leading vision-language and omni-modal baselines. We release the complete open-source pipeline including model weights, training framework, and the Meow-10K dataset, to establish a scalable paradigm for inter-species intent understanding and to advance foundation models toward real-world veterinary diagnostics and wildlife conservation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_09152 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Meow-Omni 1: A Multimodal Large Language Model for Feline Ethology Hu, Jucheng Chen, Zhangquan Chen, Yulin Hong, Chengjie Zhou, Liang Wang, Tairan Li, Sifei Zhu, Giulio Zhou, Feng Zeng, Yiheng Yang, Suorong Zhou, Dongzhan Computation and Language Neurons and Cognition Deciphering animal intent is a fundamental challenge in computational ethology, largely because of semantic aliasing, the phenomenon where identical external signals (e.g., a cat's purr) correspond to radically different internal states depending on physiological context. Existing Multimodal Large Language Models (MLLMs) are blind to high-frequency biological time-series data, restricting them to superficial behavioural pattern matching rather than genuine latent-state reasoning. To bridge this gap, we introduce Meow-Omni 1, the first open-source, quad-modal MLLM purpose-built for computational ethology. It natively fuses video, audio, and physiological time-series streams with textual reasoning. Through targeted architectural adaptation, we integrate specialized scientific encoders into a unified backbone and formalize intent inference via physiologically grounded cross-modal alignment. Evaluated on MeowBench, a novel, expert-verified quad-modal benchmark, Meow-Omni 1 achieves state-of-the-art intent-recognition accuracy (71.16%), substantially outperforming leading vision-language and omni-modal baselines. We release the complete open-source pipeline including model weights, training framework, and the Meow-10K dataset, to establish a scalable paradigm for inter-species intent understanding and to advance foundation models toward real-world veterinary diagnostics and wildlife conservation. |
| title | Meow-Omni 1: A Multimodal Large Language Model for Feline Ethology |
| topic | Computation and Language Neurons and Cognition |
| url | https://arxiv.org/abs/2605.09152 |