Predicting Brain Responses To Natural Movies With Multimodal LLMs
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866916865088946176 |
|---|---|
| author | Villanueva, Cesar Kadir Torrico Tu, Jiaxin Cindy Tripathy, Mihir Lane, Connor Iyer, Rishab Scotti, Paul S. |
| author_facet | Villanueva, Cesar Kadir Torrico Tu, Jiaxin Cindy Tripathy, Mihir Lane, Connor Iyer, Rishab Scotti, Paul S. |
| contents | We present MedARC's team solution to the Algonauts 2025 challenge. Our pipeline leveraged rich multimodal representations from various state-of-the-art pretrained models across video (V-JEPA2), speech (Whisper), text (Llama 3.2), vision-text (InternVL3), and vision-text-audio (Qwen2.5-Omni). These features extracted from the models were linearly projected to a latent space, temporally aligned to the fMRI time series, and finally mapped to cortical parcels through a lightweight encoder comprising a shared group head plus subject-specific residual heads. We trained hundreds of model variants across hyperparameter settings, validated them on held-out movies and assembled ensembles targeted to each parcel in each subject. Our final submission achieved a mean Pearson's correlation of 0.2085 on the test split of withheld out-of-distribution movies, placing our team in fourth place for the competition. We further discuss a last-minute optimization that would have raised us to second place. Our results highlight how combining features from models trained in different modalities, using a simple architecture consisting of shared-subject and single-subject components, and conducting comprehensive model selection and ensembling improves generalization of encoding models to novel movie stimuli. All code is available on GitHub. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_19956 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Predicting Brain Responses To Natural Movies With Multimodal LLMs Villanueva, Cesar Kadir Torrico Tu, Jiaxin Cindy Tripathy, Mihir Lane, Connor Iyer, Rishab Scotti, Paul S. Computer Vision and Pattern Recognition Artificial Intelligence Neurons and Cognition We present MedARC's team solution to the Algonauts 2025 challenge. Our pipeline leveraged rich multimodal representations from various state-of-the-art pretrained models across video (V-JEPA2), speech (Whisper), text (Llama 3.2), vision-text (InternVL3), and vision-text-audio (Qwen2.5-Omni). These features extracted from the models were linearly projected to a latent space, temporally aligned to the fMRI time series, and finally mapped to cortical parcels through a lightweight encoder comprising a shared group head plus subject-specific residual heads. We trained hundreds of model variants across hyperparameter settings, validated them on held-out movies and assembled ensembles targeted to each parcel in each subject. Our final submission achieved a mean Pearson's correlation of 0.2085 on the test split of withheld out-of-distribution movies, placing our team in fourth place for the competition. We further discuss a last-minute optimization that would have raised us to second place. Our results highlight how combining features from models trained in different modalities, using a simple architecture consisting of shared-subject and single-subject components, and conducting comprehensive model selection and ensembling improves generalization of encoding models to novel movie stimuli. All code is available on GitHub. |
| title | Predicting Brain Responses To Natural Movies With Multimodal LLMs |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Neurons and Cognition |
| url | https://arxiv.org/abs/2507.19956 |