AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Real-World Unsupervised Models Generalize to Predict Brain Responses to Out-of-Distribution Stimuli

Chenggang Chen, Zhiyu Yang, Xiaoqin Wang

ICML 2026 · Label-efficient and robust learning › Recognition and long-tail · approach: Data augmentation and generalization bias

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
pretrain on real-world WenetSpeech instead of curated LibriSpeech (ablation)HuBERT/Wav2Vec2 Base · NH2015 auditory fMRIexplained variance: 0.773 vs previous best 0.729Improved
pretrain on small uncurated FSD50K+AudioSet (153h) instead of LibriSpeech (960h) (ablation)HuBERT · B2021 auditory fMRIRSA: 15.5% gainImproved
fine-tune pretrained Wav2Vec2 on LibriSpeech (ablation)Wav2Vec2LS · NH2015 / B2021 auditory fMRIexplained variance: no improvement on NH2015, deteriorated on B2021No clear effect
fine-tune pretrained Wav2Vec2 for speech (ablation)Wav2Vec2 · NH2015 speech componentexplained variance rank: ranked 6thImproved
use unsupervised Mandarin-trained model instead of supervised English modelsHuBERTspeech · NH2015 RSASpearman rho: 0.506 vs 0.491Improved
use HuBERTspeech instead of CochResNet50-MultiTaskglobal RDM alignment · NH2015 CKACKA: 0.7182 vs 0.7012Improved
use HuBERTspeech instead of CochResNet50-MultiTaskglobal RDM alignment · B2021 CKACKA: 0.7331 vs 0.7013Improved
use supervised ResNet50-MultiTask instead of HuBERTspeechlocal neighborhood alignment · MkNN at small kmutual kNN overlap: k < 40 NH2015; k < 10 B2021Mixed
train on infant head-cam video instead of ImageNet (curated) (ablation)ResNeXt-50-32x4d · Brain-Score Bracci2019Brain-Score: 0.169 -> 0.249 to 0.259Improved
scale ImageNet ResNeXt depth/width (50-32x4d to 101-64x4d) (ablation)ResNeXt on ImageNet · Brain-Score Bracci2019Brain-Score: 0.169, 0.196, 0.21 with ImageNet acc 77.6%, 79.3%, 83.2%Mixed
train on infant head-cam video instead of web-curated images (DINO)baby-vision ResNeXt · Brain-Score Bracci2019Brain-Score: 0.249 to 0.259 vs 0.251 to 0.262No clear effect
train on SAYCam head-cam unsupervised models vs previous leadercamS/camA/camY/camSAY · Brain-Score tong.Coggan2024 fMRI.ITBrain-Score: 0.267 -> 0.438, 0.4167, 0.4137, 0.393Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy