Jixing Li

This study examines how hearing-impaired and normal-hearing listeners process speech in multi-talker scenarios, revealing disruptions in neural responses at shorter and longer temporal scales.

This paper surveys foundation models for non-invasive brain decoding, reviewing their applications and challenges in visual, language, and auditory domains.

The paper introduces a naturalistic neuroimaging dataset with EEG and 7T fMRI recordings from 26 Mandarin speakers listening to single- and multi-talker speech, enabling research on attention and auditory processing in complex environments.

The paper introduces a naturalistic neuroimaging dataset with EEG and 7T fMRI recordings from 26 Mandarin speakers listening to single- and multi-talker speech, enabling research on attention and auditory processing in complex environments.

This study simulates aphasia in large language models by disrupting components to explore modularity in language processing and its parallels with human systems.

The paper presents a multimodal database of neuroimaging and EEG data from older Cantonese speakers listening to The Little Prince, aiming to study language comprehension and healthy aging effects.

This study shows that scaling large language models improves their alignment with human brain activity during reading more than instruction tuning.

This study investigates whether neural signatures of semantic composition in a controlled two-word paradigm extend to naturalistic story listening using MEG, finding consistent patterns with differences in latency that disappear when accounting for contextual factors.

The paper investigates brain activity during object-extracted relative clause processing in English and Chinese, finding common and language-specific neural correlates.

The paper uses fMRI to study neural activity during zero pronoun resolution in Chinese, finding increased left middle temporal gyrus activity and representations of referents in the Precuneus and Parahippocampal Gyrus.

The paper introduces a multilingual fMRI corpus using the audiobook The Little Prince to study cross-linguistic differences in language processing.

The paper uses MEG to distinguish neural activity for semantic composition and association in the left temporal lobe, finding distinct roles for anterior and middle temporal regions.

The paper investigates neural correlates of semantic number across three languages, finding common activation patterns for plural vs. singular nouns.

This study uses fMRI and MEG to identify neural activity patterns during pronoun processing and evaluates symbolic and deep learning models to understand referential processing mechanisms.