计算语言学与意识科学实验室Computational Linguistics & Consciousness Sciences Lab

Computational Linguistics & Consciousness Sciences Lab

计算语言学与意识科学实验室

We study how human language is used, understood, learned, and evolved across time scales and dynamic environments, using computational methods to explore the deep connections among language, intelligence, and consciousness.

We are actively recruiting self-motivated graduate, undergraduate students, and post-docs.

Research Areas

Learn more & view publications
01

Representations and Mechanisms in Language Models

We analyze how models encode, organize, and retrieve linguistic knowledge to explain capability formation and model behavior.

Representation AnalysisMechanistic InterpretationBehavioral Analysis
02

Reasoning Processes and Capability Enhancement

We study reasoning paths, structured generation, and capability boundaries to develop more effective and interpretable reasoning methods.

Reasoning ProcessesStructured GenerationCapability Enhancement
03

Knowledge Boundaries and Reliable Language Models

We study when models know, when they do not, and how confidence and knowledge boundaries can support appropriate responses under uncertainty.

Knowledge BoundariesConfidenceSelective Answering
04

Generated-Text Detection and Quality Evaluation

We use spectral, statistical, and psycholinguistic signals to detect machine-generated text and evaluate generation quality and human preference.

Text DetectionGeneration EvaluationHuman Preference
05

Model Memory and Lifelong Learning

We study how models store, forget, and update information while maintaining long-term memory and adapting to new knowledge in continual settings.

Model MemoryLifelong LearningKnowledge Updating
06

Language, Cognition, and Human Preference

We connect psycholinguistics, eye tracking, behavioral experiments, and computational models to compare human and machine language processing.

PsycholinguisticsEye TrackingHuman Behavior

Latest News

View all news