Selected work

Representative Papers

Collective intelligence in science: direct elicitation of diverse information from experts with unknown information structure

Jan 20, 2026

This study addresses the challenge of efficiently aggregating dispersed private information from experts regarding complex scientific hypotheses under conditions of no shared context, unknown information structures, and unverifiable ground truth. The authors propose an incentive mechanism that integrates a self-settling virtual-currency prediction market with a real-time chat system, enabling anonymous experts to directly reveal diverse private signals and trade based on hypothesis veracity—without requiring Bayesian inference or predefined information structures. Through incentive-compatible design and an information aggregation algorithm, the mechanism achieves interpretable and efficient information synthesis at equilibrium, offering a novel paradigm for funding and decision support in large-scale collaborative scientific research.

0 citationsRead paper

EAD: An EEG Adapter for Automated Classification

May 29, 2025

To address the challenge of model non-reusability caused by heterogeneous channel counts across EEG acquisition devices, this paper proposes the EEG Adapter (EAD) framework—the first approach enabling channel-agnostic, robust embedding learning. Built upon a pre-trained EEG foundation model, EAD incorporates lightweight adapters, channel-normalized embedding alignment, and zero-shot prompt-based fine-tuning, supporting zero-shot transfer and multi-task generalization across arbitrary channel configurations. Evaluated on EEG-ImageNet and BrainLat, EAD achieves 99.33% and 92.31% classification accuracy, respectively—substantially outperforming existing baselines. Its core contribution lies in breaking the conventional reliance on fixed channel layouts, establishing the first general-purpose EEG representation learning paradigm compatible with heterogeneous recording hardware.

0 citationsRead paper

Recent publications

Latest Papers

Collective intelligence in science: direct elicitation of diverse information from experts with unknown information structure

Jan 20, 2026

This study addresses the challenge of efficiently aggregating dispersed private information from experts regarding complex scientific hypotheses under conditions of no shared context, unknown information structures, and unverifiable ground truth. The authors propose an incentive mechanism that integrates a self-settling virtual-currency prediction market with a real-time chat system, enabling anonymous experts to directly reveal diverse private signals and trade based on hypothesis veracity—without requiring Bayesian inference or predefined information structures. Through incentive-compatible design and an information aggregation algorithm, the mechanism achieves interpretable and efficient information synthesis at equilibrium, offering a novel paradigm for funding and decision support in large-scale collaborative scientific research.

0 citationsRead paper

EAD: An EEG Adapter for Automated Classification

May 29, 2025

To address the challenge of model non-reusability caused by heterogeneous channel counts across EEG acquisition devices, this paper proposes the EEG Adapter (EAD) framework—the first approach enabling channel-agnostic, robust embedding learning. Built upon a pre-trained EEG foundation model, EAD incorporates lightweight adapters, channel-normalized embedding alignment, and zero-shot prompt-based fine-tuning, supporting zero-shot transfer and multi-task generalization across arbitrary channel configurations. Evaluated on EEG-ImageNet and BrainLat, EAD achieves 99.33% and 92.31% classification accuracy, respectively—substantially outperforming existing baselines. Its core contribution lies in breaking the conventional reliance on fixed channel layouts, establishing the first general-purpose EEG representation learning paradigm compatible with heterogeneous recording hardware.

0 citationsRead paper