Colin Conwell

Researcher@Massachusetts Institute of Technology·美国·OpenReview

研究方向

computational cognitive neuroscience · deep neural network models of brain and behavior · vision and language models of brain and behavior · behavioral psychophysics · large language models (reasoning) · natural language processing · deep learning for program synthesis · mechanistic interpretability; machine psychophysics

4.0

Training the Untrainable: Introducing Inductive Bias via Representational Alignment

NeurIPS 2025Poster

三作

6.0

Training the Untrainable: Introducing Inductive Bias via Representational Alignment

ICLR 2025Rejected

三作

5.8

Rethinking Language-Alignment in Human Visual Cortex with Syntax Manipulation and Word Models

ICLR 2025Rejected

一作

4.0

Using Multimodal Deep Neural Networks to Disentangle Language from Visual Aesthetic Experience

合作者 (20)

Colin Conwell

Network of Theseus (like the ship)

How Predictable is AI Progress?

Modeling dynamic social vision highlights gaps between deep learning and humans

Training the Untrainable: Introducing Inductive Bias via Representational Alignment

Training the Untrainable: Introducing Inductive Bias via Representational Alignment

Rethinking Language-Alignment in Human Visual Cortex with Syntax Manipulation and Word Models

Using Multimodal Deep Neural Networks to Disentangle Language from Visual Aesthetic Experience