Search Results for author: Dehong Xu

Found 8 papers, 3 papers with code

A minimalistic representation model for head direction system

no code implementations15 Nov 2024 Minglu Zhao, Dehong Xu, Deqian Kong, Wen-Hao Zhang, Ying Nian Wu

We demonstrate the emergence of Gaussian-like tuning profiles and a 2D circle geometry in both versions of the model.

Aligning Large Language Models via Fine-grained Supervision

no code implementations4 Jun 2024 Dehong Xu, Liang Qiu, Minseok Kim, Faisal Ladhak, Jaeyoung Do

Pre-trained large-scale language models (LLMs) excel at producing coherent articles, yet their outputs may be untruthful, toxic, or fail to align with user expectations.

An Investigation of Conformal Isometry Hypothesis for Grid Cells

no code implementations27 May 2024 Dehong Xu, Ruiqi Gao, Wen-Hao Zhang, Xue-Xin Wei, Ying Nian Wu

As the agent moves, this vector rotates within a 2D manifold in the neural space, driven by a recurrent neural network.

From Cognition to Computation: A Comparative Review of Human Attention and Transformer Architectures

no code implementations25 Apr 2024 Minglu Zhao, Dehong Xu, Tao Gao

Attention is a cornerstone of human cognition that facilitates the efficient extraction of information in everyday life.

Latent Plan Transformer for Trajectory Abstraction: Planning as Latent Space Inference

1 code implementation7 Feb 2024 Deqian Kong, Dehong Xu, Minglu Zhao, Bo Pang, Jianwen Xie, Andrew Lizarraga, Yuhao Huang, Sirui Xie, Ying Nian Wu

We introduce the Latent Plan Transformer (LPT), a novel model that leverages a latent variable to connect a Transformer-based trajectory generator and the final return.

MuJoCo

Diverse and Faithful Knowledge-Grounded Dialogue Generation via Sequential Posterior Inference

1 code implementation1 Jun 2023 Yan Xu, Deqian Kong, Dehong Xu, Ziwei Ji, Bo Pang, Pascale Fung, Ying Nian Wu

The capability to generate responses with diversity and faithfulness using factual knowledge is paramount for creating a human-like, trustworthy dialogue system.

Dialogue Generation Diversity +1

Conformal Isometry of Lie Group Representation in Recurrent Network of Grid Cells

1 code implementation6 Oct 2022 Dehong Xu, Ruiqi Gao, Wen-Hao Zhang, Xue-Xin Wei, Ying Nian Wu

Recurrent neural networks have been proposed to explain the properties of the grid cells by updating the neural activity vector based on the velocity input of the animal.

Cannot find the paper you are looking for? You can Submit a new open access paper.