Picture for Liliang Ren

Liliang Ren

Samba: Simple Hybrid State Space Models for Efficient Unlimited Context Language Modeling

Add code
Jun 11, 2024
Viaarxiv icon

Sparse Modular Activation for Efficient Sequence Modeling

Add code
Jul 11, 2023
Figure 1 for Sparse Modular Activation for Efficient Sequence Modeling
Figure 2 for Sparse Modular Activation for Efficient Sequence Modeling
Figure 3 for Sparse Modular Activation for Efficient Sequence Modeling
Figure 4 for Sparse Modular Activation for Efficient Sequence Modeling
Viaarxiv icon

C-PMI: Conditional Pointwise Mutual Information for Turn-level Dialogue Evaluation

Add code
Jun 27, 2023
Figure 1 for C-PMI: Conditional Pointwise Mutual Information for Turn-level Dialogue Evaluation
Viaarxiv icon

Language Model Pre-Training with Sparse Latent Typing

Add code
Oct 26, 2022
Figure 1 for Language Model Pre-Training with Sparse Latent Typing
Figure 2 for Language Model Pre-Training with Sparse Latent Typing
Figure 3 for Language Model Pre-Training with Sparse Latent Typing
Figure 4 for Language Model Pre-Training with Sparse Latent Typing
Viaarxiv icon

HySPA: Hybrid Span Generation for Scalable Text-to-Graph Extraction

Add code
Jun 30, 2021
Figure 1 for HySPA: Hybrid Span Generation for Scalable Text-to-Graph Extraction
Figure 2 for HySPA: Hybrid Span Generation for Scalable Text-to-Graph Extraction
Figure 3 for HySPA: Hybrid Span Generation for Scalable Text-to-Graph Extraction
Figure 4 for HySPA: Hybrid Span Generation for Scalable Text-to-Graph Extraction
Viaarxiv icon

A Simple Fix for Convolutional Neural Network via Coordinate Embedding

Add code
Mar 24, 2020
Figure 1 for A Simple Fix for Convolutional Neural Network via Coordinate Embedding
Figure 2 for A Simple Fix for Convolutional Neural Network via Coordinate Embedding
Figure 3 for A Simple Fix for Convolutional Neural Network via Coordinate Embedding
Figure 4 for A Simple Fix for Convolutional Neural Network via Coordinate Embedding
Viaarxiv icon

RoNGBa: A Robustly Optimized Natural Gradient Boosting Training Approach with Leaf Number Clipping

Dec 05, 2019
Figure 1 for RoNGBa: A Robustly Optimized Natural Gradient Boosting Training Approach with Leaf Number Clipping
Viaarxiv icon

Scalable and Accurate Dialogue State Tracking via Hierarchical Sequence Generation

Add code
Sep 02, 2019
Figure 1 for Scalable and Accurate Dialogue State Tracking via Hierarchical Sequence Generation
Figure 2 for Scalable and Accurate Dialogue State Tracking via Hierarchical Sequence Generation
Figure 3 for Scalable and Accurate Dialogue State Tracking via Hierarchical Sequence Generation
Figure 4 for Scalable and Accurate Dialogue State Tracking via Hierarchical Sequence Generation
Viaarxiv icon

Towards Universal Dialogue State Tracking

Add code
Oct 22, 2018
Figure 1 for Towards Universal Dialogue State Tracking
Figure 2 for Towards Universal Dialogue State Tracking
Figure 3 for Towards Universal Dialogue State Tracking
Viaarxiv icon

Recurrent Soft Attention Model for Common Object Recognition

Add code
May 29, 2017
Figure 1 for Recurrent Soft Attention Model for Common Object Recognition
Figure 2 for Recurrent Soft Attention Model for Common Object Recognition
Figure 3 for Recurrent Soft Attention Model for Common Object Recognition
Figure 4 for Recurrent Soft Attention Model for Common Object Recognition
Viaarxiv icon