Ting-Han Fan
9 papers · 2021–2024 · 7 venues · across top CS/AI venues
Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.
Achievements
Jump to papers ↓
🌍
Conference Polyglot · Silver
🌉
Interdisciplinary Bridge · Silver
🗺️
Taxonomy Completionist
(21)
Top co-authors
Research topics
Keywords
length extrapolation
(5)
positional embedding
(3)
transformer architecture
(3)
attention mechanism
(3)
language model
(3)
regular language
(2)
model-based reinforcement learning
(1)
working memory
(1)
imitation learning
(1)
transformer language model
(1)
benchmark environment
(1)
positional encoding
(1)
deep generative model
(1)
discrete latent variable
(1)
cumulative reward
(1)
contraction mapping
(1)
temperature scaling
(1)
weight sharing
(1)
straight-through estimator
(1)
variance reduction
(1)
Papers
Advancing Regular Language Reasoning in Linear Recurrent Neural Networks
NAACL 2024
Attention Alignment and Flexible Positional Embeddings Improve Transformer Length Extrapolation
NAACL 2024
Dissecting Transformer Length Extrapolation via the Lens of Receptive Field Analysis
ACL 2023
Latent Positional Information is in the Self-Attention Variance of Transformer Language Models Without Positional Embeddings
ACL 2023
Transformer Working Memory Enables Regular Language Reasoning And Natural Language Length Extrapolation
EMNLP 2023
KERPLE: Kernelized Relative Positional Embedding for Length Extrapolation
NeurIPS 2022
PowerGym: A Reinforcement Learning Environment for Volt-Var Control in Power Distribution Systems
L4DC 2022
Training Discrete Deep Generative Models via Gapped Straight-Through Estimator
ICML 2022
A Contraction Approach to Model-based Reinforcement Learning
AISTATS 2021