Feiyu Duan
7 papers · 2024–2025 · 5 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 · Bronze
🗺️
Taxonomy Completionist
(15)
⚡
Prolific Year · Productive
Top co-authors
Research topics
Machine Learning
(6)
Natural Language Processing
(5)
Optimization & Theory
(5)
Resources & Methods
(5)
Artificial Intelligence
(4)
Core AI
(4)
Deep Learning
(3)
Application Areas
(3)
Learning Types · Deep Learning
(2)
Applications
(1)
Core Methods
(1)
Learning Types · Machine Learning
(1)
Models
(1)
Techniques
(1)
Keywords
large language model
(6)
data selection
(3)
language model
(2)
data quality
(2)
transfer learning
(2)
information retrieval
(1)
scaling law
(1)
continual pre-training
(1)
retrieval-augmented generation
(1)
downstream performance
(1)
mixture ratio
(1)
cross-domain generalization
(1)
corpus filtering
(1)
multi-stage training
(1)
query generation
(1)
length control
(1)
quality rating
(1)
missing information
(1)
pretraining data
(1)
knowledge density
(1)
Papers
Enhancing LLMs via High-Knowledge Data Selection
AAAI 2025
FIRE: Flexible Integration of Data Quality Ratings for Effective Pretraining
EMNLP 2025
FRAME: Boosting LLMs with A Four-Quadrant Multi-Stage Pretraining Strategy
ACL 2025
LLMs Know What They Need: Leveraging a Missing Information Guided Framework to Empower Retrieval-Augmented Generation
COLING 2025
Preference Curriculum: LLMs Should Always Be Pretrained on Their Preferred Data
ACL 2025
D-CPT Law: Domain-specific Continual Pre-Training Scaling Law for Large Language Models
NeurIPS 2024
PositionID: LLMs can Control Lengths, Copy and Paste with Explicit Positional Awareness
EMNLP 2024