Fangyuan Li
8 papers · 2016–2026 · 6 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 ↓+2 more ↓ Show less ↑
🗺️ Taxonomy Completionist (23) 🏃 Academic Marathon · Half 🌍 Conference Polyglot · Silver 🌉 Interdisciplinary Bridge · Bronze 🧭 Keyword Pioneer
⚡
Prolific Year · Productive
❓
The Questioner
Top co-authors
Research topics
Artificial Intelligence
(6)
Core AI
(6)
Machine Learning
(4)
Application Areas
(4)
Deep Learning
(3)
Optimization & Theory
(3)
Natural Language Processing
(2)
Applications
(2)
Interdisciplinary
(1)
Reinforcement Learning
(1)
Architectures
(1)
Core Methods
(1)
Learning Types
(1)
Linguistics
(1)
Methods
(1)
Keywords
reinforcement learning
(2)
model compression
(2)
model merging
(1)
code generation
(1)
parameter-efficient fine-tuning
(1)
cross-lingual nlp
(1)
tree search
(1)
credit assignment
(1)
memory efficiency
(1)
language model
(1)
neural network optimization
(1)
gradient approximation
(1)
multimodal large language model
(1)
reward shaping
(1)
multi-task learning
(1)
edge computing
(1)
task vector
(1)
mobile deployment
(1)
key-value cache
(1)
benchmark evaluation
(1)
Papers
MARS2: Scaling Multi-Agent Tree Search via Reinforcement Learning for Code Generation
ACL 2026
WIST: Web-Grounded Iterative Self-Play Tree for Domain-Targeted Reasoning Improvement
ACL 2026
EdgeInfinite: A Memory-Efficient Infinite-Context Transformer for Edge Devices
ACL 2025
Fast and Slow Gradient Approximation for Binary Neural Network Optimization
AAAI 2025
GenieBlue: Integrating both Linguistic and Multimodal Capabilities for Large Language Models on Mobile Devices
ICCV 2025
Less is More: Efficient Model Merging with Binary Task Switch
CVPR 2025
SmartBench: Is Your LLM Truly a Good Chinese Smartphone Assistant?
EMNLP 2025
Learning Event Expressions via Bilingual Structure Projection
COLING 2016