影响力指数

95.83/100

前 0.2%

全站排名 #146

发表论文64 篇

平均评分5.3

年均产出21.3 篇/年

Fei Huang

Senior Research Director@Alibaba Group US·美国·OpenReview

研究方向

Machine Learning · Machine Translation · Natural Language Processing · machine translation · natural language processing · information extraction · dialogue system

Fei Huang

Adaptive Social Learning via Mode Policy Optimization for Language Agents

Perception-Aware Policy Optimization for Multimodal Reasoning

WebWeaver: Structuring Web-Scale Evidence with Dynamic Outlines for Open-Ended Deep Research

Demystifying Deep Search: A Holistic Evaluation with Hint-free Multi-Hop Questions and Factorised Metrics

SPELL: Self-Play Reinforcement Learning for Evolving Long-Context Language Models

Scaling Generalist Data-Analytic Agents

SoLoPO: Unlocking Long-Context Capabilities in LLMs via Short-to-Long Preference Optimization

SASFT: Sparse Autoencoder-guided Supervised Finetuning to Mitigate Unexpected Code-Switching in LLMs

Expanding the Capability Frontier of LLM Agents with ZPD-Guided Data Synthesis

OSWorld-MCP: Benchmarking MCP Tool Invocation In Computer-Use Agents

AgentFold: Long-Horizon Web Agents with Proactive Context Folding

WebWatcher: Breaking New Frontiers of Vision-Language Deep Research Agent

MaskSearch: Towards Scalable Agentic Pre-Training for Search-Enhanced Reasoning

Overcoming Joint Intractability with Lossless Hierarchical Speculative Decoding

Agentic Reinforcement Learning with Implicit Step Rewards

Efficient Multimodal Planning Agent for Visual Question-Answering

ZeroSearch: Incentivize the Search Capability of LLMs without Searching

Repurposing Synthetic Data for Fine-grained Search Agent Supervision

WebSailor-V2: Bridging the Chasm to Proprietary Agents via Synthetic Data and Scalable Reinforcement Learning

P-GenRM: Personalized Generative Reward Model with Test-time User-based Scaling

Scaling Agents via Continual Pre-training

AutoHete: An Automatic and Efficient Heterogeneous Training System for LLMs

A Simple "Motivation" Can Enhance Reinforcement Finetuning of Large Reasoning Models

Writing-RL: Advancing Long-form Writing via Adaptive Curriculum Reinforcement Learning

mGRPO: Unlocking LLM Reasoning through Multilingual Thinking

MUSEG: Reinforcing Video Temporal Understanding via Timestamp-Aware Multi-Segment Grounding

ReSum: Unlocking Long-Horizon Search Intelligence via Context Summarization

Towards General Agentic Intelligence via Environment Scaling

IterResearch: Rethinking Long-Horizon Agents with Interaction Scaling

UI-S1: Advancing GUI Automation via Semi-online Reinforcement Learning

MARS: Optimizing Dual-System Deep Research via Multi-Agent Reinforcement Learning

DynamicBench: Evaluating Real-Time Report Generation in Large Language Models

End-to-End QA Construction Pipeline for Continual Pre-training of Large Language Models

TimeHC-RL: Temporal‑aware Hierarchical Cognitive Reinforcement Learning for Enhancing LLMs' Social Intelligence

WebSailor: Navigating Super-human Reasoning for Web Agent

Sampling-Efficient Test-Time Scaling: Self-Estimating the Best-of-N Sampling in Early Decoding

VRAG-RL: Empower Vision-Perception-Based RAG for Visually Rich Information Understanding via Iterative Reasoning with Reinforcement Learning

Look Before You Leap: A GUI-Critic-R1 Model for Pre-Operative Error Diagnosis in GUI Automation

Towards Efficient Online Tuning of VLM Agents via Counterfactual Soft Reinforcement Learning

On the Role of Attention Heads in Large Language Model Safety

VLM-R³: Region Recognition, Reasoning, and Refinement for Enhanced Multimodal Chain-of-Thought

OpenOmni: Advancing Open-Source Omnimodal Large Language Models with Progressive Multimodal Alignment and Real-time Emotional Speech Synthesis

WebDancer: Towards Autonomous Information Seeking Agency

StructRAG: Boosting Knowledge Intensive Reasoning of LLMs via Inference-time Hybrid Information Structurization

mPLUG-Owl3: Towards Long Image-Sequence Understanding in Multi-Modal Large Language Models

ConText: Driving In-context Learning for Text Removal and Segmentation

Benchmarking Agentic Workflow Generation

Benchmarking Multimodal Retrieval Augmented Generation with Dynamic VQA Dataset and Self-adaptive Planning Agent

MMEvol: Empowering Multimodal Large Language Models with Evol-Instruct

Extend Model Merging from Fine-Tuned to Pre-Trained Large Language Models via Weight Disentanglement

Advancing Language Multi-Agent Learning with Credit Re-Assignment for Interactive Environment Generalization

ExploraCoder: Advancing code generation for multiple unseen APIs via planning and chained exploration

In-Context Transfer Learning: Demonstration Synthesis by Transferring Similar Tasks

LaRA: Benchmarking Retrieval-Augmented Generation and Long-Context LLMs – No Silver Bullet for LC or RAG Routing

Exploiting Presentative Feature Distributions for Parameter-Efficient Continual Learning of Large Language Models

Exploring Knowledge Boundaries in Large Language Models for Retrieval Judgment

Codev-Bench: How Do LLMs Understand Developer-Centric Code Completion?

Enabling Weak LLMs to Judge Response Reliability via Meta Ranking

Enhancing Multi-Agent Learning in Real-World Interactive Environments through Process Reward Decomposition