Yankai Lin

OpenReview

Affiliations

Wechat AI, Tencent Inc.

papers

1,888

total citations

papers (12)

Proactive Agent: Shifting LLM Agents from Reactive Responses to Active Assistance

ICLR 2025arXiv

citations

Predicting Emergent Abilities with Infinite Resolution Evaluation

ICLR 2024arXiv

citations

WorkflowLLM: Enhancing Workflow Orchestration Capability of Large Language Models

ICLR 2025arXiv

citations

Exploring the Benefit of Activation Sparsity in Pre-training

ICML 2024arXiv

citations

Uncertainty and Influence aware Reward Model Refinement for Reinforcement Learning from Human Feedback

ICLR 2025

citations

Learning to Focus: Causal Attention Distillation via Gradient‐Guided Token Pruning

NEURIPS 2025arXiv

citations

Rational Decision-Making Agent with Learning Internal Utility Judgment

ICLR 2025

citations

Yankai Lin

Affiliations

papers (12)

ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs

ULTRAFEEDBACK: Boosting Language Models with Scaled AI Feedback

Advancing LLM Reasoning Generalists with Preference Trees

Topology-Imbalance Learning for Semi-Supervised Node Classification

Large Multilingual Models Pivot Zero-Shot Multimodal Learning across Languages

Proactive Agent: Shifting LLM Agents from Reactive Responses to Active Assistance

Predicting Emergent Abilities with Infinite Resolution Evaluation

WorkflowLLM: Enhancing Workflow Orchestration Capability of Large Language Models

Exploring the Benefit of Activation Sparsity in Pre-training

Uncertainty and Influence aware Reward Model Refinement for Reinforcement Learning from Human Feedback

Learning to Focus: Causal Attention Distillation via Gradient‐Guided Token Pruning

Rational Decision-Making Agent with Learning Internal Utility Judgment

papers (12)

ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs

ULTRAFEEDBACK: Boosting Language Models with Scaled AI Feedback

Advancing LLM Reasoning Generalists with Preference Trees

Topology-Imbalance Learning for Semi-Supervised Node Classification

Large Multilingual Models Pivot Zero-Shot Multimodal Learning across Languages

Proactive Agent: Shifting LLM Agents from Reactive Responses to Active Assistance

Predicting Emergent Abilities with Infinite Resolution Evaluation

WorkflowLLM: Enhancing Workflow Orchestration Capability of Large Language Models

Exploring the Benefit of Activation Sparsity in Pre-training

Uncertainty and Influence aware Reward Model Refinement for Reinforcement Learning from Human Feedback

Learning to Focus: Causal Attention Distillation via Gradient‐Guided Token Pruning

Rational Decision-Making Agent with Learning Internal Utility Judgment