Picture for Liu Leqi

Liu Leqi

AgentDistill: Training-Free Agent Distillation with Generalizable MCP Boxes

Add code
Jun 17, 2025
Viaarxiv icon

Hypothesis Testing for Quantifying LLM-Human Misalignment in Multiple Choice Settings

Add code
Jun 17, 2025
Viaarxiv icon

Learning Composable Chains-of-Thought

Add code
May 28, 2025
Viaarxiv icon

Teaching with Lies: Curriculum DPO on Synthetic Negatives for Hallucination Detection

Add code
May 23, 2025
Viaarxiv icon

EquivaMap: Leveraging LLMs for Automatic Equivalence Checking of Optimization Formulations

Add code
Feb 20, 2025
Viaarxiv icon

A Common Pitfall of Margin-based Language Model Alignment: Gradient Entanglement

Add code
Oct 17, 2024
Viaarxiv icon

Accounting for AI and Users Shaping One Another: The Role of Mathematical Models

Add code
Apr 18, 2024
Figure 1 for Accounting for AI and Users Shaping One Another: The Role of Mathematical Models
Figure 2 for Accounting for AI and Users Shaping One Another: The Role of Mathematical Models
Viaarxiv icon

Steering LLMs Towards Unbiased Responses: A Causality-Guided Debiasing Framework

Add code
Mar 13, 2024
Figure 1 for Steering LLMs Towards Unbiased Responses: A Causality-Guided Debiasing Framework
Figure 2 for Steering LLMs Towards Unbiased Responses: A Causality-Guided Debiasing Framework
Figure 3 for Steering LLMs Towards Unbiased Responses: A Causality-Guided Debiasing Framework
Figure 4 for Steering LLMs Towards Unbiased Responses: A Causality-Guided Debiasing Framework
Viaarxiv icon

Personalized Language Modeling from Personalized Human Feedback

Add code
Feb 06, 2024
Viaarxiv icon

A Field Test of Bandit Algorithms for Recommendations: Understanding the Validity of Assumptions on Human Preferences in Multi-armed Bandits

Add code
Apr 16, 2023
Figure 1 for A Field Test of Bandit Algorithms for Recommendations: Understanding the Validity of Assumptions on Human Preferences in Multi-armed Bandits
Figure 2 for A Field Test of Bandit Algorithms for Recommendations: Understanding the Validity of Assumptions on Human Preferences in Multi-armed Bandits
Figure 3 for A Field Test of Bandit Algorithms for Recommendations: Understanding the Validity of Assumptions on Human Preferences in Multi-armed Bandits
Figure 4 for A Field Test of Bandit Algorithms for Recommendations: Understanding the Validity of Assumptions on Human Preferences in Multi-armed Bandits
Viaarxiv icon