Picture for Xiaozhi Wang

Xiaozhi Wang

Precise Localization of Memories: A Fine-grained Neuron-level Knowledge Editing Technique for LLMs

Add code
Mar 03, 2025
Viaarxiv icon

Sparse Auto-Encoder Interprets Linguistic Features in Large Language Models

Add code
Feb 27, 2025
Viaarxiv icon

Agentic Reward Modeling: Integrating Human Preferences with Verifiable Correctness Signals for Reliable Reward Systems

Add code
Feb 26, 2025
Viaarxiv icon

LongBench v2: Towards Deeper Understanding and Reasoning on Realistic Long-context Multitasks

Add code
Dec 19, 2024
Figure 1 for LongBench v2: Towards Deeper Understanding and Reasoning on Realistic Long-context Multitasks
Figure 2 for LongBench v2: Towards Deeper Understanding and Reasoning on Realistic Long-context Multitasks
Figure 3 for LongBench v2: Towards Deeper Understanding and Reasoning on Realistic Long-context Multitasks
Figure 4 for LongBench v2: Towards Deeper Understanding and Reasoning on Realistic Long-context Multitasks
Viaarxiv icon

Constraint Back-translation Improves Complex Instruction Following of Large Language Models

Add code
Oct 31, 2024
Figure 1 for Constraint Back-translation Improves Complex Instruction Following of Large Language Models
Figure 2 for Constraint Back-translation Improves Complex Instruction Following of Large Language Models
Figure 3 for Constraint Back-translation Improves Complex Instruction Following of Large Language Models
Figure 4 for Constraint Back-translation Improves Complex Instruction Following of Large Language Models
Viaarxiv icon

Configurable Foundation Models: Building LLMs from a Modular Perspective

Add code
Sep 04, 2024
Figure 1 for Configurable Foundation Models: Building LLMs from a Modular Perspective
Figure 2 for Configurable Foundation Models: Building LLMs from a Modular Perspective
Figure 3 for Configurable Foundation Models: Building LLMs from a Modular Perspective
Figure 4 for Configurable Foundation Models: Building LLMs from a Modular Perspective
Viaarxiv icon

OpenEP: Open-Ended Future Event Prediction

Add code
Aug 14, 2024
Viaarxiv icon

MAVEN-Fact: A Large-scale Event Factuality Detection Dataset

Add code
Jul 22, 2024
Viaarxiv icon

Finding Safety Neurons in Large Language Models

Add code
Jun 20, 2024
Viaarxiv icon

R-Eval: A Unified Toolkit for Evaluating Domain Knowledge of Retrieval Augmented Large Language Models

Add code
Jun 17, 2024
Viaarxiv icon