Picture for Mengshu Sun

Mengshu Sun

LookAhead Tuning: Safer Language Models via Partial Answer Previews

Add code
Mar 24, 2025
Viaarxiv icon

Exploring Typographic Visual Prompts Injection Threats in Cross-Modality Generation Models

Add code
Mar 14, 2025
Viaarxiv icon

Bi'an: A Bilingual Benchmark and Model for Hallucination Detection in Retrieval-Augmented Generation

Add code
Feb 26, 2025
Viaarxiv icon

LightThinker: Thinking Step-by-Step Compression

Add code
Feb 21, 2025
Viaarxiv icon

K-ON: Stacking Knowledge On the Head Layer of Large Language Model

Add code
Feb 10, 2025
Figure 1 for K-ON: Stacking Knowledge On the Head Layer of Large Language Model
Figure 2 for K-ON: Stacking Knowledge On the Head Layer of Large Language Model
Figure 3 for K-ON: Stacking Knowledge On the Head Layer of Large Language Model
Figure 4 for K-ON: Stacking Knowledge On the Head Layer of Large Language Model
Viaarxiv icon

OntoTune: Ontology-Driven Self-training for Aligning Large Language Models

Add code
Feb 08, 2025
Figure 1 for OntoTune: Ontology-Driven Self-training for Aligning Large Language Models
Figure 2 for OntoTune: Ontology-Driven Self-training for Aligning Large Language Models
Figure 3 for OntoTune: Ontology-Driven Self-training for Aligning Large Language Models
Figure 4 for OntoTune: Ontology-Driven Self-training for Aligning Large Language Models
Viaarxiv icon

MAQInstruct: Instruction-based Unified Event Relation Extraction

Add code
Feb 06, 2025
Viaarxiv icon

Improving Natural Language Understanding for LLMs via Large-Scale Instruction Synthesis

Add code
Feb 06, 2025
Figure 1 for Improving Natural Language Understanding for LLMs via Large-Scale Instruction Synthesis
Figure 2 for Improving Natural Language Understanding for LLMs via Large-Scale Instruction Synthesis
Figure 3 for Improving Natural Language Understanding for LLMs via Large-Scale Instruction Synthesis
Figure 4 for Improving Natural Language Understanding for LLMs via Large-Scale Instruction Synthesis
Viaarxiv icon

Have We Designed Generalizable Structural Knowledge Promptings? Systematic Evaluation and Rethinking

Add code
Dec 31, 2024
Viaarxiv icon

OneKE: A Dockerized Schema-Guided LLM Agent-based Knowledge Extraction System

Add code
Dec 28, 2024
Figure 1 for OneKE: A Dockerized Schema-Guided LLM Agent-based Knowledge Extraction System
Figure 2 for OneKE: A Dockerized Schema-Guided LLM Agent-based Knowledge Extraction System
Figure 3 for OneKE: A Dockerized Schema-Guided LLM Agent-based Knowledge Extraction System
Viaarxiv icon