Picture for Ziwei Ji

Ziwei Ji

Calibrating Verbal Uncertainty as a Linear Feature to Reduce Hallucinations

Add code
Mar 18, 2025
Viaarxiv icon

Delusions of Large Language Models

Add code
Mar 09, 2025
Viaarxiv icon

Relaxed Recursive Transformers: Effective Parameter Sharing with Layer-wise LoRA

Add code
Oct 28, 2024
Figure 1 for Relaxed Recursive Transformers: Effective Parameter Sharing with Layer-wise LoRA
Figure 2 for Relaxed Recursive Transformers: Effective Parameter Sharing with Layer-wise LoRA
Figure 3 for Relaxed Recursive Transformers: Effective Parameter Sharing with Layer-wise LoRA
Figure 4 for Relaxed Recursive Transformers: Effective Parameter Sharing with Layer-wise LoRA
Viaarxiv icon

ANAH-v2: Scaling Analytical Hallucination Annotation of Large Language Models

Add code
Jul 05, 2024
Figure 1 for ANAH-v2: Scaling Analytical Hallucination Annotation of Large Language Models
Figure 2 for ANAH-v2: Scaling Analytical Hallucination Annotation of Large Language Models
Figure 3 for ANAH-v2: Scaling Analytical Hallucination Annotation of Large Language Models
Figure 4 for ANAH-v2: Scaling Analytical Hallucination Annotation of Large Language Models
Viaarxiv icon

LLM Internal States Reveal Hallucination Risk Faced With a Query

Add code
Jul 03, 2024
Viaarxiv icon

Efficient Document Ranking with Learnable Late Interactions

Add code
Jun 25, 2024
Figure 1 for Efficient Document Ranking with Learnable Late Interactions
Figure 2 for Efficient Document Ranking with Learnable Late Interactions
Figure 3 for Efficient Document Ranking with Learnable Late Interactions
Figure 4 for Efficient Document Ranking with Learnable Late Interactions
Viaarxiv icon

ANAH: Analytical Annotation of Hallucinations in Large Language Models

Add code
May 30, 2024
Viaarxiv icon

High-Dimension Human Value Representation in Large Language Models

Add code
Apr 11, 2024
Figure 1 for High-Dimension Human Value Representation in Large Language Models
Figure 2 for High-Dimension Human Value Representation in Large Language Models
Figure 3 for High-Dimension Human Value Representation in Large Language Models
Figure 4 for High-Dimension Human Value Representation in Large Language Models
Viaarxiv icon

Contrastive Learning for Inference in Dialogue

Add code
Oct 19, 2023
Viaarxiv icon

Towards Mitigating Hallucination in Large Language Models via Self-Reflection

Add code
Oct 10, 2023
Figure 1 for Towards Mitigating Hallucination in Large Language Models via Self-Reflection
Figure 2 for Towards Mitigating Hallucination in Large Language Models via Self-Reflection
Figure 3 for Towards Mitigating Hallucination in Large Language Models via Self-Reflection
Figure 4 for Towards Mitigating Hallucination in Large Language Models via Self-Reflection
Viaarxiv icon