Search Results for author: Zeping Yu

Found 11 papers, 7 papers with code

Locate-then-Merge: Neuron-Level Parameter Fusion for Mitigating Catastrophic Forgetting in Multimodal LLMs

no code implementations22 May 2025 Zeping Yu, Sophia Ananiadou

Although multimodal large language models (MLLMs) have achieved impressive performance, the multimodal instruction tuning stage often causes catastrophic forgetting of the base LLM's language ability, even in strong models like Llama3.

Hallucination

Back Attention: Understanding and Enhancing Multi-Hop Reasoning in Large Language Models

no code implementations15 Feb 2025 Zeping Yu, Yonatan Belinkov, Sophia Ananiadou

We investigate how large language models perform latent multi-hop reasoning in prompts like "Wolfgang Amadeus Mozart's mother's spouse is".

Attribute Attribute Extraction +2

Understanding and Mitigating Gender Bias in LLMs via Interpretable Neuron Editing

no code implementations24 Jan 2025 Zeping Yu, Sophia Ananiadou

Existing methods to mitigate bias lack a comprehensive understanding of its mechanisms or compromise the model's core capabilities.

Understanding Multimodal LLMs: the Mechanistic Interpretability of Llava in Visual Question Answering

1 code implementation17 Nov 2024 Zeping Yu, Sophia Ananiadou

Understanding the mechanisms behind Large Language Models (LLMs) is crucial for designing improved models and strategies.

Hallucination In-Context Learning +2

Interpreting Arithmetic Mechanism in Large Language Models through Comparative Neuron Analysis

2 code implementations21 Sep 2024 Zeping Yu, Sophia Ananiadou

We find arithmetic ability resides within a limited number of attention heads, with each head specializing in distinct operations.

Model Editing Prediction

How do Large Language Models Learn In-Context? Query and Key Matrices of In-Context Heads are Two Towers for Metric Learning

2 code implementations5 Feb 2024 Zeping Yu, Sophia Ananiadou

We investigate the mechanism of in-context learning (ICL) on sentence classification tasks with semantically-unrelated labels ("foo"/"bar").

In-Context Learning Metric Learning +3

Neuron-Level Knowledge Attribution in Large Language Models

3 code implementations19 Dec 2023 Zeping Yu, Sophia Ananiadou

Additionally, since most static methods typically only identify "value neurons" directly contributing to the final prediction, we propose a method for identifying "query neurons" which activate these "value neurons".

knowledge editing

Emotion Detection for Misinformation: A Review

no code implementations1 Nov 2023 Zhiwei Liu, Tianlin Zhang, Kailai Yang, Paul Thompson, Zeping Yu, Sophia Ananiadou

The emotions and sentiments of netizens, as expressed in social media posts and news, constitute important factors that can help to distinguish fake news from genuine news and to understand the spread of rumors.

Fake News Detection Misinformation

CodeCMR: Cross-Modal Retrieval For Function-Level Binary Source Code Matching

1 code implementation NeurIPS 2020 Zeping Yu, Wenxin Zheng, Jiaqi Wang, Qiyi Tang, Sen Nie, Shi Wu

We adopt Deep Pyramid Convolutional Neural Network (DPCNN) for source code feature extraction and Graph Neural Network (GNN) for binary code feature extraction.

Computer Security Cross-Modal Retrieval +3

Sliced Recurrent Neural Networks

3 code implementations COLING 2018 Zeping Yu, Gongshen Liu

In this paper, we introduce sliced recurrent neural networks (SRNNs), which could be parallelized by slicing the sequences into many subsequences.

Sentiment Analysis

Cannot find the paper you are looking for? You can Submit a new open access paper.