Imagen de Portada

Do LLMs and VLMs Share Neurons for Inference? Evidence and Mechanisms of Cross-Modal Transfer

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Cui, Chenhang, Zhang, An, Chen, Yuxin, Deng, Gelei, Zheng, Jingnan, Liang, Zhenkai, Wang, Xiang, Chua, Tat-Seng
Formato:	Preprint
Publicado:	2026
Materias:	Computation and Language
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models
por: Cui, Chenhang, et al.
Publicado: (2024)

Transport and Merge: Cross-Architecture Merging for Large Language Models
por: Cui, Chenhang, et al.
Publicado: (2026)

Self-Guard: Defending Large Reasoning Models via enhanced self-reflection
por: Zheng, Jingnan, et al.
Publicado: (2026)

RSafe: Incentivizing proactive reasoning to build robust and adaptive LLM safeguards
por: Zheng, Jingnan, et al.
Publicado: (2025)

Fine-Grained Verifiers: Preference Modeling as Next-token Prediction in Vision-Language Alignment
por: Cui, Chenhang, et al.
Publicado: (2024)

Risky-Bench: Probing Agentic Safety Risks under Real-World Deployment
por: Zheng, Jingnan, et al.
Publicado: (2026)

Understanding Multilingualism in Mixture-of-Experts LLMs: Routing Mechanism, Expert Specialization, and Layerwise Steering
por: Chen, Yuxin, et al.
Publicado: (2026)

ALI-Agent: Assessing LLMs' Alignment with Human Values via Agent-based Evaluation
por: Zheng, Jingnan, et al.
Publicado: (2024)

Do Hallucination Neurons Generalize? Evidence from Cross-Domain Transfer in LLMs
por: Vaddi, Snehit, et al.
Publicado: (2026)

MolCA: Molecular Graph-Language Modeling with Cross-Modal Projector and Uni-Modal Adapter
por: Liu, Zhiyuan, et al.
Publicado: (2023)

MASKDROID: Robust Android Malware Detection with Masked Graph Representations
por: Zheng, Jingnan, et al.
Publicado: (2024)

Who Transfers Safety? Identifying and Targeting Cross-Lingual Shared Safety Neurons
por: Zhang, Xianhui, et al.
Publicado: (2026)

Generative Cross-Modal Retrieval: Memorizing Images in Multimodal Language Models for Retrieval and Beyond
por: Li, Yongqi, et al.
Publicado: (2024)

Hello Again! LLM-powered Personalized Agent for Long-term Dialogue
por: Li, Hao, et al.
Publicado: (2024)

SemCORE: A Semantic-Enhanced Generative Cross-Modal Retrieval Framework with MLLMs
por: Li, Haoxuan, et al.
Publicado: (2025)

STYLE: Improving Domain Transferability of Asking Clarification Questions in Large Language Model Powered Conversational Agents
por: Chen, Yue, et al.
Publicado: (2024)

Aligning Large Language Models for Faithful Integrity Against Opposing Argument
por: Zhao, Yong, et al.
Publicado: (2025)

IGD: Token Decisiveness Modeling via Information Gain in LLMs for Personalized Recommendation
por: Lin, Zijie, et al.
Publicado: (2025)

Search-in-the-Chain: Interactively Enhancing Large Language Models with Search for Knowledge-intensive Tasks
por: Xu, Shicheng, et al.
Publicado: (2023)

Towards Human-centered Proactive Conversational Agents
por: Deng, Yang, et al.
Publicado: (2024)

Beyond Persuasion: Towards Conversational Recommender System with Credible Explanations
por: Qin, Peixin, et al.
Publicado: (2024)

XNLP: An Interactive Demonstration System for Universal Structured NLP
por: Fei, Hao, et al.
Publicado: (2023)

Exploring Training and Inference Scaling Laws in Generative Retrieval
por: Cai, Hongru, et al.
Publicado: (2025)

Language-specific Neurons Do Not Facilitate Cross-Lingual Transfer
por: Mondal, Soumen Kumar, et al.
Publicado: (2025)

Don't Just Say "I don't know"! Self-aligning Large Language Models for Responding to Unknown Questions with Explanations
por: Deng, Yang, et al.
Publicado: (2024)

Plug-and-Play Policy Planner for Large Language Model Powered Dialogue Agents
por: Deng, Yang, et al.
Publicado: (2023)

Ask-before-Plan: Proactive Language Agents for Real-World Planning
por: Zhang, Xuan, et al.
Publicado: (2024)

A Survey on Neural Question Generation: Methods, Applications, and Prospects
por: Guo, Shasha, et al.
Publicado: (2024)

Length Controlled Generation for Black-box LLMs
por: Gu, Yuxuan, et al.
Publicado: (2024)

GIFT: Games as Informal Training for Generalizable LLMs
por: Lyu, Nuoyan, et al.
Publicado: (2026)

Do LLMs Signal When They're Right? Evidence from Neuron Agreement
por: Chen, Kang, et al.
Publicado: (2025)

Concept -- An Evaluation Protocol on Conversational Recommender Systems with System-centric and User-centric Factors
por: Huang, Chen, et al.
Publicado: (2024)

Chart-based Reasoning: Transferring Capabilities from LLMs to VLMs
por: Carbune, Victor, et al.
Publicado: (2024)

Strength Lies in Differences! Improving Strategy Planning for Non-collaborative Dialogues via Diversified User Simulation
por: Zhang, Tong, et al.
Publicado: (2024)

A Picture Is Worth a Graph: A Blueprint Debate Paradigm for Multimodal Reasoning
por: Zheng, Changmeng, et al.
Publicado: (2024)

SCTc-TE: A Comprehensive Formulation and Benchmark for Temporal Event Forecasting
por: Ma, Yunshan, et al.
Publicado: (2023)

On the Multi-turn Instruction Following for Conversational Web Agents
por: Deng, Yang, et al.
Publicado: (2024)

Cross-Modal Safety Mechanism Transfer in Large Vision-Language Models
por: Xu, Shicheng, et al.
Publicado: (2024)

Efficient Inference for Large Language Model-based Generative Recommendation
por: Lin, Xinyu, et al.
Publicado: (2024)

Beware of Your Po! Measuring and Mitigating AI Safety Risks in Role-Play Fine-Tuning of LLMs
por: Zhao, Weixiang, et al.
Publicado: (2025)