Cover Image

Unveiling Instruction-Specific Neurons & Experts: An Analytical Framework for LLM's Instruction-Following Capabilities

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Zhang, Junyan, Gao, Yubo, Yan, Yibo, Li, Jungang, Hou, Zhaorui, Tao, Sicheng, Liu, Shuliang, Dai, Song, Hei, Yonghua, Li, Junzhuo, Hu, Xuming
Format:	Preprint
Published:	2025
Subjects:	Computation and Language Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Unlocking Speech Instruction Data Potential with Query Rewriting
by: Hei, Yonghua, et al.
Published: (2025)

PhysicsArena: The First Multimodal Physics Reasoning Benchmark Exploring Variable, Process, and Solution Dimensions
by: Dai, Song, et al.
Published: (2025)

CoheMark: A Novel Sentence-Level Watermark for Enhanced Text Quality
by: Zhang, Junyan, et al.
Published: (2025)

Do BERT-Like Bidirectional Models Still Perform Better on Text Classification in the Era of LLMs?
by: Zhang, Junyan, et al.
Published: (2025)

MOSS-ChatV: Reinforcement Learning with Process Reasoning Reward for Video Temporal Reasoning
by: Tao, Sicheng, et al.
Published: (2025)

EffiReason-Bench: A Unified Benchmark for Evaluating and Advancing Efficient Reasoning in Large Language Models
by: Huang, Junquan, et al.
Published: (2025)

Self-Review Framework for Enhancing Instruction Following Capability of LLM
by: Park, Sihyun
Published: (2025)

Thinking Economically: A Hierarchical Framework for Adaptive-Complexity Reasoning in LLMs
by: Gao, Yubo, et al.
Published: (2026)

Dynamic Expert Specialization: Towards Catastrophic Forgetting-Free Multi-Domain MoE Adaptation
by: Li, Junzhuo, et al.
Published: (2025)

When Instructions Multiply: Measuring and Estimating LLM Capabilities of Multiple Instructions Following
by: Harada, Keno, et al.
Published: (2025)

Temporal Gains, Spatial Costs: Revisiting Video Fine-Tuning in Multimodal Large Language Models
by: Zhang, Linghao, et al.
Published: (2026)

Unveiling Language Routing Isolation in Multilingual MoE Models for Interpretable Subnetwork Adaptation
by: Zheng, Kening, et al.
Published: (2026)

Decoding Knowledge Attribution in Mixture-of-Experts: A Framework of Basic-Refinement Collaboration and Efficiency Analysis
by: Li, Junzhuo, et al.
Published: (2025)

VideoMark: A Distortion-Free Robust Watermarking Framework for Video Diffusion Models
by: Hu, Xuming, et al.
Published: (2025)

Optimal Expert-Attention Allocation in Mixture-of-Experts: A Scalable Law for Dynamic Model Design
by: Li, Junzhuo, et al.
Published: (2026)

SAVEn-Vid: Synergistic Audio-Visual Integration for Enhanced Understanding in Long Video Context
by: Li, Jungang, et al.
Published: (2024)

Quantifying LLM Biases Across Instruction Boundary in Mixed Question Forms
by: Ling, Zipeng, et al.
Published: (2025)

MMNeuron: Discovering Neuron-Level Domain-Specific Interpretation in Multimodal Large Language Model
by: Huo, Jiahao, et al.
Published: (2024)

PMark: Towards Robust and Distortion-free Semantic-level Watermarking with Channel Constraints
by: Huo, Jiahao, et al.
Published: (2025)

DocPruner: A Storage-Efficient Framework for Multi-Vector Visual Document Retrieval via Adaptive Patch-Level Embedding Pruning
by: Yan, Yibo, et al.
Published: (2025)

MINER: Mining the Underlying Pattern of Modality-Specific Neurons in Multimodal Large Language Models
by: Huang, Kaichen, et al.
Published: (2024)

Enhancing LLM Instruction Following: An Evaluation-Driven Multi-Agentic Workflow for Prompt Instructions Optimization
by: Purpura, Alberto, et al.
Published: (2026)

Internal Chain-of-Thought: Empirical Evidence for Layer-wise Subtask Scheduling in LLMs
by: Yang, Zhipeng, et al.
Published: (2025)

CodeIF: Benchmarking the Instruction-Following Capabilities of Large Language Models for Code Generation
by: Yan, Kaiwen, et al.
Published: (2025)

A Visual Semantic Adaptive Watermark grounded by Prefix-Tuning for Large Vision-Language Model
by: Zheng, Qi, et al.
Published: (2026)

Marco-Bench-MIF: On Multilingual Instruction-Following Capability of Large Language Models
by: Zeng, Bo, et al.
Published: (2025)

Non-instructional Fine-tuning: Enabling Instruction-Following Capabilities in Pre-trained Language Models without Instruction-Following Data
by: Xie, Juncheng, et al.
Published: (2024)

Instruction-Following Pruning for Large Language Models
by: Hou, Bairu, et al.
Published: (2025)

CodeIF-Bench: Evaluating Instruction-Following Capabilities of Large Language Models in Interactive Code Generation
by: Wang, Peiding, et al.
Published: (2025)

Enhancing Instruction-Following Capability of Visual-Language Models by Reducing Image Redundancy
by: Yang, Te, et al.
Published: (2024)

VerIF: Verification Engineering for Reinforcement Learning in Instruction Following
by: Peng, Hao, et al.
Published: (2025)

Natural Language based Specification and Verification
by: Li, Zhaorui, et al.
Published: (2026)

Visual Instruction Pretraining for Domain-Specific Foundation Models
by: Li, Yuxuan, et al.
Published: (2025)

Towards Better Instruction Following Retrieval Models
by: Zhuang, Yuchen, et al.
Published: (2025)

Instruction Following without Instruction Tuning
by: Hewitt, John, et al.
Published: (2024)

Enhancing and Assessing Instruction-Following with Fine-Grained Instruction Variants
by: Yang, Jiuding, et al.
Published: (2024)

CATMark: A Context-Aware Thresholding Framework for Robust Cross-Task Watermarking in Large Language Models
by: Zhang, Yu, et al.
Published: (2025)

GENMANIP: LLM-driven Simulation for Generalizable Instruction-Following Manipulation
by: Gao, Ning, et al.
Published: (2025)

Task-Specific Data Selection for Instruction Tuning via Monosemantic Neuronal Activations
by: Ma, Da, et al.
Published: (2025)

On the Paradoxical Interference between Instruction-Following and Task Solving
by: Qi, Yunjia, et al.
Published: (2026)