Cover Image

Fragile Knowledge, Robust Instruction-Following: The Width Pruning Dichotomy in Llama-3.2

Fuente: arXiv

Saved in:

Bibliographic Details
Main Author:	Martra, Pere
Format:	Preprint
Published:	2025
Subjects:	Computation and Language Artificial Intelligence Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

FragileFlow: Spectral Control of Correct-but-Fragile Predictions for Foundation Model Robustness
by: Li, Zhuoyun, et al.
Published: (2026)

Open Llama2 Model for the Lithuanian Language
by: Nakvosas, Artūras, et al.
Published: (2024)

Physics of Language Models: Part 3.2, Knowledge Manipulation
by: Allen-Zhu, Zeyuan, et al.
Published: (2023)

Steering Llama 2 via Contrastive Activation Addition
by: Panickssery, Nina, et al.
Published: (2023)

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

Financial Instruction Following Evaluation (FIFE)
by: Matlin, Glenn, et al.
Published: (2025)

Llama-Nemotron: Efficient Reasoning Models
by: Bercovich, Akhiad, et al.
Published: (2025)

Mixture of Universal Experts: Scaling Virtual Width via Depth-Width Transformation
by: Chen, Yilong, et al.
Published: (2026)

Is In-Context Learning Sufficient for Instruction Following in LLMs?
by: Zhao, Hao, et al.
Published: (2024)

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

BanglaLlama: LLaMA for Bangla Language
by: Zehady, Abdullah Khan, et al.
Published: (2024)

Compact Language Models via Pruning and Knowledge Distillation
by: Muralidharan, Saurav, et al.
Published: (2024)

ReIFE: Re-evaluating Instruction-Following Evaluation
by: Liu, Yixin, et al.
Published: (2024)

From Language Modeling to Instruction Following: Understanding the Behavior Shift in LLMs after Instruction Tuning
by: Wu, Xuansheng, et al.
Published: (2023)

Instruction Following by Principled Boosting Attention of Large Language Models
by: Guardieiro, Vitoria, et al.
Published: (2025)

Improving Instruction-Following in Language Models through Activation Steering
by: Stolfo, Alessandro, et al.
Published: (2024)

Infer Human's Intentions Before Following Natural Language Instructions
by: Wan, Yanming, et al.
Published: (2024)

Inspection and Control of Self-Generated-Text Recognition Ability in Llama3-8b-Instruct
by: Ackerman, Christopher, et al.
Published: (2024)

Is Bigger Edit Batch Size Always Better? -- An Empirical Study on Model Editing with Llama-3
by: Yoon, Junsang, et al.
Published: (2024)

Instruction-Tuned, but Not More Verifiable Instruction-Following: A Cross-Task Diagnosis for LoRA Adapters
by: Zou, Junyi
Published: (2026)

Lugha-Llama: Adapting Large Language Models for African Languages
by: Buzaaba, Happy, et al.
Published: (2025)

Efficient Mathematical Reasoning Models via Dynamic Pruning and Knowledge Distillation
by: Yu, Fengming, et al.
Published: (2025)

Replay Failures as Successes: Sample-Efficient Reinforcement Learning for Instruction Following
by: Zhang, Kongcheng, et al.
Published: (2025)

From Pruning to Grafting: Dynamic Knowledge Redistribution via Learnable Layer Fusion
by: Pei, Zehua, et al.
Published: (2024)

A Practice of Post-Training on Llama-3 70B with Optimal Selection of Additional Language Mixture Ratio
by: Xi, Ningyuan, et al.
Published: (2024)

Forbidden Facts: An Investigation of Competing Objectives in Llama-2
by: Wang, Tony T., et al.
Published: (2023)

Conifer: Improving Complex Constrained Instruction-Following Ability of Large Language Models
by: Sun, Haoran, et al.
Published: (2024)

Improving Instruction Following in Language Models through Proxy-Based Uncertainty Estimation
by: Lee, JoonHo, et al.
Published: (2024)

Badllama 3: removing safety finetuning from Llama 3 in minutes
by: Volkov, Dmitrii
Published: (2024)

Instruction-tuned Language Models are Better Knowledge Learners
by: Jiang, Zhengbao, et al.
Published: (2024)

SAIF: A Sparse Autoencoder Framework for Interpreting and Steering Instruction Following of Language Models
by: He, Zirui, et al.
Published: (2025)

Pragmatic Instruction Following and Goal Assistance via Cooperative Language-Guided Inverse Planning
by: Zhi-Xuan, Tan, et al.
Published: (2024)

Depth-Width tradeoffs in Algorithmic Reasoning of Graph Tasks with Transformers
by: Yehudai, Gilad, et al.
Published: (2025)

Domain Adaptation of Llama3-70B-Instruct through Continual Pre-Training and Model Merging: A Comprehensive Evaluation
by: Siriwardhana, Shamane, et al.
Published: (2024)

Light-IF: Endowing LLMs with Generalizable Reasoning via Preview and Self-Checking for Complex Instruction Following
by: Wang, Chenyang, et al.
Published: (2025)

GPT-3 Powered Information Extraction for Building Robust Knowledge Bases
by: Choudhury, Ritabrata Roy, et al.
Published: (2024)

SPaR: Self-Play with Tree-Search Refinement to Improve Instruction-Following in Large Language Models
by: Cheng, Jiale, et al.
Published: (2024)

OmniJARVIS: Unified Vision-Language-Action Tokenization Enables Open-World Instruction Following Agents
by: Wang, Zihao, et al.
Published: (2024)

LLM Self-Correction with DeCRIM: Decompose, Critique, and Refine for Enhanced Following of Instructions with Multiple Constraints
by: Ferraz, Thomas Palmeira, et al.
Published: (2024)

IDEA Prune: An Integrated Enlarge-and-Prune Pipeline in Generative Language Model Pretraining
by: Li, Yixiao, et al.
Published: (2025)