Imagen de Portada

Is Bigger Edit Batch Size Always Better? -- An Empirical Study on Model Editing with Llama-3

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Yoon, Junsang, Gupta, Akshat, Anumanchipalli, Gopala
Formato:	Preprint
Publicado:	2024
Materias:	Computation and Language Artificial Intelligence Machine Learning
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

A Unified Framework for Model Editing
por: Gupta, Akshat, et al.
Publicado: (2024)

Lifelong Knowledge Editing requires Better Regularization
por: Gupta, Akshat, et al.
Publicado: (2025)

Geometric Interpretation of Layer Normalization and a Comparative Analysis with RMSNorm
por: Gupta, Akshat, et al.
Publicado: (2024)

Rebuilding ROME : Resolving Model Collapse during Sequential Model Editing
por: Gupta, Akshat, et al.
Publicado: (2024)

Model Editing at Scale leads to Gradual and Catastrophic Forgetting
por: Gupta, Akshat, et al.
Publicado: (2024)

Efficient Knowledge Editing via Minimal Precomputation
por: Gupta, Akshat, et al.
Publicado: (2025)

Evolutionary Strategies lead to Catastrophic Forgetting in LLMs
por: Abdi, Immanuel, et al.
Publicado: (2026)

Self-Assessment Tests are Unreliable Measures of LLM Personality
por: Gupta, Akshat, et al.
Publicado: (2023)

FineZip : Pushing the Limits of Large Language Models for Practical Lossless Text Compression
por: Mittu, Fazal, et al.
Publicado: (2024)

How Do LLMs Use Their Depth?
por: Gupta, Akshat, et al.
Publicado: (2025)

Norm Growth and Stability Challenges in Localized Sequential Knowledge Editing
por: Gupta, Akshat, et al.
Publicado: (2025)

The Oracle Has Spoken: A Multi-Aspect Evaluation of Dialogue in Pythia
por: Chen, Zixun, et al.
Publicado: (2025)

Resolving UnderEdit & OverEdit with Iterative & Neighbor-Assisted Model Editing
por: Baghel, Bhiman Kumar, et al.
Publicado: (2025)

Unsupervised TTS Acoustic Modeling for TTS with Conditional Disentangled Sequential VAE
por: Lian, Jiachen, et al.
Publicado: (2022)

FLEKE: Federated Locate-then-Edit Knowledge Editing
por: Zhao, Zongkai, et al.
Publicado: (2025)

PokerBench: Training Large Language Models to become Professional Poker Players
por: Zhuang, Richard, et al.
Publicado: (2025)

UltraEdit: Training-, Subject-, and Memory-Free Lifelong Editing in Language Models
por: Gu, Xiaojie, et al.
Publicado: (2025)

BiasEdit: Debiasing Stereotyped Language Models via Model Editing
por: Xu, Xin, et al.
Publicado: (2025)

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

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

Identifying Multiple Personalities in Large Language Models with External Evaluation
por: Song, Xiaoyang, et al.
Publicado: (2024)

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

Always Learning, Always Mixing: Efficient and Simple Data Mixing All The Time
por: Hu, Michael Y., et al.
Publicado: (2026)

Power Scheduler: A Batch Size and Token Number Agnostic Learning Rate Scheduler
por: Shen, Yikang, et al.
Publicado: (2024)

Power Lines: Scaling Laws for Weight Decay and Batch Size in LLM Pre-training
por: Bergsma, Shane, et al.
Publicado: (2025)

Reasoning Models Don't Always Say What They Think
por: Chen, Yanda, et al.
Publicado: (2025)

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

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

Fragile Knowledge, Robust Instruction-Following: The Width Pruning Dichotomy in Llama-3.2
por: Martra, Pere
Publicado: (2025)

MultiMedEdit: A Scenario-Aware Benchmark for Evaluating Knowledge Editing in Medical VQA
por: Wen, Shengtao, et al.
Publicado: (2025)

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

Batched Low-Rank Adaptation of Foundation Models
por: Wen, Yeming, et al.
Publicado: (2023)

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

Chain-of-Thought Reasoning In The Wild Is Not Always Faithful
por: Arcuschin, Iván, et al.
Publicado: (2025)

RSQ: Learning from Important Tokens Leads to Better Quantized LLMs
por: Sung, Yi-Lin, et al.
Publicado: (2025)

Perplexity Cannot Always Tell Right from Wrong
por: Veličković, Petar, et al.
Publicado: (2026)

Improving Summarization with Human Edits
por: Yao, Zonghai, et al.
Publicado: (2023)

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

OneEdit: A Neural-Symbolic Collaboratively Knowledge Editing System
por: Zhang, Ningyu, et al.
Publicado: (2024)

Constraining Sequential Model Editing with Editing Anchor Compression
por: Xu, Hao-Xiang, et al.
Publicado: (2025)