Imagen de Portada

Llama SLayer 8B: Shallow Layers Hold the Key to Knowledge Injection

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Chen, Tianxiang, Tan, Zhentao, Gong, Tao, Wu, Yue, Chu, Qi, Liu, Bin, Ye, Jieping, Yu, Nenghai
Formato:	Preprint
Publicado:	2024
Materias:	Computation and Language
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Flora: Effortless Context Construction to Arbitrary Length and Scale
por: Chen, Tianxiang, et al.
Publicado: (2025)

MiM-ISTD: Mamba-in-Mamba for Efficient Infrared Small Target Detection
por: Chen, Tianxiang, et al.
Publicado: (2024)

Bootstrapping Audio-Visual Segmentation by Strengthening Audio Cues
por: Chen, Tianxiang, et al.
Publicado: (2024)

TCI-Former: Thermal Conduction-Inspired Transformer for Infrared Small Target Detection
por: Chen, Tianxiang, et al.
Publicado: (2024)

Learning to Focus and Precise Cropping: A Reinforcement Learning Framework with Information Gaps and Grounding Loss for MLLMs
por: Zhao, Xuanpu, et al.
Publicado: (2026)

Structure-aware Domain Knowledge Injection for Large Language Models
por: Liu, Kai, et al.
Publicado: (2024)

When Agents Look the Same: Quantifying Distillation-Induced Similarity in Tool-Use Behaviors
por: Yang, Chenghao, et al.
Publicado: (2026)

Llama Scope: Extracting Millions of Features from Llama-3.1-8B with Sparse Autoencoders
por: He, Zhengfu, et al.
Publicado: (2024)

Multi-spectral Class Center Network for Face Manipulation Detection and Localization
por: Miao, Changtao, et al.
Publicado: (2023)

Llama-Mob: Instruction-Tuning Llama-3-8B Excels in City-Scale Mobility Prediction
por: Tang, Peizhi, et al.
Publicado: (2024)

MGH Radiology Llama: A Llama 3 70B Model for Radiology
por: Shi, Yucheng, et al.
Publicado: (2024)

Context-Aware Weakly Supervised Image Manipulation Localization with SAM Refinement
por: Wang, Xinghao, et al.
Publicado: (2025)

Freeze Deep, Train Shallow: Interpretable Layer Allocation for Continued Pre-Training
por: Wu, Yu-Hang, et al.
Publicado: (2026)

Shallow Preference Signals: Large Language Model Aligns Even Better with Truncated Data?
por: Qi, Xuan, et al.
Publicado: (2025)

Towards More Unified In-context Visual Understanding
por: Sheng, Dianmo, et al.
Publicado: (2023)

INSIDE: LLMs' Internal States Retain the Power of Hallucination Detection
por: Chen, Chao, et al.
Publicado: (2024)

BadLlama: cheaply removing safety fine-tuning from Llama 2-Chat 13B
por: Gade, Pranav, et al.
Publicado: (2023)

Latent Paraphrasing: Perturbation on Layers Improves Knowledge Injection in Language Models
por: Kang, Minki, et al.
Publicado: (2024)

ChocoLlama: Lessons Learned From Teaching Llamas Dutch
por: Meeus, Matthieu, et al.
Publicado: (2024)

Mixture-of-Noises Enhanced Forgery-Aware Predictor for Multi-Face Manipulation Detection and Localization
por: Miao, Changtao, et al.
Publicado: (2024)

MARS-Bench: A Multi-turn Athletic Real-world Scenario Benchmark for Dialogue Evaluation
por: Yang, Chenghao, et al.
Publicado: (2025)

Targeted Lexical Injection: Unlocking Latent Cross-Lingual Alignment in Lugha-Llama via Early-Layer LoRA Fine-Tuning
por: Ngugi, Stanley
Publicado: (2025)

SLayerGen: a Crystal Generative Model for all Space and Layer Groups
por: Chang, Rees, et al.
Publicado: (2026)

Teaching Llama a New Language Through Cross-Lingual Knowledge Transfer
por: Kuulmets, Hele-Andra, et al.
Publicado: (2024)

Extending Llama-3's Context Ten-Fold Overnight
por: Zhang, Peitian, et al.
Publicado: (2024)

SpindleKV: A Novel KV Cache Reduction Method Balancing Both Shallow and Deep Layers
por: Tang, Zicong, et al.
Publicado: (2025)

DR.EHR: Dense Retrieval for Electronic Health Record with Knowledge Injection and Synthetic Data
por: Zhao, Zhengyun, et al.
Publicado: (2025)

Llama See, Llama Do: A Mechanistic Perspective on Contextual Entrainment and Distraction in LLMs
por: Niu, Jingcheng, et al.
Publicado: (2025)

KORE: Enhancing Knowledge Injection for Large Multimodal Models via Knowledge-Oriented Controls
por: Jiang, Kailin, et al.
Publicado: (2025)

Provably Secure Disambiguating Neural Linguistic Steganography
por: Qi, Yuang, et al.
Publicado: (2024)

TableLlama: Towards Open Large Generalist Models for Tables
por: Zhang, Tianshu, et al.
Publicado: (2023)

Zebra-Llama: A Context-Aware Large Language Model for Democratizing Rare Disease Knowledge
por: Soman, Karthik, et al.
Publicado: (2024)

Code Llama: Open Foundation Models for Code
por: Rozière, Baptiste, et al.
Publicado: (2023)

Llama-Embed-Nemotron-8B: A Universal Text Embedding Model for Multilingual and Cross-Lingual Tasks
por: Babakhin, Yauhen, et al.
Publicado: (2025)

SAC-KG: Exploiting Large Language Models as Skilled Automatic Constructors for Domain Knowledge Graphs
por: Chen, Hanzhu, et al.
Publicado: (2024)

Question-guided Knowledge Graph Re-scoring and Injection for Knowledge Graph Question Answering
por: Zhang, Yu, et al.
Publicado: (2024)

Efficient Speculative Decoding for Llama at Scale: Challenges and Solutions
por: Tang, Bangsheng, et al.
Publicado: (2025)

LlamaCare: A Large Medical Language Model for Enhancing Healthcare Knowledge Sharing
por: Sun, Maojun
Publicado: (2024)

Understanding Parametric Knowledge Injection in Retrieval-Augmented Generation
por: Tang, Minghao, et al.
Publicado: (2025)

Bidirectional LMs are Better Knowledge Memorizers? A Benchmark for Real-world Knowledge Injection
por: Zhang, Yuwei, et al.
Publicado: (2025)