paper-with-me

Papers

IKnow: Instruction-Knowledge-Aware Continual Pretraining for Effective Domain Adaptation

2025-10-23 · Tianyi Zhang, Florian Mai, Lucie Flek arxiv

Continual pretraining promises to adapt large language models (LLMs) to new domains using only unlabeled test-time data, but naively applying standard self-supervised objectives to instruction-tuned models is known to degrade their instruction-following capability and semantic representations. Existing fixes assume access to the original base model or rely on knowledge from an external domain-specific database - both of which pose a realistic barrier in settings where the base model weights are withheld for safety reasons or reliable external corpora are unavailable. In this work, we propose Instruction-Knowledge-Aware Continual Adaptation (IKnow), a simple and general framework that formulates novel self-supervised objectives in the instruction-response dialogue format. Rather than depend- ing on external resources, IKnow leverages domain knowledge embedded within the text itself and learns to encode it at a deeper semantic level.

📄 PDF Abstract BibTeX arXiv:2510.20377

Code (0)

등록된 구현이 없습니다.

Tasks

Continual PretrainingDomain Adaptation

Similar Papers 제목 키워드 기반

UniKnow: A Unified Framework for Reliable Language Model Behavior across Parametric and External Knowledge

2025-02-19 · Youna Kim, Hyuhng Joon Kim, Minjoon Choi, Sungmin Cho 외

Language models often benefit from external knowledge beyond parametric knowledge. While this combination enhances performance, achieving reliable knowledge utilization remains challenging, as it requires assessing the s…

InformativenessLanguage ModelingLanguage Modelling

LLaVA-c: Continual Improved Visual Instruction Tuning

2025-06-10 · Wenzhuo LIU, Fei Zhu, Haiyang Guo, Longhui Wei 외

Multimodal models like LLaVA-1.5 achieve state-of-the-art visual understanding through visual instruction tuning on multitask datasets, enabling strong instruction-following and multimodal performance. However, multitask…

Continual LearningContinual PretrainingInstruction Following

The Construction of Instruction-tuned LLMs for Finance without Instruction Data Using Continual Pretraining and Model Merging

2024-09-30 · Masanori Hirano, Kentaro Imajo

This paper proposes a novel method for constructing instruction-tuned large language models (LLMs) for finance without instruction data. Traditionally, developing such domain-specific LLMs has been resource-intensive, re…

Continual Pretraining

Typhoon: Thai Large Language Models

2023-12-21 · Kunat Pipatanakul, Phatrasek Jirabovonvisut, Potsawee Manakul, Sittipong Sripaisarnmongkol 외

Typhoon is a series of Thai large language models (LLMs) developed specifically for the Thai language. This technical report presents challenges and insights in developing Thai LLMs, including data preparation, pretraini…

Question AnsweringWorld Knowledge

Instruction-based Hypergraph Pretraining

2024-03-28 · Mingdai Yang, Zhiwei Liu, Liangwei Yang, Xiaolong Liu 외

Pretraining has been widely explored to augment the adaptability of graph learning models to transfer knowledge from large datasets to a downstream task, such as link prediction or classification. However, the gap betwee…

DiversityGraph LearningLink PredictionRepresentation Learning