paper-with-me

홈 › Papers

RE-Adapt: Reverse Engineered Adaptation of Large Language Models

2024-05-23 · William Fleshman, Benjamin Van Durme

We introduce RE-Adapt, an approach to fine-tuning large language models on new domains without degrading any pre-existing instruction-tuning. We reverse engineer an adapter which isolates what an instruction-tuned model has learned beyond its corresponding pretrained base model. Importantly, this requires no additional data or training. We can then fine-tune the base model on a new domain and readapt it to instruction following with the reverse engineered adapter. RE-Adapt and our low-rank variant LoRE-Adapt both outperform other methods of fine-tuning, across multiple popular LLMs and datasets, even when the models are used in conjunction with retrieval-augmented generation.

📄 PDF Abstract BibTeX arXiv:2405.15007

Code (0)

등록된 구현이 없습니다.

Tasks

Instruction FollowingRetrievalRetrieval-augmented Generation

Methods 이 논문이 사용한 방법론

BASE 설명 없음
Adapter 설명 없음

Similar Papers 제목 키워드 기반

RE-AdaptIR: Improving Information Retrieval through Reverse Engineered Adaptation

2024-06-20 · William Fleshman, Benjamin Van Durme

Large language models (LLMs) fine-tuned for text-retrieval have demonstrated state-of-the-art results across several information retrieval (IR) benchmarks. However, supervised training for improving these models requires…

Information RetrievalRetrievalText Retrieval

LLM4CodeRE: Generative AI for Code Decompilation Analysis and Reverse Engineering

2026-04-07 · Hamed Jelodar, Samita Bai, Tochukwu Emmanuel Nwankwo, Parisa Hamedi 외 arxiv

Code decompilation analysis is a fundamental yet challenging task in malware reverse engineering, particularly due to the pervasive use of sophisticated obfuscation techniques. Although recent large language models (LLMs…

Classification of Reverse-Engineered Class Diagram and Forward-Engineered Class Diagram using Machine Learning

2020-11-14 · Kaushil Mangaroliya, Het Patel

UML Class diagram is very important to visualize the whole software we are working on and helps understand the whole system in the easiest way possible by showing the system classes, its attributes, methods, and relation…

BIG-bench Machine LearningGeneral Classification

VPA: Fully Test-Time Visual Prompt Adaptation

2023-09-26 · Jiachen Sun, Mark Ibrahim, Melissa Hall, Ivan Evtimov 외

Textual prompt tuning has demonstrated significant performance improvements in adapting natural language processing models to a variety of downstream tasks by treating hand-engineered prompts as trainable parameters. Ins…

Domain AdaptationPseudo LabelTest-time AdaptationVisual Prompting+2

Learning to Learn-at-Test-Time: Language Agents with Learnable Adaptation Policies

2026-04-01 · Zhanzhi Lou, Hui Chen, Yibo Li, Qian Wang 외 arxiv

Test-Time Learning (TTL) enables language agents to iteratively refine their performance through repeated interactions with the environment at inference time. At the core of TTL is an adaptation policy that updates the a…