paper-with-me

홈 › Papers

Activated LoRA: Fine-tuned LLMs for Intrinsics

2025-04-16 · Kristjan Greenewald, Luis Lastras, Thomas Parnell, Vraj Shah, Lucian Popa, Giulio Zizzo, Chulaka Gunasekara, Ambrish Rawat, David Cox

Low-Rank Adaptation (LoRA) has emerged as a highly efficient framework for finetuning the weights of large foundation models, and has become the go-to method for data-driven customization of LLMs. Despite the promise of highly customized behaviors and capabilities, switching between relevant LoRAs in a multiturn setting is highly inefficient, as the key-value (KV) cache of the entire turn history must be recomputed with the LoRA weights before generation can begin. To address this problem, we propose Activated LoRA (aLoRA), which modifies the LoRA framework to only adapt weights for the tokens in the sequence \emph{after} the aLoRA is invoked. This change crucially allows aLoRA to accept the base model's KV cache of the input string, meaning that aLoRA can be instantly activated whenever needed in a chain without recomputing the cache. This enables building what we call \emph{intrinsics}, i.e. highly specialized models invoked to perform well-defined operations on portions of an input chain or conversation that otherwise uses the base model by default. We use aLoRA to train a set of intrinsics models, demonstrating competitive accuracy with standard LoRA while achieving significant inference benefits.

📄 PDF Abstract BibTeX arXiv:2504.12397

Code (1)

IBM/activated-lora 공식 구현 pytorch

Methods 이 논문이 사용한 방법론

BASE 설명 없음
SET Dynamic Sparse Training method where weight mask is updated randomly periodically

Similar Papers 제목 키워드 기반

LoRA Land: 310 Fine-tuned LLMs that Rival GPT-4, A Technical Report

2024-04-29 · Justin Zhao, Timothy Wang, Wael Abid, Geoffrey Angus 외

Low Rank Adaptation (LoRA) has emerged as one of the most widely adopted methods for Parameter Efficient Fine-Tuning (PEFT) of Large Language Models (LLMs). LoRA reduces the number of trainable parameters and memory usag…

GPUparameter-efficient fine-tuning

LLaVA-MoLE: Sparse Mixture of LoRA Experts for Mitigating Data Conflicts in Instruction Finetuning MLLMs

2024-01-29 · Shaoxiang Chen, Zequn Jie, Lin Ma

Instruction finetuning on a variety of image-text instruction data is the key to obtaining a versatile Multimodal Large Language Model (MLLM), and different configurations of the instruction data can lead to finetuned mo…

Language ModellingLarge Language ModelMixture-of-ExpertsMultimodal Large Language Model

Finetuning-Activated Backdoors in LLMs

2025-05-22 · Thibaud Gloaguen, Mark Vero, Robin Staab, Martin Vechev

Finetuning openly accessible Large Language Models (LLMs) has become standard practice for achieving task-specific performance improvements. Until now, finetuning has been regarded as a controlled and secure process in w…

Meta-Learning

Bayesian Low-rank Adaptation for Large Language Models

2023-08-24 · Adam X. Yang, Maxime Robeyns, Xi Wang, Laurence Aitchison

Low-rank adaptation (LoRA) has emerged as a new paradigm for cost-efficient fine-tuning of large language models (LLMs). However, fine-tuned LLMs often become overconfident especially when fine-tuned on small datasets. B…

Investigating the Effectiveness of Task-Agnostic Prefix Prompt for Instruction Following

2023-02-28 · Seonghyeon Ye, Hyeonbin Hwang, Sohee Yang, Hyeongu Yun 외

In this paper, we present our finding that prepending a Task-Agnostic Prefix Prompt (TAPP) to the input improves the instruction-following ability of various Large Language Models (LLMs) during inference. TAPP is differe…

Instruction FollowingZero-shot Generalization