paper-with-me

홈 › Papers

TeleLoRA: Teleporting Model-Specific Alignment Across LLMs

2025-03-26 · Xiao Lin, Manoj Acharya, Anirban Roy, Susmit Jha

Mitigating Trojans in Large Language Models (LLMs) is one of many tasks where alignment data is LLM specific, as different LLMs have different Trojan triggers and trigger behaviors to be removed. In this paper, we introduce TeleLoRA (Teleporting Low-Rank Adaptation), a novel framework that synergizes model-specific alignment data across multiple LLMs to enable zero-shot Trojan mitigation on unseen LLMs without alignment data. TeleLoRA learns a unified generator of LoRA adapter weights by leveraging local activation information across multiple LLMs. This generator is designed to be permutation symmetric to generalize across models with different architectures and sizes. We optimize the model design for memory efficiency, making it feasible to learn with large-scale LLMs with minimal computational resources. Experiments on LLM Trojan mitigation benchmarks demonstrate that TeleLoRA effectively reduces attack success rates while preserving the benign performance of the models.

📄 PDF Abstract BibTeX arXiv:2503.20228

Code (0)

등록된 구현이 없습니다.

Tasks

model

Methods 이 논문이 사용한 방법론

Adapter 설명 없음

Similar Papers 제목 키워드 기반

ABRA: Teleporting Fine-Tuned Knowledge Across Domains for Open-Vocabulary Object Detection

2026-03-12 · Mattia Bernardi, Chiara Cappellino, Matteo Mosconi, Enver Sangineto 외 arxiv

Although recent Open-Vocabulary Object Detection architectures, such as Grounding DINO, demonstrate strong zero-shot capabilities, their performance degrades significantly under domain shifts. Moreover, many domains of p…

Object Detection

Neural Fields as World Models

2026-02-21 · Joshua Nunley arxiv

Humans rehearse possible futures offline, as in mental practice and perhaps dreaming, suggesting that world models may support task learning away from the environment. Standard machine learning world models compress visu…

AR-MAP: Are Autoregressive Large Language Models Implicit Teachers for Diffusion Large Language Models?

2026-02-02 · Liang Lin, Feng Xiong, Zengbin Wang, Kun Wang 외 arxiv

Diffusion Large Language Models (DLLMs) have emerged as a powerful alternative to autoregressive models, enabling parallel token generation across multiple positions. However, preference alignment of DLLMs remains challe…

Transfer Learning

Improving the Distributional Alignment of LLMs using Supervision

2025-07-01 · Gauri Kambhatla, Sanjana Gautam, Angela Zhang, Alex Liu 외 arxiv

The ability to accurately align LLMs with diverse population groups on subjective questions would have great value. In this work, we show that adding simple supervision can more consistently improve the alignment of LLM-…

Improving Adversarial Transferability in MLLMs via Dynamic Vision-Language Alignment Attack

2025-02-27 · Chenhe Gu, Jindong Gu, Andong Hua, Yao Qin

Multimodal Large Language Models (MLLMs), built upon LLMs, have recently gained attention for their capabilities in image recognition and understanding. However, while MLLMs are vulnerable to adversarial attacks, the tra…