Search Results for author: Tiansheng Huang

Found 27 papers, 20 papers with code

Safety Tax: Safety Alignment Makes Your Large Reasoning Models Less Reasonable

1 code implementation1 Mar 2025 Tiansheng Huang, Sihao Hu, Fatih Ilhan, Selim Furkan Tekin, Zachary Yahn, Yichang Xu, Ling Liu

While safety alignment has been extensively studied for LLM, there is still a large research gap for Large Reasoning Models (LRMs) that equip with improved reasoning capability.

Language Modeling Language Modelling +2

Multi-Agent Reinforcement Learning with Focal Diversity Optimization

1 code implementation6 Feb 2025 Selim Furkan Tekin, Fatih Ilhan, Tiansheng Huang, Sihao Hu, Zachary Yahn, Ling Liu

First, we develop an agent-fusion framework for encouraging multiple LLM based agents to collaborate in producing the final inference output for each LLM query.

Diversity Multi-agent Reinforcement Learning +3

TeZO: Empowering the Low-Rankness on the Temporal Dimension in the Zeroth-Order Optimization for Fine-tuning LLMs

no code implementations31 Jan 2025 Yan Sun, Tiansheng Huang, Liang Ding, Li Shen, DaCheng Tao

Zeroth-order optimization (ZO) has demonstrated remarkable promise in efficient fine-tuning tasks for Large Language Models (LLMs).

Virus: Harmful Fine-tuning Attack for Large Language Models Bypassing Guardrail Moderation

1 code implementation29 Jan 2025 Tiansheng Huang, Sihao Hu, Fatih Ilhan, Selim Furkan Tekin, Ling Liu

By designing a new red-teaming method, we in this paper show that purely relying on the moderation guardrail for data filtration is not reliable.

Red Teaming Safety Alignment

$H^3$Fusion: Helpful, Harmless, Honest Fusion of Aligned LLMs

1 code implementation26 Nov 2024 Selim Furkan Tekin, Fatih Ilhan, Tiansheng Huang, Sihao Hu, Zachary Yahn, Ling Liu

The former penalizes the selection errors of the expert-router, and the latter mediates the expert weights drifting during fine-tuning and dynamically adjusts the fusion behavior of the resulting model by canalizing the activations on the experts.

Targeted Vaccine: Safety Alignment for Large Language Models against Harmful Fine-Tuning via Layer-wise Perturbation

1 code implementation13 Oct 2024 Guozhi Liu, Weiwei Lin, Tiansheng Huang, Ruichao Mo, Qi Mu, Li Shen

Second, instead of applying uniform perturbation across all layers, T-Vaccine only applies perturbation to the safety-critical layers while keeping other layers frozen during training.

Safety Alignment TAR

LLM-TOPLA: Efficient LLM Ensemble by Maximising Diversity

1 code implementation4 Oct 2024 Selim Furkan Tekin, Fatih Ilhan, Tiansheng Huang, Sihao Hu, Ling Liu

This paper presents LLM-TOPLA, a diversity-optimized LLM ensemble method with three unique properties: (i) We introduce the focal diversity metric to capture the diversity-performance correlation among component LLMs of an ensemble.

Diversity Ensemble Pruning +2

Harmful Fine-tuning Attacks and Defenses for Large Language Models: A Survey

2 code implementations26 Sep 2024 Tiansheng Huang, Sihao Hu, Fatih Ilhan, Selim Furkan Tekin, Ling Liu

To clear up concern, this paper provide a comprehensive overview to three aspects of harmful fine-tuning: attacks setting, defense design and evaluation methodology.

Safety Alignment

Booster: Tackling Harmful Fine-tuning for Large Language Models via Attenuating Harmful Perturbation

1 code implementation3 Sep 2024 Tiansheng Huang, Sihao Hu, Fatih Ilhan, Selim Furkan Tekin, Ling Liu

For the first time in the literature, we in this paper show that \textit{harmful perturbation} over the model weights should be the root cause of alignment-broken of harmful fine-tuning.

Antidote: Post-fine-tuning Safety Alignment for Large Language Models against Harmful Fine-tuning

3 code implementations18 Aug 2024 Tiansheng Huang, Gautam Bhattacharya, Pratik Joshi, Josh Kimball, Ling Liu

To this end, we propose Antidote, a post-fine-tuning stage solution, which remains \textbf{\textit{agnostic to the training hyper-parameters in the fine-tuning stage}}.

Philosophy Safety Alignment

Personalized Privacy Protection Mask Against Unauthorized Facial Recognition

1 code implementation19 Jul 2024 Ka-Ho Chow, Sihao Hu, Tiansheng Huang, Ling Liu

Second, we incorporate a perceptibility optimization to preserve the visual quality of the protected facial images.

Diversity Ensemble Learning +1

Lisa: Lazy Safety Alignment for Large Language Models against Harmful Fine-tuning Attack

1 code implementation28 May 2024 Tiansheng Huang, Sihao Hu, Fatih Ilhan, Selim Furkan Tekin, Ling Liu

Recent studies show that Large Language Models (LLMs) with safety alignment can be jail-broken by fine-tuning on a dataset mixed with harmful data.

Safety Alignment

Robust Few-Shot Ensemble Learning with Focal Diversity-Based Pruning

no code implementations5 Apr 2024 Selim Furkan Tekin, Fatih Ilhan, Tiansheng Huang, Sihao Hu, Ka-Ho Chow, Margaret L. Loper, Ling Liu

This paper presents FusionShot, a focal diversity optimized few-shot ensemble learning approach for boosting the robustness and generalization performance of pre-trained few-shot models.

Diversity Ensemble Learning +2

Vaccine: Perturbation-aware Alignment for Large Language Models against Harmful Fine-tuning Attack

1 code implementation2 Feb 2024 Tiansheng Huang, Sihao Hu, Ling Liu

The new paradigm of finetuning-as-a-service introduces a new attack surface for Large Language Models (LLMs): a few harmful data uploaded by users can easily trick the finetuning to produce an alignment-broken model.

Language Modelling Large Language Model

PokeLLMon: A Human-Parity Agent for Pokemon Battles with Large Language Models

1 code implementation2 Feb 2024 Sihao Hu, Tiansheng Huang, Ling Liu

We introduce PokeLLMon, the first LLM-embodied agent that achieves human-parity performance in tactical battle games, as demonstrated in Pokemon battles.

Action Generation Decision Making +1

Resource-Efficient Transformer Pruning for Finetuning of Large Models

1 code implementation CVPR 2024 Fatih Ilhan, Gong Su, Selim Furkan Tekin, Tiansheng Huang, Sihao Hu, Ling Liu

With the recent advances in vision transformers and large language models (LLMs) finetuning costly large models on downstream learning tasks poses significant challenges under limited computational resources.

Natural Language Understanding

Large Language Model-Powered Smart Contract Vulnerability Detection: New Perspectives

1 code implementation2 Oct 2023 Sihao Hu, Tiansheng Huang, Fatih İlhan, Selim Furkan Tekin, Ling Liu

The goal of auditor is to yield a broad spectrum of vulnerabilities with the hope of encompassing the correct answer, whereas the goal of critic that evaluates the validity of identified vulnerabilities is to minimize the number of false positives.

Language Modeling Language Modelling +2

Fusion of Global and Local Knowledge for Personalized Federated Learning

1 code implementation21 Feb 2023 Tiansheng Huang, Li Shen, Yan Sun, Weiwei Lin, DaCheng Tao

Personalized federated learning, as a variant of federated learning, trains customized models for clients using their heterogeneously distributed data.

Personalized Federated Learning

FedSpeed: Larger Local Interval, Less Communication Round, and Higher Generalization Accuracy

2 code implementations21 Feb 2023 Yan Sun, Li Shen, Tiansheng Huang, Liang Ding, DaCheng Tao

Federated learning is an emerging distributed machine learning framework which jointly trains a global model via a large number of local devices with data privacy protections.

Federated Learning

Adaptive Deep Neural Network Inference Optimization with EENet

1 code implementation15 Jan 2023 Fatih Ilhan, Ka-Ho Chow, Sihao Hu, Tiansheng Huang, Selim Tekin, Wenqi Wei, Yanzhao Wu, Myungjin Lee, Ramana Kompella, Hugo Latapie, Gaowen Liu, Ling Liu

Instead of having every sample go through all DNN layers during prediction, EENet learns an early exit scheduler, which can intelligently terminate the inference earlier for certain predictions, which the model has high confidence of early exit.

Inference Optimization Scheduling +1

Achieving Personalized Federated Learning with Sparse Local Models

no code implementations27 Jan 2022 Tiansheng Huang, Shiwei Liu, Li Shen, Fengxiang He, Weiwei Lin, DaCheng Tao

To counter this issue, personalized FL (PFL) was proposed to produce dedicated local models for each individual user.

Personalized Federated Learning

On Heterogeneously Distributed Data, Sparsity Matters

no code implementations29 Sep 2021 Tiansheng Huang, Shiwei Liu, Li Shen, Fengxiang He, Weiwei Lin, DaCheng Tao

Federated learning (FL) is particularly vulnerable to heterogeneously distributed data, since a common global model in FL may not adapt to the heterogeneous data distribution of each user.

Personalized Federated Learning

Adaptive Processor Frequency Adjustment for Mobile Edge Computing with Intermittent Energy Supply

no code implementations10 Feb 2021 Tiansheng Huang, Weiwei Lin, Xiaobin Hong, Xiumin Wang, Qingbo Wu, Rui Li, Ching-Hsien Hsu, Albert Y. Zomaya

With astonishing speed, bandwidth, and scale, Mobile Edge Computing (MEC) has played an increasingly important role in the next generation of connectivity and service delivery.

Deep Reinforcement Learning Edge-computing

Stochastic Client Selection for Federated Learning with Volatile Clients

no code implementations17 Nov 2020 Tiansheng Huang, Weiwei Lin, Li Shen, Keqin Li, Albert Y. Zomaya

Federated Learning (FL), arising as a privacy-preserving machine learning paradigm, has received notable attention from the public.

Fairness Federated Learning +1

An Efficiency-boosting Client Selection Scheme for Federated Learning with Fairness Guarantee

no code implementations3 Nov 2020 Tiansheng Huang, Weiwei Lin, Wentai Wu, Ligang He, Keqin Li, Albert Y. Zomaya

The client selection policy is critical to an FL process in terms of training efficiency, the final model's quality as well as fairness.

Distributed Computing Fairness +1

Cannot find the paper you are looking for? You can Submit a new open access paper.