Towards Federated RLHF with Aggregated Client Preference for LLMs

• URL: http://arxiv.org/abs/2407.03038v3
• Date: Tue, 08 Apr 2025 18:13:57 GMT
• Title: Towards Federated RLHF with Aggregated Client Preference for LLMs
• Authors: Feijie Wu, Xiaoze Liu, Haoyu Wang, Xingchen Wang, Lu Su, Jing Gao,
• Abstract summary: Reinforcement learning with human feedback (RLHF) fine-tunes a pretrained large language model (LLM) using user preference data.<n>Due to privacy concerns, users may be reluctant to share sensitive preference data.<n>We propose utilizing Federated Learning (FL) techniques, allowing large-scale preference collection from diverse real-world users.
• Score: 16.97734775088073
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Reinforcement learning with human feedback (RLHF) fine-tunes a pretrained large language model (LLM) using user preference data, enabling it to generate content aligned with human preferences. However, due to privacy concerns, users may be reluctant to share sensitive preference data. To address this, we propose utilizing Federated Learning (FL) techniques, allowing large-scale preference collection from diverse real-world users without requiring them to transmit data to a central server. Our federated RLHF methods (i.e., FedBis and FedBiscuit) encode each client's preferences into binary selectors and aggregate them to capture common preferences. In particular, FedBiscuit overcomes key challenges, such as preference heterogeneity and reward hacking, through innovative solutions like grouping clients with similar preferences to reduce heterogeneity and using multiple binary selectors to enhance LLM output quality. To evaluate the performance of the proposed methods, we establish the first federated RLHF benchmark with a heterogeneous human preference dataset. Experimental results show that by integrating the LLM with aggregated client preferences, FedBis and FedBiscuit significantly enhance the professionalism and readability of the generated content.

Related papers

• FSPO: Few-Shot Preference Optimization of Synthetic Preference Data in LLMs Elicits Effective Personalization to Real Users [111.56469697145519]
  We propose Few-Shot Preference Optimization, which reframes reward modeling as a meta-learning problem. Under this framework, an LLM learns to quickly adapt to a user via a few labeled preferences from that user, constructing a personalized reward function for them. We generate over 1M synthetic personalized preferences using publicly available LLMs. We evaluate FSPO on personalized open-ended generation for up to 1,500 synthetic users across three domains: movie reviews, pedagogical adaptation based on educational background, and general question answering, along with a controlled human study.
  arXiv  Detail & Related papers  (2025-02-26T17:08:46Z)
• Federated Fine-Tuning of LLMs: Framework Comparison and Research Directions [59.5243730853157]
  Federated learning (FL) provides a privacy-preserving solution for fine-tuning pre-trained large language models (LLMs) using distributed private datasets. This article conducts a comparative analysis of three advanced federated LLM (FedLLM) frameworks that integrate knowledge distillation (KD) and split learning (SL) to mitigate these issues.
  arXiv  Detail & Related papers  (2025-01-08T11:37:06Z)
• Few-shot Steerable Alignment: Adapting Rewards and LLM Policies with Neural Processes [50.544186914115045]
  Large language models (LLMs) are increasingly embedded in everyday applications. Ensuring their alignment with the diverse preferences of individual users has become a critical challenge. We present a novel framework for few-shot steerable alignment.
  arXiv  Detail & Related papers  (2024-12-18T16:14:59Z)
• FedReMa: Improving Personalized Federated Learning via Leveraging the Most Relevant Clients [13.98392319567057]
  Federated Learning (FL) is a distributed machine learning paradigm that achieves a globally robust model through decentralized computation and periodic model synthesis. Despite their wide adoption, existing FL and PFL works have yet to comprehensively address the class-imbalance issue. We propose FedReMa, an efficient PFL algorithm that can tackle class-imbalance by utilizing an adaptive inter-client co-learning approach.
  arXiv  Detail & Related papers  (2024-11-04T05:44:28Z)
• Co-clustering for Federated Recommender System [33.70723179405055]
  Federated Recommender System (FRS) offers a solution that strikes a balance between providing high-quality recommendations and preserving user privacy. The presence of statistical heterogeneity in FRS, commonly observed due to personalized decision-making patterns, can pose challenges. We propose CoFedRec, a novel Co-clustering Federated Recommendation mechanism.
  arXiv  Detail & Related papers  (2024-11-03T21:32:07Z)
• PersonalLLM: Tailoring LLMs to Individual Preferences [11.717169516971856]
  We present a public benchmark, PersonalLLM, focusing on adapting LLMs to provide maximal benefits for a particular user. We curate open-ended prompts paired with many high-quality answers over which users would be expected to display heterogeneous latent preferences. Our dataset and generated personalities offer an innovative testbed for developing personalization algorithms.
  arXiv  Detail & Related papers  (2024-09-30T13:55:42Z)
• Aligning Large Language Models with Self-generated Preference Data [72.99676237703099]
  We propose a new framework that boosts the alignment of large language models (LLMs) with human preferences. Our key idea is leveraging the human prior knowledge within the small (seed) data. We introduce a noise-aware preference learning algorithm to mitigate the risk of low quality within generated preference data.
  arXiv  Detail & Related papers  (2024-06-06T18:01:02Z)
• FedSelect: Personalized Federated Learning with Customized Selection of Parameters for Fine-Tuning [9.22574528776347]
  FedSelect is a novel PFL algorithm inspired by the iterative subnetwork discovery procedure used for the Lottery Ticket Hypothesis. We show that FedSelect outperforms recent state-of-the-art PFL algorithms under challenging client data heterogeneity settings.
  arXiv  Detail & Related papers  (2024-04-03T05:36:21Z)
• FedClust: Optimizing Federated Learning on Non-IID Data through Weight-Driven Client Clustering [28.057411252785176]
  Federated learning (FL) is an emerging distributed machine learning paradigm enabling collaborative model training on decentralized devices without exposing their local data. This paper proposes FedClust, a novel CFL approach leveraging correlations between local model weights and client data distributions.
  arXiv  Detail & Related papers  (2024-03-07T01:50:36Z)
• Personalized Federated Learning with Attention-based Client Selection [57.71009302168411]
  We propose FedACS, a new PFL algorithm with an Attention-based Client Selection mechanism. FedACS integrates an attention mechanism to enhance collaboration among clients with similar data distributions. Experiments on CIFAR10 and FMNIST validate FedACS's superiority.
  arXiv  Detail & Related papers  (2023-12-23T03:31:46Z)
• FedRec+: Enhancing Privacy and Addressing Heterogeneity in Federated Recommendation Systems [15.463595798992621]
  FedRec+ is an ensemble framework for federated recommendation systems. It enhances privacy and reduces communication costs for edge users. Experimental results demonstrate the state-of-the-art performance of FedRec+.
  arXiv  Detail & Related papers  (2023-10-31T05:36:53Z)
• FedJETs: Efficient Just-In-Time Personalization with Federated Mixture of Experts [48.78037006856208]
  FedJETs is a novel solution by using a Mixture-of-Experts (MoE) framework within a Federated Learning (FL) setup. Our method leverages the diversity of the clients to train specialized experts on different subsets of classes, and a gating function to route the input to the most relevant expert(s) Our approach can improve accuracy up to 18% in state of the art FL settings, while maintaining competitive zero-shot performance.
  arXiv  Detail & Related papers  (2023-06-14T15:47:52Z)
• FedABC: Targeting Fair Competition in Personalized Federated Learning [76.9646903596757]
  Federated learning aims to collaboratively train models without accessing their client's local private data. We propose a novel and generic PFL framework termed Federated Averaging via Binary Classification, dubbed FedABC. In particular, we adopt the one-vs-all'' training strategy in each client to alleviate the unfair competition between classes.
  arXiv  Detail & Related papers  (2023-02-15T03:42:59Z)
• Fed-CBS: A Heterogeneity-Aware Client Sampling Mechanism for Federated Learning via Class-Imbalance Reduction [76.26710990597498]
  We show that the class-imbalance of the grouped data from randomly selected clients can lead to significant performance degradation. Based on our key observation, we design an efficient client sampling mechanism, i.e., Federated Class-balanced Sampling (Fed-CBS) In particular, we propose a measure of class-imbalance and then employ homomorphic encryption to derive this measure in a privacy-preserving way.
  arXiv  Detail & Related papers  (2022-09-30T05:42:56Z)
• FLIS: Clustered Federated Learning via Inference Similarity for Non-IID Data Distribution [7.924081556869144]
  We present a new algorithm, FLIS, which groups the clients population in clusters with jointly trainable data distributions. We present experimental results to demonstrate the benefits of FLIS over the state-of-the-art benchmarks on CIFAR-100/10, SVHN, and FMNIST datasets.
  arXiv  Detail & Related papers  (2022-08-20T22:10:48Z)
• FedSPLIT: One-Shot Federated Recommendation System Based on Non-negative Joint Matrix Factorization and Knowledge Distillation [7.621960305708476]
  We present the first unsupervised one-shot federated CF implementation, named FedSPLIT, based on NMF joint factorization. FedSPLIT can obtain similar results than the state of the art (and even outperform it in certain situations) with a substantial decrease in the number of communications.
  arXiv  Detail & Related papers  (2022-05-04T23:42:14Z)
• FedCL: Federated Contrastive Learning for Privacy-Preserving Recommendation [98.5705258907774]
  FedCL can exploit high-quality negative samples for effective model training with privacy well protected. We first infer user embeddings from local user data through the local model on each client, and then perturb them with local differential privacy (LDP) Since individual user embedding contains heavy noise due to LDP, we propose to cluster user embeddings on the server to mitigate the influence of noise.
  arXiv  Detail & Related papers  (2022-04-21T02:37:10Z)
• On the Convergence of Clustered Federated Learning [57.934295064030636]
  In a federated learning system, the clients, e.g. mobile devices and organization participants, usually have different personal preferences or behavior patterns. This paper proposes a novel weighted client-based clustered FL algorithm to leverage the client's group and each client in a unified optimization framework.
  arXiv  Detail & Related papers  (2022-02-13T02:39:19Z)
• FedGroup: Efficient Clustered Federated Learning via Decomposed Data-Driven Measure [18.083188787905083]
  We propose a novel clustered federated learning (CFL) framework FedGroup. We show that FedGroup can significantly improve absolute test accuracy by +14.1% on FEMNIST compared to FedAvg. We also evaluate FedGroup and FedGrouProx (combined with FedProx) on several open datasets.
  arXiv  Detail & Related papers  (2020-10-14T08:15:34Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.