# Federated Learning with Ad-hoc Adapter Insertions: The Case of Soft-Embeddings for Training Classifier-as-Retriever

arXiv: https://arxiv.org/abs/2509.16508 · HTML: https://mlsanity.com/paper/2509.16508/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in RAG pipeline design and evaluation, one of 15 directions in Retrieval-augmented generation.
- Builds on 24 of its 35 references that are in the map.
- First posted on arXiv 20 Sep 2025.

## Placed in

- Retrieval-augmented generation / [RAG pipeline design and evaluation](https://mlsanity.com/rag/rag-pipeline-design-and-evaluation/)

## Builds on (most cited first, 24 of 24)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [Communication-Efficient Learning of Deep Networks from Decentralized Data](https://mlsanity.com/paper/1602.05629/) (2016)
- [Parameter-Efficient Transfer Learning for NLP](https://mlsanity.com/paper/1902.00751/) (2019)
- [Prefix-Tuning: Optimizing Continuous Prompts for Generation](https://mlsanity.com/paper/2101.00190/) (2021)
- [The Power of Scale for Parameter-Efficient Prompt Tuning](https://mlsanity.com/paper/2104.08691/) (2021)
- [Advances and Open Problems in Federated Learning](https://mlsanity.com/paper/1912.04977/) (2019)
- [Retrieval-Augmented Generation for Large Language Models: A Survey](https://mlsanity.com/paper/2312.10997/) (2023)
- [Language Models as Knowledge Bases?](https://mlsanity.com/paper/1909.01066/) (2019)
- [ColBERT: Efficient and Effective Passage Search via Contextualized Late Interaction over BERT](https://mlsanity.com/paper/2004.12832/) (2020)
- [Optimization Methods for Large-Scale Machine Learning](https://mlsanity.com/paper/1606.04838/) (2016)
- [Towards a Unified View of Parameter-Efficient Transfer Learning](https://mlsanity.com/paper/2110.04366/) (2021)
- [Large Language Models Can Be Strong Differentially Private Learners](https://mlsanity.com/paper/2110.05679/) (2021)
- [FederatedScope-LLM: A Comprehensive Package for Fine-tuning Large Language Models in Federated Learning](https://mlsanity.com/paper/2309.00363/) (2023)
- [FLoRA: Federated Fine-Tuning Large Language Models with Heterogeneous Low-Rank Adaptations](https://mlsanity.com/paper/2409.05976/) (2024)
- [Petals: Collaborative Inference and Fine-tuning of Large Models](https://mlsanity.com/paper/2209.01188/) (2022)
- [Federated Learning Meets Blockchain in Edge Computing: Opportunities and Challenges](https://mlsanity.com/paper/2104.01776/) (2021)
- [Understanding Clipping for Federated Learning: Convergence and Client-Level Differential Privacy](https://mlsanity.com/paper/2106.13673/) (2021)
- [When Does Differentially Private Learning Not Suffer in High Dimensions?](https://mlsanity.com/paper/2207.00160/) (2022)
- [Differentially Private Federated Learning with Local Regularization and Sparsification](https://mlsanity.com/paper/2203.03106/) (2022)
- [Make Landscape Flatter in Differentially Private Federated Learning](https://mlsanity.com/paper/2303.11242/) (2023)
- [Thinking Forward: Memory-Efficient Federated Finetuning of Language Models](https://mlsanity.com/paper/2405.15551/) (2024)
- [Universality of Layer-Level Entropy-Weighted Quantization Beyond Model Architecture and Size](https://mlsanity.com/paper/2503.04704/) (2025)

## Abstract (arXiv)

When existing retrieval-augmented generation (RAG) solutions are intended to be used for new knowledge domains, it is necessary to update their encoders, which are taken to be pretrained large language models (LLMs). However, fully finetuning these large models is compute- and memory-intensive, and even infeasible when deployed on resource-constrained edge devices. We propose a novel encoder architecture in this work that addresses this limitation by using a frozen small language model (SLM), which satisfies the memory constraints of edge devices, and inserting a small adapter network before the transformer blocks of the SLM. The trainable adapter takes the token embeddings of the new corpus and learns to produce enhanced soft embeddings for it, while requiring significantly less compute power to update than full fine-tuning. We further propose a novel retrieval mechanism by attaching a classifier head to the SLM encoder, which is trained to learn a similarity mapping of the input embeddings to their corresponding documents. Finally, to enable the online fine-tuning of both (i) the encoder soft embeddings and (ii) the classifier-as-retriever on edge devices, we adopt federated learning (FL) and differential privacy (DP) to achieve an efficient, privacy-preserving, and product-grade training solution. We conduct a theoretical analysis of our methodology, establishing convergence guarantees under mild assumptions on gradient variance when deployed for general smooth nonconvex loss functions. Through extensive numerical experiments, we demonstrate (i) the efficacy of obtaining soft embeddings to enhance the encoder, (ii) training a classifier to improve the retriever, and (iii) the role of FL in achieving speedup.