# FedSDR: Federated Self-Distillation with Rectification

arXiv: https://arxiv.org/abs/2605.18028 · HTML: https://mlsanity.com/paper/2605.18028/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Federated, robust and private distillation, one of 12 directions in Knowledge distillation.
- Builds on 20 of its 57 references that are in the map.
- First posted on arXiv 18 May 2026.

## Placed in

- Knowledge distillation / [Federated, robust and private distillation](https://mlsanity.com/distill/federated-robust-and-private-distillation/)

## Builds on (most cited first, 20 of 20)

- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Qwen2.5-Coder Technical Report](https://mlsanity.com/paper/2409.12186/) (2024)
- [WizardLM: Empowering large pre-trained language models to follow complex instructions](https://mlsanity.com/paper/2304.12244/) (2023)
- [Federated Learning with Non-IID Data](https://mlsanity.com/paper/1806.00582/) (2018)
- [Adaptive Federated Optimization](https://mlsanity.com/paper/2003.00295/) (2020)
- [DROP: A Reading Comprehension Benchmark Requiring Discrete Reasoning Over Paragraphs](https://mlsanity.com/paper/1903.00161/) (2019)
- [Measuring the Effects of Non-Identical Data Distribution for Federated Visual Classification](https://mlsanity.com/paper/1909.06335/) (2019)
- [Federated Learning with Personalization Layers](https://mlsanity.com/paper/1912.00818/) (2019)
- [MedAlpaca -- An Open-Source Collection of Medical Conversational AI Models and Training Data](https://mlsanity.com/paper/2304.08247/) (2023)
- [Self-Distillation Bridges Distribution Gap in Language Model Fine-Tuning](https://mlsanity.com/paper/2402.13669/) (2024)
- [FinGPT: Democratizing Internet-scale Data for Financial Large Language Models](https://mlsanity.com/paper/2307.10485/) (2023)
- [pFedLoRA: Model-Heterogeneous Personalized Federated Learning with LoRA Tuning](https://mlsanity.com/paper/2310.13283/) (2023)
- [Advances and Open Challenges in Federated Foundation Models](https://mlsanity.com/paper/2404.15381/) (2024)
- [Federated Large Language Models: Current Progress and Future Directions](https://mlsanity.com/paper/2409.15723/) (2024)
- [Personalized Federated Fine-Tuning for LLMs via Data-Driven Heterogeneous Model Architectures](https://mlsanity.com/paper/2411.19128/) (2024)
- [Merge-of-Thought Distillation](https://mlsanity.com/paper/2509.08814/) (2025)
- [FedDistill: Global Model Distillation for Local Model De-Biasing in Non-IID Federated Learning](https://mlsanity.com/paper/2404.09210/) (2024)
- [Training-Trajectory-Aware Token Selection](https://mlsanity.com/paper/2601.10348/) (2026)
- [Co-LoRA: Collaborative Model Personalization on Heterogeneous Multi-Modal Clients](https://mlsanity.com/paper/2506.11024/) (2025)

## Abstract (arXiv)

Federated fine-tuning of Large Language Models faces severe statistical heterogeneity. However, existing model-level defenses often overlook the root cause: intrinsic data distribution mismatches. In this work, we first establish Federated Self-Distillation (FedSD) as a fundamental and potent strategy. By projecting client representations into a smoothed ``model-understanding space,'' FedSD alone serves as a universal booster, demonstrating superior performance over conventional algorithms. Despite its success, we identify a subtle trade-off termed the Rewrite Paradox -- unconstrained self-distillation can inadvertently increase hallucinations and redundancy. To refine this paradigm, we further propose FedSDR (Federated Self-Distillation with Rectification), the ultimate reinforced framework. It augments FedSD with a dual-stream mechanism: a local LoRA-S (Smoothing) branch to implicitly absorb heterogeneity via distilled data, and a parallel global LoRA-R (Rectification) branch anchored to raw data to enforce factual correctness. By selectively aggregating only LoRA-R, FedSDR yields a globally aligned and faithful model. Extensive experiments verify its superior performance.