Breaking Robustness: Free-Rider Attacks Against Contribution Evaluation in Federated Learning
Bibliographic record
Abstract
Federated Learning (FL) enables privacy-preserving collaborative model training across multiple clients, with recent efforts increasingly focused on integrating contribution evaluation methods into its framework, which not only quantifies each client's contribution but also serves as a defence mechanism against malicious behaviour. While existing FL contribution evaluation techniques have proven effective in detecting attacks like label flipping or data replication, we identify a critical gap: they are vulnerable to strategically designed free-rider attacks that mimic positively contributive behaviours. In this paper, we propose GradPred, a novel free-rider attack framework that leverages time-series prediction to forecast future global gradient updates of the FL model, based on past gradient updates. This allows a malicious client to entirely bypass local training while still appearing to make meaningful contributions to the FL system, making the attack difficult to detect using state-of-the-art FL contribution evaluation methods. Through empirical evaluation, we reveal an overlooked aspect of vulnerability in FL and call for further investigation into detecting and defending against this class of stealthy, fine-grained free-rider attacks.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.019 | 0.060 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.002 | 0.003 |
| Scholarly communication | 0.003 | 0.007 |
| Open science | 0.003 | 0.007 |
| Research integrity | 0.003 | 0.005 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".