- Rating
- 2
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 3
- Soundness
- 2
Read review
Paper summary
The paper studies black-box tracing of leaked client copies of federated language models. It trains a shared trigger embedding and assigns it to client-specific token rows while clients fine-tune adapters. Verification compares responses to registered triggers. Experiments span 11 datasets and language and vision-language models, reporting verification rates of approximately 95.2% to 100% with generally modest task degradation. The main contribution is a reusable embedding-based marking mechanism that reduces the need for separate watermark training for each client.
Critical strengths
- Reusing a trained embedding across client-specific trigger tokens is a simple and potentially useful way to reduce per-client marking effort in adapter-based federated fine-tuning.
- The evaluation covers multiple task types, language and vision-language models, and several common model modifications. This provides meaningful evidence of basic functionality beyond a single task.
Critical weaknesses
- The evaluation does not establish whether the mark survives separating the adapters from the marked embeddings. The client receives model parameters, the identity mark is concentrated in embedding rows, and useful fine-tuning occurs in adapters. Restoring public base embeddings or attaching the adapters to a clean base is therefore a directly relevant removal scenario. The current perturbation tests do not resolve it. This is an untested concern, not a demonstrated successful attack.
- The identification evidence does not match the decision rule. The rule requires other clients’ triggers to remain below a threshold, whereas low average cross-client response does not bound the worst competing identity or false identification on innocent models. Without threshold calibration and suitable null models, high verification rates alone do not establish reliable leakage tracing.
Questions for authors
- Under the stated threat model, what happens to task utility and verification when public base embeddings are restored or client adapters are attached to an unmarked base? If this operation is excluded, please clarify the restriction and its practical implications.
- What threshold is actually used for identification, and what is the probability of accusing any wrong client or an independently trained innocent model at that threshold? Please distinguish the maximum cross-client response from the reported average.