The paper introduces Watermark Basin Alignment (WM‑Basin), a black‑box method for detecting distilled large language models (LLMs). It exploits a persistent watermark basin—a key‑specific geometric signal embedded in the teacher’s parameter space during fine‑tuning—that remains detectable even when token‑level signals vanish. WM‑Basin maps suspect model outputs to gradient representations, removes ordinary fine‑tuning directions, and tests alignment with the watermark basin, achieving significantly higher detection rates than token‑level baselines across a range of teacher and student models.
arXiv:2607. 09692v1 Announce Type: new Abstract: Model distillation -- training on outputs from stronger third-party models -- is widely used to boost performance, but raises concerns about unfair advantages and policy violations.
arXiv:2606. 18430v1 Announce Type: new Abstract: Statistical watermarks help organizations attribute large language model (LLM) outputs, yet existing detectors often struggle when watermark signals are weak, texts are repetitive, or watermarks are edited.
arXiv:2610.15685v1 Announce Type: new Abstract: Conventional distillation transfers capability through teacher-generated demonstrations. LLM-as-a-Judge knowledge distillation offers a distinct and in...
arXiv:2605.10889v2 Announce Type: replace-cross Abstract: On-policy distillation offers dense, per-token supervision for training reasoning models; however, it remains unclear under which conditions...
arXiv:2608.29030v1 Announce Type: new Abstract: In-context watermarking (ICW) prepends an instruction to a query asking the model to embed a statistically detectable signal in its response. It thus e...
arXiv:2607. 06855v1 Announce Type: new Abstract: On-policy distillation is a practical post-training recipe for large language models, supplying dense teacher supervision on the student's own trajectories.
arXiv:2604.23238v3 Announce Type: replace-cross Abstract: Distillation via reasoning traces exposes frontier models to adversarial third parties, who can bypass their guardrails and misappropriate th...
On-Policy distillation (OPD) transfers teacher capabilities by supervising student-sampled trajectories with dense token-level teacher signals. Recent selective OPD methods improve this process by prioritizing signals that are confident, informative, or learnable.
arXiv:2608. 03632v1 Announce Type: new Abstract: On-Policy distillation (OPD) transfers teacher capabilities by supervising student-sampled trajectories with dense token-level teacher signals.
MarkTune is a model‑embedded watermarking framework that improves the trade‑off between generation quality and detectability. By regularizing toward a high‑quality reference distribution, it decouples quality preservation from proximity to the base model and uses the GaussMark detection statistic as a reward during on‑policy fine‑tuning. Empirical results show that MarkTune enhances the quality‑detectability frontier, maintains downstream task performance, and remains robust against text‑editing and open‑weight attacks across diverse datasets.
arXiv:2607. 20440v1 Announce Type: cross Abstract: Proprietary large language models (LLMs) entail substantial intellectual and financial investment, making them valuable intellectual property (IP).
arXiv:2605.21856v2 Announce Type: replace-cross Abstract: Large language models (LLMs) have demonstrated impressive reasoning abilities across a wide range of tasks, but data contamination undermines...