arXiv:2607. 04432v1 Announce Type: cross Abstract: A student model trained on pure uniform noise can still inherit its teacher's digit-classification ability, provided the two share initialization.
arXiv:2606. 12171v1 Announce Type: cross Abstract: Knowledge Distillation (KD) and mixup have proven effective at inducing smoothness in class boundaries; KD captures inherent class relationships in probability distributions, and mixup enforces them through convex combinations of inputs.
Direct feedback alignment (DFA) trains hidden layers via fixed random projections of output error, but with tanh hidden units and independent sigmoid outputs, plain stochastic gradient descent can stall near a constant predictor of class frequencies. This stall is traced to the error’s common mode—a rank‑one component shared across inputs—that drives tanh units toward saturation. The study shows that calibration of the baseline readout to class priors suppresses collapse and speeds learning, while other interventions such as using Adam, adjusting feedback strength, or subtracting batch means affect the severity and recovery of collapse across MNIST, CIFAR‑10, and deeper networks.
The paper investigates subliminal learning, where hidden traits from a teacher model are transferred to a student during distillation. It introduces trait‑direction drift as the underlying mechanism, showing that biased generation creates measurable preference gaps that accumulate into behavioral transfer during fine‑tuning. The authors propose probe‑space corridor regularization, a targeted defense that constrains drift along a calibrated trait direction, significantly reducing hidden‑trait transfer while maintaining task performance.
The paper investigates how teacher signals influence parameter updates in Multi‑Teacher On‑Policy Distillation (MOPD) by analyzing Qwen3‑1.7B and SmolLM3‑3B. It shows that loss averaging, Adam’s first‑moment bias, BF16 rounding, and the choice of averaging rule all shape the gradients and ultimately affect task performance. The study quantifies these effects, revealing, for example, that token‑averaging favors longer responses and that BF16 rounding masks most weight changes.
The paper investigates self‑distillation techniques for language models by systematically varying three key design choices: the source of rollout tokens (student vs. teacher), the teacher coupling strategy (frozen or exponential moving average), and the KL divergence direction (reverse or forward). Experiments on Qwen2.5‑7B and Ministral‑3‑3B across 1,200 adaptation runs reveal that rollout source mainly affects acquisition on contradictory tasks, teacher coupling most strongly influences acquisition across all tasks, and KL direction impacts retention differently depending on the model. A controlled theoretical model reproduces these empirical trends, offering a unified framework for understanding acquisition‑retention trade‑offs in self‑distillation.
arXiv:2606. 02237v1 Announce Type: new Abstract: Distribution Matching Distillation (DMD) compresses pretrained diffusion models into efficient few-step generators by aligning their noised distributions across all scales.
Reinforcement learning with verifiable rewards yields no group-relative signal when rollout groups are uniformly correct or uniformly wrong, which account for 63. 0-68.
The paper investigates how knowledge distillation from Vision Transformers to smaller CNNs can cause dimensional collapse in the student’s representation space. Using SVD and Shannon entropy, the authors show that cosine‑based distillation leads to a drastic reduction in effective rank, while adding an InfoNCE objective can double the rank but harms downstream accuracy due to signal dilution. They further demonstrate that a label‑aware contrastive objective (Supervised Contrastive distillation) can maintain or improve accuracy without unnecessary rank expansion, indicating that effective rank alone is not a reliable indicator of representation quality.
arXiv:2609.39338v1 Announce Type: new Abstract: Knowledge distillation transfers knowledge by encouraging a student to match a teacher's predicted class probabilities. These probabilities express not...
arXiv:2608. 09826v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards yields no group-relative signal when rollout groups are uniformly correct or uniformly wrong, which account for 63.
arXiv:2609.36240v1 Announce Type: new Abstract: Does representation learning stop when the training loss stops improving? We study this question for matrix Muon, whose polar-normalised updates have a...
The paper investigates how scaling the amount of model-generated, off‑task distillation data affects the recoverability of teacher‑induced traits in student models. In a controlled subliminal‑learning setup, teachers are prompted to express a target trait, producing restricted data such as number‑only completions. Students trained on larger independent datasets show a clearer manifestation of the teacher’s trait in a separate evaluation domain, with the effect being strongest when the trait is already favored or when alternative traits are present. The authors find this trend holds across model families, trait types, multi‑trait settings, and cross‑model transfer, and suggest that scaling should be coupled with trait‑aware curation and evaluation.