Unverified50% confidenceOpinionExact time
LEAF-OPD relies on a high-quality teacher model, and the teacher's capability ceiling may constrain the student's unlearning effectiveness.
1
Sources
50%
Confidence
Long-term
Relevance
11/17/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe paper observes that supervision signals from a fixed teacher become increasingly less worthwhile as the student model's capabilities improve76% similarUnverifiedD²-OPD determines the extrapolation direction by adaptively weighted combination of the teacher model's post-training offset and the initial student-teacher gap.76% similarUnverifiedThe researchers found that even starting from the states where the student deviates the most, the teacher's own takeover can still complete the task, whereas the student, after training on the teacher's per-token scores, often runs out of turns.74% similarUnverifiedExisting OPD-based memory training schemes face a cold-start dilemma: memory tools and protocols are not within the pretraining distribution of any model, so the teacher model cannot provide effective guidance, and the student model cannot sample enough memory actions.74% similarUnverifiedThe core challenge of On-Policy Distillation (OPD) is how to extract effective supervisory signals from the teacher model; traditional methods rely on token probabilities or directly aligning hidden states, each with its own limitations: token probabilities may underestimate the teacher's internal knowledge, while directly aligning hidden states forces accommodation to the teacher's feature space, which is especially tricky in cross-architecture transfer.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1179617API
curl https://kongchang.com/api/v1/knowledge/claims/1179617MCP
get_claim(id=1179617)