Unverified50% confidenceFactExact time
OC-OPD constructs a four-observation matrix: treating the student and teacher as candidate source and evaluator respectively, self-observations (student evaluating student, teacher evaluating teacher) measure uncertainty, and cross-observations correspond to the reverse-KL and forward-KL objectives respectively.
1
Sources
50%
Confidence
Long-term
Relevance
11/3/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedIn Squint OPSD, the teacher's token-level distribution supervises the student model's learning along the on-policy trajectory.69% similarUnverifiedBranchOPD leverages the student model's own uncertainty and the divergence between the teacher and student to locate the decision points most worth branching.68% similarUnverifiedThe reverse KL divergence has a "mode-seeking" property: when the teacher distribution is multimodal, the student is pulled toward one of the teacher's dominant modes.64% similarUnverifiedThe researchers provide a theoretical analysis under an idealized bidirectional KL objective, proving that the expected reward of the optimal policy is no lower than the teacher's, while retaining a guaranteed proportion of the teacher's probability mass.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1129549API
curl https://kongchang.com/api/v1/knowledge/claims/1129549MCP
get_claim(id=1129549)