The paper introduces Contrastive Vision On‑policy Distillation (CV‑OPD), a method that enhances on‑policy distillation for vision‑language models by adding a contrastive divergence loss and a discrepancy‑aware gating mechanism. CV‑OPD contrasts teacher distributions from positive and negative visual crops, forcing the student to rely on genuine visual evidence rather than shortcuts. Experiments on fine‑grained visual benchmarks show that CV‑OPD reduces shortcut learning and improves visual reasoning across existing VLMs.
arXiv:2606. 05718v1 Announce Type: cross Abstract: On-policy distillation (OPD) improves reasoning by training a student on trajectories sampled from its own policy under supervision from a teacher.
arXiv:2608. 14144v1 Announce Type: cross Abstract: Visual on-policy distillation relies heavily on an informative teacher-student asymmetry, through either a larger, stronger teacher or privileged supervision, such as reference answers or ground-truth regions of interest.
arXiv:2609.36838v1 Announce Type: cross Abstract: Visual agents solve problems by interleaving reasoning with image operations, and on-policy distillation (OPD) provides guidance from a strong teache...
arXiv:2607. 21556v1 Announce Type: cross Abstract: On-policy self-distillation (OPSD) is promising as it removes the external teacher required by on-policy distillation (OPD), yet it still needs asymmetric information between teacher and student to ensure that the self-teacher provides a stronger learning signal than the student.
arXiv:2610.14939v1 Announce Type: new Abstract: GRPO-style reinforcement learning has advanced vision-language reasoning, but its sequence-level rewards provide limited credit assignment for individu...
OPD‑Aha is a privileged on‑policy distillation method that improves multimodal reasoning by reconstructing the distillation target from the teacher’s isolated visual preference instead of relying on fragile teacher‑student discrepancies. It suppresses continuations that contradict the image, encouraging students to interrupt flawed reasoning with reflection tokens such as "wait" and "actually." This approach leads to consistent improvements across fine‑grained perception and complex multimodal reasoning benchmarks.
arXiv:2609.39120v1 Announce Type: new Abstract: On-policy distillation (OPD) improves reasoning by providing token-level supervision from a teacher on a student's own trajectories. Existing methods p...
Privileged on-policy distillation improves multimodal reasoning by allowing a teacher to evaluate student trajectories using rich, training-only visual evidence. Both models score these trajectories w...
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.
arXiv:2608. 01263v1 Announce Type: new Abstract: On-policy distillation (OPD) samples trajectories from the current student policy and minimizes token-level divergence between student and teacher next-token distributions at prefixes along those trajectories.
arXiv:2610.23469v1 Announce Type: new Abstract: Compact medical vision-language models (VLMs) can answer clinical questions correctly without reliably using the image, relying instead on question wor...
arXiv:2609.38777v1 Announce Type: new Abstract: A central goal of vision-language model (VLM) distillation is to transfer both the teacher's language capabilities and its visual understanding. Howeve...