Metraipolis › archive › dc_39238e4886bf

CROP: Task Relevance via Counterfactuals for Selective On-Policy Distillation — arXiv:2608.13387

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.13387
License (as archived)
CC0-1.0
Archived
2026-10-19T04:24:38.043Z
SHA-256 of the archived text
97a294f7dbc24d275c87b7c10a4acb643d1cb1d4603880053c0a8496bf5b3e3c
Stable URL of this page
https://metraipolis.com/documento/dc_39238e4886bf

How to cite

CROP: Task Relevance via Counterfactuals for Selective On-Policy Distillation — arXiv:2608.13387 — arXiv (en), https://arxiv.org/abs/2608.13387 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_39238e4886bf (Metraipolis archive dc_39238e4886bf, sha256 97a294f7dbc24d275c87b7c10a4acb643d1cb1d4603880053c0a8496bf5b3e3c)
BibTeX
@misc{dc39238e4886bf,
  title = {CROP: Task Relevance via Counterfactuals for Selective On-Policy Distillation — arXiv:2608.13387},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.13387},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_39238e4886bf; sha256 97a294f7dbc24d275c87b7c10a4acb643d1cb1d4603880053c0a8496bf5b3e3c},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.13387. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

On-policy distillation (OPD) supervises a student language model on trajectories sampled from its current policy, but assigns equal credit to response tokens with unequal supervision value. Selective OPD addresses this limitation by allocating supervision non-uniformly across response tokens according to their estimated training value. Most existing criteria, however, focus primarily on optimization need, such as uncertainty or teacher-student disagreement, while task relevance, namely whether the supervision is tied to the semantic content of the current input, remains less directly characterized as a complementary dimension. To address this gap, we introduce Counterfactual Relevance for On-Policy Distillation (CROP), which operationalizes task relevance through a paraphrase-calibrated counterfactual sensitivity margin. For each source prompt, CROP constructs a validated original-paraphrase-counterfactual triplet, holds the student rollout fixed, and measures each response position by its sensitivity to a task-relevant condition change calibrated by its sensitivity to a meaning-preserving rewrite. Matched selection controls show that CROP identifies more useful supervision positions than random or lowest-relevance selection, while component comparisons confirm the value of both counterfactual sensitivity and paraphrase calibration. Across two teacher-student settings, CROP improves aggregate performance by 1.92 and 2.96 points over the strongest non-CROP selector. These results support task relevance as a complementary criterion for selective OPD and establish CROP as a model-internal, contrast-specific method for allocating token-level supervision.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-14 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2608.13387 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-19T04:17:42.399Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-08-14?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).