Pith. sign in

REVIEW 6 cited by

f-Divergence Minimization for Sequence-Level Knowledge Distillation

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2307.15190 v1 pith:Z7CFYMPH submitted 2023-07-27 cs.CL cs.LG

classification cs.CLcs.LG
keywords knowledgedistillationf-distillsequence-levelapproachesdistillingexistingf-divergence
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

Knowledge distillation (KD) is the process of transferring knowledge from a large model to a small one. It has gained increasing attention in the natural language processing community, driven by the demands of compressing ever-growing language models. In this work, we propose an f-DISTILL framework, which formulates sequence-level knowledge distillation as minimizing a generalized f-divergence function. We propose four distilling variants under our framework and show that existing SeqKD and ENGINE approaches are approximations of our f-DISTILL methods. We further derive step-wise decomposition for our f-DISTILL, reducing intractable sequence-level divergence to word-level losses that can be computed in a tractable manner. Experiments across four datasets show that our methods outperform existing KD approaches, and that our symmetric distilling losses can better force the student to learn from the teacher distribution.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 6 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Cross-Tokenizer On-Policy Distillation via Byte-Prefix Marginalization

    cs.LG 2026-07 conditional novelty 7.0 of 10

    Byte-Prefix Marginalization maps a teacher's next-token distribution onto the student's vocabulary through shared byte prefixes plus an explicit residual, giving a mass-preserving target for on-policy distillation acr...

  2. Weak-to-Strong Generalization via Direct On-Policy Distillation

    cs.LG 2026-07 conditional novelty 7.0 of 10

    Transferring the log-ratio of a small model's pre-RL and post-RL checkpoints provides a dense implicit reward that improves stronger student models at a fraction of the cost of direct RL.

  3. Stable On-Policy Distillation through Adaptive Target Reformulation

    cs.LG 2026-01 unverdicted novelty 7.0 of 10

    Veto replaces the teacher's target distribution with a product of teacher and student probabilities, stabilizing on-policy knowledge distillation and improving small-model benchmarks.

  4. ABKD: Pursuing a Proper Allocation of the Probability Mass in Knowledge Distillation via $\alpha$-$\beta$-Divergence

    cs.LG 2025-05 conditional novelty 6.0 of 10

    ABKD replaces forward and reverse KL distillation with a two-parameter alpha-beta divergence that interpolates between them, improving results on 17 language and vision benchmarks.

  5. Being Strong Progressively! Enhancing Knowledge Distillation of Large Language Models through a Curriculum Learning Framework

    cs.CL 2025-06 conditional novelty 4.0 of 10

    POCL wraps LLM knowledge distillation in a curriculum that increases data difficulty and temperature over stages, improving Rouge-L on small GPT-2 and OPT students, though ablations show temperature drives the gains.

  6. Feature Alignment and Representation Transfer in Knowledge Distillation for Large Language Models

    cs.CL 2025-04 reject

    A broad survey of knowledge distillation for LLMs that summarizes published methods but contains no new results and several citation errors.

Pith tools