Pith. sign in

REVIEW 1 cited by

Sparse Progressive Distillation: Resolving Overfitting under Pretrain-and-Finetune Paradigm

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 2110.08190 v4 pith:E4GKYU5F submitted 2021-10-15 cs.CL

classification cs.CL
keywords pruningoverfittingparadigmpretrain-and-finetuneunderdistillationprogressiverisk
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Conventional wisdom in pruning Transformer-based language models is that pruning reduces the model expressiveness and thus is more likely to underfit rather than overfit. However, under the trending pretrain-and-finetune paradigm, we postulate a counter-traditional hypothesis, that is: pruning increases the risk of overfitting when performed at the fine-tuning phase. In this paper, we aim to address the overfitting problem and improve pruning performance via progressive knowledge distillation with error-bound properties. We show for the first time that reducing the risk of overfitting can help the effectiveness of pruning under the pretrain-and-finetune paradigm. Ablation studies and experiments on the GLUE benchmark show that our method outperforms the leading competitors across different tasks.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

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

  1. ElastiFormer: Learned Redundancy Reduction in Transformer via Self-Distillation

    cs.LG 2024-11 conditional novelty 5.0 of 10

    A post-training routing method that uses self-distillation to let frozen pretrained Transformers process only a subset of parameters and tokens, cutting active compute by 20 to 50 percent.

Pith tools