Pith. sign in

REVIEW 1 cited by

Random Masking Finds Winning Tickets for Parameter Efficient Fine-tuning

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 2405.02596 v1 pith:7EOKS72J submitted 2024-05-04 cs.LG cs.AIcs.CL

Random Masking Finds Winning Tickets for Parameter Efficient Fine-tuning

classification cs.LG cs.AIcs.CL
keywords maskingrandomfine-tuningparameterspeftlargelearningmodels
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Fine-tuning large language models (LLM) can be costly. Parameter-efficient fine-tuning (PEFT) addresses the problems by training a fraction of the parameters, whose success reveals the expressiveness and flexibility of pretrained models. This paper studies the limit of PEFT, by further simplifying its design and reducing the number of trainable parameters beyond standard setups. To this end, we use Random Masking to fine-tune the pretrained model. Despite its simplicity, we show that Random Masking is surprisingly effective: with a larger-than-expected learning rate, Random Masking can match the performance of standard PEFT algorithms such as LoRA on various tasks, using fewer trainable parameters. We provide both empirical and theoretical explorations into the success of Random Masking. We show that masking induces a flatter loss landscape and more distant solutions, which allows for and necessitates large learning rates.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 1 Pith paper

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

  1. Enhancing Model Privacy in Federated Learning with Random Masking and Quantization

    cs.LG 2025-08 reject novelty 5.0

    FedQSN hides part of the server model with random masks and quantizes the remainder to give clients a degraded proxy, reporting a large global-vs-proxy performance gap with modest loss in the final global model.