AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

MiSS: Revisiting the Trade-off in LoRA with an Efficient Shard-Sharing Structure

Jiale Kang, Qingyu Yin

ICLR 2026 · Architectures and efficiency › Pruning and efficient tuning · approach: LoRA variants and theory · 13 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
raise MiSS rank from 16 to 128 (ablation)MiSS on Llama2-7B · GSM8Kaccuracy: 45.90 -> 53.49Improved
raise MiSS rank from 16 to 128 (ablation)MiSS on Llama2-7B · Mathaccuracy: 3.77 -> 10.08Improved
replace LoRA with MiSSRoBERTa-base · GLUE CoLAscore: 62.40 -> 72.86Improved
replace PiSSA with MiSSRoBERTa-base · GLUE CoLAscore: 67.28 -> 72.86Improved
replace PiSSA with MiSSRoBERTa-base · GLUE MNLIaccuracy: 85.72 -> 85.71No clear effect
replace LoRA with MiSSLlama2-7B · NLG averageaverage score: 24.72 -> 29.30Improved
replace LoRA with MiSSQwen3-4B · Mathaccuracy: 15.20 -> 34.82Improved
replace PiSSA with MiSSMistral-7B · NLG averageaverage score: 44.45 -> 47.79Improved
replace PiSSA with MiSSLlama2-13B · NLG averageaverage score: 39.52 -> 42.11Improved
replace PiSSA with MiSSQwen3-4B · GSM8Kaccuracy: 85.78 -> 85.52Hurt
replace LoRA with MiSSViT-Base on VTAB-1K · image averageaccuracy: 88.08 -> 88.02No clear effect
replace LoRA with MiSSLlama-3.2-3B · PEFT benchmarktest accuracy: 0.4822 -> 0.5080Improved

Later papers that compared with this method

1 later accepted paper reports MiSS as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy