AI Paper Atlas

Research map / Large language models / Instruction and fine-tuning

SecP-Tuning: Efficient Privacy-Preserving Prompt Tuning for Large Language Models via MPC

Jinglong Luo, Zhuo Zhang, Yehong Zhang, Shiyu Liu, Ye Dong, Hui Wang, Yue Yu, Xun Zhou, Zenglin Xu

ICLR 2026 · Large language models › Instruction and fine-tuning · approach: Safety alignment after fine-tuning

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
replace gradient-based tuning with forward-only tuning (no secure backward pass or optimizer) (ablation)MPC prompt tuning · RoBERTa-large LAN, seq len 512total time (s): 882.075 -> 174.138 (Prompt Tuning vs FoT)Improved
add privacy-preserving Random Feature Attention in place of softmax attention (ablation)SecP-Tuning (FoT) · RoBERTa-large LAN, seq len 512total time (s) and communication (GB): 174.138 -> 55.172 s; 205.359 -> 56.545 GBImproved
use SecP-Tuning instead of gradient-based prompt tuning in MPCPrompt Tuning · RoBERTa-large LANspeedup and communication: 16x speedup, 20x less communicationImproved
use gradient-based prompt tuning instead of full SFT under MPCSFT · RoBERTa-large LANtotal time (s): 651.598 -> 882.075Hurt
use the proposed cosine protocol inside RFA (ablation)privacy-preserving RFA · sequence lengths 64 to 512time and communicationImproved
lower bandwidth and higher latency (WAN 100Mbps/80ms vs LAN)SecP-Tuning · RoBERTa-large WANspeedup over Prompt Tuning: 16x (LAN) -> 34x (WAN)Improved
initialize FoT with pre-trained prompt embeddings (ablation)FoT · RTE and MRPC, 16 shotsaccuracy / F1: RTE 52.59 -> 77.62; MRPC 61.56 -> 75.51Improved
use SecP-Tuning (FoT+RFA) instead of plaintext FoT without pre-trained promptFoT · Yelp Polarity, 16 shotsaccuracy: 91.50 -> 85.23Hurt
use SecP-Tuning instead of gradient-based prompt tuning without pre-trained promptPrompt Tuning · SST-2 and Yelp P., 16 shotsaccuracy: SST-2 68.23 -> 88.11; Yelp 61.02 -> 85.23Improved
use SecP-Tuning instead of SFT on harder tasksSFT · AG's News, MRPC, RTE, Yelp P. 16 shotsaccuracy / F1: AG's News 86.36 -> 81.27; Yelp 91.82 -> 85.23; MRPC 77.35 -> 75.33Mixed

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