{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/nimbus-secure-and-efficient-two-party","title":"Nimbus: Secure and Efficient Two-Party Inference for Transformers","arxiv_id":"2411.15707","date":"2024-11-24","proceeding":null,"authors":["Zhengyi Li","Kang Yang","Jin Tan","Wen-jie Lu","Haoqi Wu","Xiao Wang","Yu Yu","Derun Zhao","Yancheng Zheng","Minyi Guo","Jingwen Leng"],"abstract":"Transformer models have gained significant attention due to their power in machine learning tasks. Their extensive deployment has raised concerns about the potential leakage of sensitive information during inference. However, when being applied to Transformers, existing approaches based on secure two-party computation (2PC) bring about efficiency limitations in two folds: (1) resource-intensive matrix multiplications in linear layers, and (2) complex non-linear activation functions like $\\mathsf{GELU}$ and $\\mathsf{Softmax}$. This work presents a new two-party inference framework $\\mathsf{Nimbus}$ for Transformer models. For the linear layer, we propose a new 2PC paradigm along with an encoding approach to securely compute matrix multiplications based on an outer-product insight, which achieves $2.9\\times \\sim 12.5\\times$ performance improvements compared to the state-of-the-art (SOTA) protocol. For the non-linear layer, through a new observation of utilizing the input distribution, we propose an approach of low-degree polynomial approximation for $\\mathsf{GELU}$ and $\\mathsf{Softmax}$, which improves the performance of the SOTA polynomial approximation by $2.9\\times \\sim 4.0\\times$, where the average accuracy loss of our approach is 0.08\\% compared to the non-2PC inference without privacy. Compared with the SOTA two-party inference, $\\mathsf{Nimbus}$ improves the end-to-end performance of \\bert{} inference by $2.7\\times \\sim 4.7\\times$ across different network settings.","url_abs":"https://arxiv.org/abs/2411.15707v1","url_pdf":"https://arxiv.org/pdf/2411.15707v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"nimbus-secure-and-efficient-two-party","repo_url":"https://github.com/secretflow/spu","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[],"methods":[{"method_slug":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"transformer","method_name":"Transformer"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2411.15707","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2411.15707"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/secretflow/spu","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"unverified":5},"by_repo_kind":{"official":{"samples":5,"ran":0,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"f927f22c2fdc59cb","entry":"drop_cached_var","repo":"secretflow/spu","repo_kind":"official","path":"spu/experimental/drop_cached_var_impl.py","file_url":"https://github.com/secretflow/spu/blob/HEAD/spu/experimental/drop_cached_var_impl.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"f927f22c2fdc59cb"}},{"code_sha256_prefix":"46f470fa1f8dac1e","entry":"example","repo":"secretflow/spu","repo_kind":"official","path":"spu/intrinsic/example_impl.py","file_url":"https://github.com/secretflow/spu/blob/HEAD/spu/intrinsic/example_impl.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"46f470fa1f8dac1e"}},{"code_sha256_prefix":"62d289c312817f45","entry":"example_binary","repo":"secretflow/spu","repo_kind":"official","path":"spu/intrinsic/example_binary_impl.py","file_url":"https://github.com/secretflow/spu/blob/HEAD/spu/intrinsic/example_binary_impl.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"62d289c312817f45"}},{"code_sha256_prefix":"c3e0fd0811d9680c","entry":"make_cached_var","repo":"secretflow/spu","repo_kind":"official","path":"spu/experimental/make_cached_var_impl.py","file_url":"https://github.com/secretflow/spu/blob/HEAD/spu/experimental/make_cached_var_impl.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"c3e0fd0811d9680c"}},{"code_sha256_prefix":"19cee4cd7c4ad5b4","entry":"reveal","repo":"secretflow/spu","repo_kind":"official","path":"spu/experimental/reveal_impl.py","file_url":"https://github.com/secretflow/spu/blob/HEAD/spu/experimental/reveal_impl.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"19cee4cd7c4ad5b4"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}