# jaguar Published articles for jaguar. This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles. ## Jaguar: Fast Private Neural Network Inference with Power-of-Two Homomorphic Arithmetic DevFeed: [Jaguar: Fast Private Neural Network Inference with Power-of-Two Homomorphic Arithmetic]() Original publisher: [Read original article]() Author: Yewon Jeong, Nayoung Jung, Hyeri Roh, Woo-Seok Choi Published: 2026-11-04T05:00:00Z Content type: article Language: en Sources: [arXiv: Cryptography and Security]() Topics: [homomorphic encryption](), [FHE](), [Cryptography]() Tags: [2pc](), [arithmetic](), [avx](), [client](), [communication](), [convolution](), [cs-cr](), [design](), [domain](), [fast](), [fixed](), [flow](), [hybrid](), [inference](), [jaguar](), [kernel](), [network](), [neural](), [neural-network](), [power](), [private]() ### AI overview Jaguar targets two bottlenecks in hybrid homomorphic encryption/two-party computation inference: prime-modulus convolution and doubled-precision ReLU followed by separate truncation. Its power-of-two ciphertext ring enables SPA-Conv, which uses Scalar-Polynomial Accumulation in the coefficient domain, and exact ciphertext-side truncation by local right shifts, allowing ReLU at target fixed-point precision without a post-ReLU truncation protocol. For client-side decryption, Jaguar uses an auxiliary NTT prime for the single polynomial multiplication, retaining O(N log N) decryption. On ImageNet-scale ResNet-18, ResNet-50, and MobileNetV2 with AVX disabled, the abstract reports 2.07-3.72x lower end-to-end latency than Cheetah and 2.16-3.36x lower than Rhombus, plus 1.16-1.76x lower communication than Cheetah. ### Source excerpt arXiv:2606.11827v2 Announce Type: replace Abstract: Hybrid HE/2PC private neural network inference remains bottlenecked by prime-modulus homomorphic arithmetic in convolution and by a precision flow that runs ReLU at doubled bitwidth before invoking a separate truncation protocol. We present Jaguar, a system built on a single design choice--a power-of-two ciphertext ring--that addresses both. The choice enables SPA-Conv, a coefficient-domain convolution kernel that replaces NTT-centric polynomial multiplication with Scalar-Polynomial Accumulation, and an exact ciphertext-side truncation by local right shifts that lets ReLU run directly at the target fixed-point precision and eliminates the post-ReLU truncation protocol. Where NTT remains genuinely useful--at the client, for the single polynomial multiplication during decryption--we recover it through an auxiliary NTT prime, preserving the power-of-two protocol substrate while keeping decryption O(N log N). On ImageNet-scale ResNet-18, ResNet-50, and MobileNetV2 with AVX disabled, Jaguar achieves 2.07-3.72x lower end-to-end latency than Cheetah and 2.16-3.36x lower than Rhombus, with 1.16-1.76x lower communication than Cheetah. ## Tech Firms Dismantle Botnet That Turned 1.3M TV Boxes Into Proxies DevFeed: [Tech Firms Dismantle Botnet That Turned 1.3M TV Boxes Into Proxies]() Original publisher: [Read original article]() Author: BrianKrebs Published: 2026-11-03T02:24:31Z Content type: news Language: en Sources: [Krebs on Security]() Topics: [IoT botnets](), [Internet of things](), [pay-per-install](), [malware protection](), [account takeover]() Tags: [a-little-sunshine](), [advertising](), [alarum-technologies](), [android-tv](), [appsalt](), [ben-brundage](), [bmw-ipweb](), [botnet](), [earnfm](), [gaganode](), [google](), [google-threat-intelligence-group](), [honeygain](), [internet-of-things-iot](), [jaguar](), [kimwolf](), [linkfog](), [mecool](), [motion-picture-association](), [ne-er-do-well-news](), [netnut](), [skyworth-technology](), [source](), [superbox](), [synthient](), [takedown](), [traffmonetizer](), [web-fraud-2-0](), [xcpe-botnet]() ### AI overview Google and industry partners dismantled the XCPE botnet, which enrolled more than 1.3 million Android TV boxes and may have connected to over 10 million additional compromised devices. Google says Videostrong Technology Co. used malware to sell infected devices' internet connections through residential proxy services, enable advertising fraud, and distribute affiliates' malware. The same malware was found in other Android IoT products, including telehealth devices, karaoke machines, and digital signage. Google took legal action against the domains and servers used to control XCPE and says it fixed a weakness that let malware load on first boot despite device security requirements. The report names several proxy providers linked to infected devices and cautions that affiliates also distributed their modules through channels beyond XCPE, so the 1.3 million figure may understate the wider impact. ### Source excerpt Google said today it worked with industry partners to dismantle a botnet made up of more than 1.3 million Android TV set-top boxes that ship with malicious software which secretly rents the user's Internet connection out to strangers and conducts advertising fraud. Experts involved in the takedown say the action is likely to sever the botnet's access to more than 10 million additional compromised systems, including other Internet of Things (IoT) products such as Android-based healthcare devices and karaoke machines.