English(EN) Deterministic, Reproducible Inference in Exact Integer Arithmetic, and Verifiable Distributed Inference as Its Consequence

新的运行时确保跨硬件的可复现AI模型推理

作者 PulseAugur 编辑部 · [1 个来源] · 2026-11-11 05:00

一个名为tritify的新开源运行时已被开发出来，以确保AI模型的确定性和可复现推理。该运行时完全在精确整数算术中执行模型，将浮点运算限制在离线量化和验证过程中。它已在各种模型上进行了测试，包括仅解码器语言模型、混合专家解码器以及Segment Anything 2.1图像/视频堆栈，在不同的硬件、指令集架构甚至跨多台联网机器分布式部署的情况下，都展示了位相同的输出。 AI

影响实现了可验证的分布式推理，可能增强AI部署的信任和安全性。

排序理由详细介绍可复现AI推理新方法的论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.LG 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 34 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 详细介绍可复现AI推理新方法的论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, infra; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.LG TIER_1 English(EN) · Sayed Amir Karim (Pelloran Inc.) · 2026-11-11 05:00

精确整数算术中的确定性、可复现推理，及其可验证分布式推理的推论

arXiv:2611.13730v1 Announce Type: new Abstract: Optimized floating-point inference is not reproducible across environments: the same model on different hardware, compilers or thread counts can produce different bits. We present tritify, an open-source runtime that executes the fo…

本文如何被排名

报道来源 [1]

精确整数算术中的确定性、可复现推理，及其可验证分布式推理的推论

相关实体

相关话题