1
Why LLM Decisions Should Be Deterministic
LLM决策的确定性边界关键不是一致性，而是可审计性——每个决策都有可复现的实现。
TL;DR: I originally treated deterministic boundaries around LLMs as a consistency mechanism. I now think their real value is auditability. If the syst…
2
Gypscie: A Cross-Platform AI Artifact Management System
跨平台AI工件管理系统Gypscie，统一管理实验产物，提升AI开发效率与可复现性。
arXiv:2604.10311v2 Announce Type: replace Abstract: Artificial Intelligence (AI) models, encompassing both traditional machine learning (ML) and more …
3
Show HN: Open Science, open-source alternative to Claude Science
对抗AI研究封闭化，开源、本地优先、模型无关的科研工作台，MIT许可，完全可复现。
Claude Science just launched. It’s a sign AI research is moving toward centralized, closed systems. So we built the opposite: Open Science: a local fi…
4
RouteJudge: An Open Platform for Reproducible and Preference-Aware LLM Routing
RouteJudge平台让LLM路由可复现且感知用户偏好，有助于精准高效地选择模型，已被ICML 2026研讨接收。
arXiv:2606.18774v1 Announce Type: new Abstract: We present RouteJudge, an online pairwise preference evaluation framework for LLM routing systems, wit…
5
SafeRun: Enabling Determinism in LLM Planning for Running
ICML 2026 workshop论文，聚焦如何让LLM在跑步规划中摆脱随机性、实现可复现的确定性输出，提升安全性与可靠性。
arXiv:2606.09027v1 Announce Type: new Abstract: Large Language Models enable flexible natural-language planning but remain unreliable in determinism-c…
6
VisualLeakBench: Reproducible Action-Boundary Propagation Failures in Vision-Language Agents
VisualLeakBench 为视觉语言代理提供可复现的动作边界传播失败评估，帮助研究人员系统检测和修复模型漏洞。
arXiv:2606.07595v1 Announce Type: new Abstract: Vision-language agents increasingly consume screenshots, documents, and user interfaces before writing…
7
BrainSurgery: Reproducible and Reliable Declarative Weight Manipulations for Model Editing and Upcycling
可复现且可靠的声明式模型权重操作方法，为AI模型编辑与升级提供新范式
arXiv:2606.09707v1 Announce Type: new Abstract: As deep learning models scale, managing, inspecting, and modifying large checkpoints has become increa…
8
Reproducibility is the New Copyleft: Defining AGI-oriented Reproducible Builds
从Copyleft到AGI可复现构建，重新定义AI开源许可的演进方向。
arXiv:2606.03019v1 Announce Type: cross Abstract: Copyleft, as implemented in licenses such as the GNU General Public License, was a legal hack that u…
9
A Unified and Reproducible Experimentation Framework for Speech Understanding
面向语音理解的开源可复现实验框架，统一评估标准，加速研究迭代。
arXiv:2605.30899v1 Announce Type: cross Abstract: Speech foundation models and Speech LLMs have advanced speech understanding, yet deployment-oriented…
10
From Text to Voice: A Reproducible and Verifiable Framework for Evaluating Tool Calling LLM Agents
一项可复现、可验证的评估框架，让工具调用型LLM智能体的性能对比不再模糊。
arXiv:2605.15104v2 Announce Type: replace Abstract: Voice agents increasingly require reliable tool use from speech, whereas prominent tool-calling be…
11
Hawkeye: Reproducing GPU-Level Non-Determinism
揭秘GPU计算的黑箱，在CPU上精确复现NVIDIA矩阵乘法，解决机器学习非确定性难题。
arXiv:2603.20421v2 Announce Type: replace-cross Abstract: We present Hawkeye, a system for analyzing and reproducing GPU-level arithmetic operations. …