亲爱的研友该休息了!由于当前在线用户较少,发布求助请尽量完整地填写文献信息,科研通机器人24小时在线,伴您度过漫漫科研夜!身体可是革命的本钱,早点休息,好梦!

The Last Human-Written Paper: Agent-Native Research Artifacts

作者
Jiachen Liu,Jiaxin Pei,Jintao Huang,Chenglei Si,Ao Qu,Xiangru Tang,Runyu Lu,Lichang Chen,Xiaoyan Bai,Haizhong Zheng,Carl Chen,Zhiyang Chen,Haojie Ye,Yujuan Fu,Zexue He,Zijian Jin,ZHANG Zhenyu,Shangquan Sun,Maestro Harmon,John Dianzhuo Wang
出处
期刊:Cornell University - arXiv [Cornell University]
被引量:30
标识
DOI:10.48550/arxiv.2604.24658
摘要

Scientific publication compresses a branching, iterative research process into a linear narrative, discarding the majority of what was discovered along the way. This compilation imposes two structural costs: a Storytelling Tax, where failed experiments, rejected hypotheses, and the branching exploration process are discarded to fit a linear narrative; and an Engineering Tax, where the gap between reviewer-sufficient prose and agent-sufficient specification leaves critical implementation details unwritten. Tolerable for human readers, these costs become critical when AI agents must understand, reproduce, and extend published work. We introduce the Agent-Native Research Artifact (ARA), a protocol that replaces the narrative paper with a machine-executable research package structured around four layers: scientific logic, executable code with full specifications, an exploration graph that preserves the failures compilation discards, and evidence grounding every claim in raw outputs. Three mechanisms support the ecosystem: a Live Research Manager that captures decisions and dead ends during ordinary development; an ARA Compiler that translates legacy PDFs and repos into ARAs; and an ARA-native review system that automates objective checks so human reviewers can focus on significance, novelty, and taste. On PaperBench and RE-Bench, ARA raises question-answering accuracy from 72.4% to 93.7% and reproduction success from 57.4% to 64.4%. On RE-Bench's five open-ended extension tasks, preserved failure traces in ARA accelerate progress, but can also constrain a capable agent from stepping outside the prior-run box depending on the agent's capabilities. Our code is open-sourced at https://github.com/Orchestra-Research/Agent-Native-Research-Artifact.

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
molihuakai的应助被科研通管家采纳,获得10
12秒前
13秒前
英俊的谷蓝完成签到,获得积分10
17秒前
明理的璎完成签到,获得积分10
35秒前
39秒前
美丽的忆彤完成签到,获得积分10
40秒前
Ava的应助被sun采纳,获得10
56秒前
1分钟前
动听一德完成签到,获得积分10
1分钟前
领导范儿的应助被ClaraClark591采纳,获得30
1分钟前
1分钟前
1分钟前
sun发布了新的文献求助10
1分钟前
ClaraClark591发布了新的文献求助30
1分钟前
威武的又琴完成签到,获得积分10
1分钟前
闪闪小凡完成签到,获得积分10
1分钟前
1分钟前
369ninja发布了新的文献求助10
1分钟前
Tashanzhishi完成签到,获得积分10
2分钟前
优美的烙完成签到,获得积分10
2分钟前
打打的应助被科研通管家采纳,获得10
2分钟前
2分钟前
2分钟前
科目三的应助被科研通管家采纳,获得10
2分钟前
奔跑的应助被科研通管家采纳,获得10
2分钟前
愉快的惋庭完成签到,获得积分10
2分钟前
等待剑完成签到,获得积分10
2分钟前
2分钟前
Maestro_S的应助被爱学习的doge采纳,获得10
2分钟前
2分钟前
3分钟前
孙老师完成签到 ,获得积分10
3分钟前
AiQi完成签到 ,获得积分10
3分钟前
sfwrbh完成签到,获得积分10
3分钟前
多情的中蓝完成签到,获得积分10
3分钟前
甜蜜语堂完成签到,获得积分10
3分钟前
李爱国的应助被科研通管家采纳,获得10
4分钟前
Maestro_S的应助被科研通管家采纳,获得10
4分钟前
大医仁心完成签到 ,获得积分0
4分钟前
Ali发布了新的文献求助10
4分钟前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 1700
The World Atlas of Wine 9th edition 1000
奈特基础组织学:配套组织病理学(第4版)(Netter's Essential Histology With Correlated Histopathology) 800
John McDowells Bildungstheorie 800
Social Psychology in the Real World 600
八旗制度の研究 600
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8179956
求助须知:如何正确求助?哪些是违规求助? 9570170
关于积分的说明 21808444
捐赠科研通 7775818
什么是DOI,文献DOI怎么找? 3433251
关于科研通互助平台的介绍 2557352
邀请新用户注册赠送积分活动 2478958