Discuz! Board

 找回密码
 立即注册
搜索
热搜: 活动 交友 discuz
查看: 5|回复: 0

“前程质远”2026前列腺癌质控水平高质提升项目在进博会签约

[复制链接]

7万

主题

7万

帖子

21万

积分

论坛元老

Rank: 8Rank: 8

积分
219781
发表于 4 天前 | 显示全部楼层 |阅读模式
“前程质远”2026前列腺癌质控水平高质提升项目在进博会签约

机器人控制测试发现智能体会误报任务成功

RoboARCade把长序列操作任务放进同一套MuJoCo仿真环境,涉及三种机器人本体。团队扩大了原任务的初始条件范围,让智能体必须在执行过程中利用记忆、判断和视觉信息。综合测试中,GPT-6 Astra得分49.8%,Claude Fable 5.1为43.6%。更丰富的视觉界面带来1.5至6.8个百分点的改善,使用成本同时增至原来的1.7至2.3倍。 Satvik Sharma等人的这项研究收录在11月6日更新的计算机科学预印本列表中。

错误分析显示,主要困难在于选择操作策略,或把正确策略执行得足够精确。智能体在尝试过的子任务中,有24%至39%出现错误宣告成功的情况。在另一个五任务子集上,提示中加入一段成功示范后,Astra从53.5%提高到73.0%。后者属于子集测试,不能当作整套基准上的得分。团队公开了代码、视频与执行轨迹。

https://lightroom.adobe.com/shares/4dcc360d1d6cf14cbc3102b8da649cb8
https://lightroom.adobe.com/shares/d225a7dcace72043861a56f4a51f47dd
https://lightroom.adobe.com/shares/94c71396951936449e4854555a3e5b66
https://lightroom.adobe.com/shares/2799d597e8c61844b62ed81e2eb1b00b
https://lightroom.adobe.com/shares/8607d84ad42a5243bda4ae57bc83b51e
https://lightroom.adobe.com/shares/cacf131bee20b440ad1998d3edd6d418
https://lightroom.adobe.com/shares/76391ff6be0764479bdfd48c3a489a09
https://lightroom.adobe.com/shares/4eb9e1f1b036e5498958270b934ece14
https://lightroom.adobe.com/shares/4b4d9b0c2c71e44c87bd55cb967e7cc8
https://lightroom.adobe.com/shares/b1db6d904a97804eb4a6cf41854fcd97
https://lightroom.adobe.com/shares/369acdc3df54d946909743c86371e6a3
https://lightroom.adobe.com/shares/dbd79fad8430a9479ed1314e5bcae657
https://lightroom.adobe.com/shares/3f77468fef75644a88548ffd6d97d27b
https://lightroom.adobe.com/shares/96bf8f2e3b230b4f8a3516e97a483bca
https://lightroom.adobe.com/shares/75009d43b3f5934e8333d22e62cd065f
https://lightroom.adobe.com/shares/2b3c6a437ba3be45a5eaea2348957780
https://lightroom.adobe.com/shares/77ab1d847603044bb1ed92fa8d386183
https://lightroom.adobe.com/shares/b6c871b436ee0446b31d661723841f74
https://lightroom.adobe.com/shares/aba34f39ab97d247a568cb5d3d08893d
https://lightroom.adobe.com/shares/5416cf93533e7d4989b4b67bf03a8371
https://lightroom.adobe.com/shares/6c31e1f09040d84da88238a37fcfef9d
https://lightroom.adobe.com/shares/78b985d6cfb00845ac6ca16df8ae884e
https://lightroom.adobe.com/shares/837911a1f3a0a742995a492dd8efe731
https://lightroom.adobe.com/shares/02ba2d15261af04d9604e0e4ee3ebe0e
https://lightroom.adobe.com/shares/21f33b5b2ecf9048a1937eb35d02c3c6
https://lightroom.adobe.com/shares/d9db4f4919ca9940bb1e699d448faad7
https://lightroom.adobe.com/shares/f9986a93b27f6346b855e628f97377ad
https://lightroom.adobe.com/shares/f64b351729598241aace90d19ff024aa
https://lightroom.adobe.com/shares/dd1188534c725c4a94b830061a85bd20
https://lightroom.adobe.com/shares/3324ad77ec43b641a3c8a5308732876b
https://lightroom.adobe.com/shares/ad3675c24dd6c442ad20ddb716a938b0
https://lightroom.adobe.com/shares/144efd2d0346474a90439766d1695f10
https://lightroom.adobe.com/shares/dc1e1874082ac64db29c3510ce40fd00
https://lightroom.adobe.com/shares/4d31e558cf441d48b5c3cbf8a32f41dd
https://lightroom.adobe.com/shares/12330bcdd471a440bfe9f94dd77c3dca
https://lightroom.adobe.com/shares/02585e12402c354ca388b4b4bc3126b1
https://lightroom.adobe.com/shares/3b5285a1f458ed4dbcbb9eece4fb9166
https://lightroom.adobe.com/shares/c7058db66fd4d34b84fb288c7b91e300
https://lightroom.adobe.com/shares/d1b88020719bc848adffacc923a28cbf
https://lightroom.adobe.com/shares/3ca9b1f886c3604b901f317ea131d468
https://lightroom.adobe.com/shares/1b9aa024158b4d4ebcd1b9ad16255fca
https://lightroom.adobe.com/shares/27a55cb53029b64d9ddb128887a73d43
https://lightroom.adobe.com/shares/53d0b2ef3d2e04428821655addafff8a
https://lightroom.adobe.com/shares/68424b44ee79f64e9249992fa680b024
https://lightroom.adobe.com/shares/1a2ad66184942d449c5fab09334f7a7f
https://lightroom.adobe.com/shares/ecce30e71645f4498ca5414c0fa09600
https://lightroom.adobe.com/shares/4e46ee2d4dcab34dbdf595d8410b35a2
https://lightroom.adobe.com/shares/256ad12b8da83446b67b48cb8fc89b5e
https://lightroom.adobe.com/shares/68f13c7aafebb141a613566ca7087530
https://lightroom.adobe.com/shares/4e6216de0fa26b4eb947fb4a536504ef
回复

使用道具 举报

您需要登录后才可以回帖 登录 | 立即注册

本版积分规则

Archiver|手机版|小黑屋|DiscuzX

GMT+8, 2026-11-14 19:44 , Processed in 0.045551 second(s), 18 queries .

Powered by Discuz! X3.4

Copyright © 2001-2020, Tencent Cloud.

快速回复 返回顶部 返回列表