Latest

Every published item, in order · 407 items
Nov 13
Chinese contentSingularity-Bench 发布:测试智能体能否端到端构建 AI 系统

这个基准让智能体从零写训练、推理、编排系统,再和人类基线硬碰硬。结果挺扎眼:Opus 5.5 配 Claude Code 六项全过人类线,其他智能体平均只有 0.58。做 agent 评测或基础设施的工程师都该看看。

arXiv cs.AI@Yang Zhou et al. (19)Source
ModelsBuildersMulti-source13:19
Chinese contentSaidzoda-Kazakh-EmbeddingGemma-300M:单轮微调将哈萨克语检索 recall@1 提升至 78.0%

339M 参数的小模型,用一个 epoch 的合成问答对微调,哈萨克语检索 recall@1 从 60.8% 拉到 78.0%,还压过 OpenAI text-embedding-3-large。做低资源语言检索或想复刻这套合成数据微调流程的工程师都该看看细节。

Event
arXiv: OpenAI@Tohir Saidzoda et al. (6)Source
ModelsBuildersMulti-sourceSelected74°11:46
Neuro-VL vision-language model trained on one million neuroimaging studies

Training on real clinical archives lets a specialized model beat general frontier systems on both capability and resistance to leading clinical prompts — a notable result for medical AI evaluation methodology. Radiology teams, hospital AI buyers and health AI developers should examine its benchmark design, especially the longitudinal comparison task, when assessing clinical imaging products.

Event
arXiv: OpenAI@Chenhui Zhao et al. (22)Source
Nov 11
Chinese contentTimeSeek 提出多任务多域预训练,统一时间序列预测、异常检测与分类

这个 TimeSeek 把预测、异常检测、分类三个任务塞进了一次预训练里,架构上按领域自适应提取特征,不再只靠堆 Transformer 层。做时间序列的工程师可以对比一下 TimesFM、Chronos 这类单任务模型,看 few-shot 和 zero-shot 成绩是否真能省掉多套模型的维护成本。

arXiv cs.LG@Peng Chen et al. (9)Source
ModelsBuildersMulti-source10:51
Chinese contentExpertCrowd-SWE发布:100个专家工程任务测试编程智能体

想让编程智能体在陌生大仓库里干真活的人该看看这个基准。它不是改几行的小修小补,而是工程师写的真实变更请求,评分器能识别偷懒和作弊补丁。榜单上 Claude Opus 5.5 和 GPT-6 Astra 基本打平,选型做对比时很实用。

Event
arXiv: DeepSeek@Josip Vrdoljak et al. (7)Source
Nov 10
Nov 9
Tomorrow 08:00, the daily lands in your inboxTelegram channel or email. Same screening as this page, no ads.
Nov 6
ModelsBuildersSelected11:29
PlayGen2 Real-Time World Model Uses Explicit 3D State

The architectural split between 3D state prediction and rendering avoids the drift that plagues video-frame-recurrent world models, and the laptop-scale latency shows the approach runs on consumer hardware rather than data-centre GPUs. Developers building playable-world or interactive-simulation demos can download the code and weights and extend the framework to other games or rule-based gameplay. It compares with video-only world models like Genie-style systems by trading appearance realism for spatial persistence and control.

arXiv cs.LG@Jihai Zhang et al. (6)Source
Nov 3
Nov 2
Oct 28
Oct 27
ModelsBuildersSelected09:20
ProtAgent-8B Matches Frontier Models on Protein Tool Tasks

Interaction-level credit assignment, not raw model scale, is what closes the gap with frontier systems here — a 671B model loses by roughly 20 points under the same protocol. Researchers building tool-using agents in specialized domains can apply the same step-level reward scheme without frontier API budgets. The authors' caveat that annotation coverage does not prove biological correctness is worth keeping in mind when comparing reported scores.

arXiv: DeepSeek@Yuxiang Wang et al. (6)Source
Oct 26
Oct 23
Oct 22
Oct 20
Oct 19
Oct 15
Oct 14
Oct 12
ModelsBuildersMulti-sourceSelected10:30
Adaptive DHiSparse Speculative Decoding

This approach solves a critical bottleneck in long-context AI serving by enabling more efficient memory usage while maintaining high throughput. The technology allows models to handle significantly larger contexts without hitting memory limits, which directly impacts real-world applications requiring extensive context windows. Developers working with large language models should consider this optimization technique to improve performance in production environments.

Event
arXiv: DeepSeek@Naji Alazhar, Wassel AlazharSource
Oct 9
ModelsBuildersSelected10:21
StoreBench Live-Commerce Agent Evaluation

StoreBench introduces a dynamic, production-grade environment that reveals significant gaps between current LLM capabilities and human performance in e-commerce operations. The benchmark's realistic challenges - including supplier failures and market shocks - provide a more rigorous test than static evaluations. Researchers and developers can access five training tasks and sample trajectories to improve agent performance in complex, time-sensitive commercial scenarios.

Event
arXiv: DeepSeek@Daksh Raghuvanshi et al. (3)Source
Oct 7
Oct 6
Oct 5
Oct 2
Oct 1
Sep 17
AITOP16:09
Anthropic拟10月在新加坡设亚太第五个办事处 开...

Anthropic拟10月在新加坡设亚太第五个办事处 开放本地招聘 数据显示,按人口比例计算,新加坡的Claude.ai使用水平在121个国家中排名第二。 我们观察到,人工智能公司Anthropic计划于10月在新加坡设立其亚太地区的第五…

AITOPRead
Aug 13
Aug 7
AITOP17:04
出口结构剧变:中国卖给世界的,不再是产品而是解法

中国出口"新新三样":从打火机到爬墙机器人,世界不再只是买便宜货 我们观察到一组清晰的数据信号:中国出口的"新新三样"已经从过去的服装、家具、家电,换成机器人、AI模型和创新药物。这不是产业升级的修辞游戏,而是发生在澳大利亚和巴西的实证样…

AITOPRead
Aug 2