Latest

Every published item, in order · 19 items

Blogs News Posts Papers Papers →

10月22日

Papers Builders11:03

Chinese contentEnvironment Duels：评估 LLM 出题与解题双重能力的动态基准

让 9 个模型互相出题、互相解题，结果最强的解题者反而不是最好的出题者，小模型出题更狠，挺有意思的视角。

#LLM #benchmark #智能体 #Qwen3.8-27B

Event

arXiv cs.AI@Shibo Jin et al. (5)2 sourcesSource

10月21日

Papers BuildersSelected13:22

Meta ProPerAgent Harness Lifts Proactive Agent Accuracy to 38%

The core insight is that timing, not memory volume, is the bottleneck for proactive assistants: a dedicated filtering call before the model sees a memory does most of the work. Builders of always-on or wearable agents can apply this retrieve-then-filter pattern without fine-tuning, and RL on top adds a further gain. It is a useful comparison point against standard RAG pipelines that flood the model with context.

#ProPerAgent #Meta #agents #RAG

DAIR.AI@dair_aiSource

Papers BuildersMulti-sourceSelected74°10:11

Agent S4 Evolves Its Own Computer-Use Interfaces

Interface design itself becomes a learnable, evolving capability, meaning smaller open models can beat frontier systems on computer-use tasks at a fraction of the cost. Builders working on agents can reuse the open-source repository and apply the evolution loop to their own workflows. The results set a concrete cost-performance benchmark against GPT-5.5 and Claude Opus 4.8 that teams choosing an agent stack can weigh directly.

#Agent S4 #Qwen3.8-27B #computer-use agents #open-source models

Event

arXiv cs.AI@Saaket Agashe et al. (8)3 sourcesSource

10月20日

Papers Builders11:48

Chinese content研究：GEPA 与 SkillOpt 在 DeepSeek V4 Flash 和 Qwen3.8-27B 本地部署上的效果差异

GEPA 在 MATH-500 上能从 0.828 拉到 0.904，但在 HotpotQA 上完全没用，SkillOpt 还会翻车——如果你在本地跑推理模型做提示词优化，这篇能帮你避坑。

#GEPA #SkillOpt #DeepSeek #Qwen

Event

arXiv: DeepSeek@Andrea Giummarra et al. (6)2 sourcesSource

Papers BuildersMulti-source09:30

Chinese contentBellman Language Models：把 LLM Agent 决策当 MDP 一次性求解

把 Agent 反复遇到的决策当成 MDP 求解一次然后复用，Qwen3.8-27B 在 τ²-bench 上从 29% 冲到 96%，成本降 98%，思路挺新的。

#Bellman Language Models #τ²-bench #Claude Opus 5.5 #Qwen3.8-27B

Event

arXiv cs.AI@Adarsh Vatsa, William Eiers3 sourcesSource

10月19日

Papers Builders10:04

Chinese contentLM2LM：用 RL 训练 LLM 提出语言模型改进方案

用世界模型当奖励模型训练 AI 自己改进模型，Qwen3.8-27B 训完居然超过 GPT-6-sol，成本还降了 33%。

#LM2LM #Qwen3.8-27B #GPT-6-sol #强化学习

Event

arXiv cs.AI@Minhua Lin et al. (10)2 sourcesSource

10月18日

Papers BuildersSelected00:20

Tencent paper details recursive self-improvement for terminal agents

The reward-design audit is the most transferable part: dense progress rewards let failed runs harvest half the reward, so teams training RL agents should measure how much reward failure earns before trusting their curves. RL and agent-training engineers can apply the strict-success-versus-partial-credit split to their own reward pipelines. Compared with generic terminal-agent RL reports, this one pairs benchmark gains with a concrete failure-mode analysis.

#Tencent #Qwen3 #Terminal-Bench #PPO

elvis@omarsar0Source

10月15日

Papers BuildersSelected12:52

BizBench Benchmark Tests Agent Execution Validity in Payments Workflows

The benchmark shows that outcome-only evaluation masks invalid execution, since models can hit the final goal while violating constraints mid-run — a gap that matters most in regulated domains like payments where intermediate errors carry financial or compliance consequences. Agent builders can use the 50-task suite and deterministic checks to measure constraint compliance and tool-call overhead, not just task success. Teams choosing models for workflow automation get a direct comparison of frontier and open models under identical conditions.

#BizBench #agents #benchmark #Claude Opus 5

arXiv: DeepSeek@Enhan Zhao et al. (9)Source

10月9日

Papers BuildersMulti-source10:21

Chinese content评估LLM生成评分标准的干预转移方法

论文提出了一种创新的评分标准评估方法，揭示了不同LLM生成评分标准之间的不对称性，对AI性能监控有重要启示。

#评分标准 #干预转移 #LLM评估 #HealthBench

Event

arXiv: DeepSeek@Erik Skalnes et al. (4)5 sourcesSource

10月8日

Papers Builders10:52

Chinese contentSOTA：用期权隐含收益分布引导的期权交易智能体

用 Qwen 微调出的期权交易智能体，把上千份合约简化成策略选择，六个月样本外赚 18.3%，新闻用法还挺反直觉。

#SOTA #Qwen #期权交易 #LLM智能体

arXiv cs.AI@Yizhen Xie, Mengyang LiuSource

10月7日

PapersMulti-source10:36

Chinese content研究实测 JEV 与 LLM 在七项政治学复现任务中的准确率与成本

想用模型标注文本的研究者可以看看：JEV 号称便宜快，实测发现便宜不成立，校准也只赢了一半。

#JEV #GPT-6 Luna #Qwen3.8-27B #文本标注

Event

arXiv: OpenAI@Steven Denney, Matthew DiGiuseppe11 sourcesSource

10月6日

Papers BuildersMulti-source13:11

Chinese contentJEV 与 GPT-6 Luna、Qwen3.8-27B 在政治学文本标注任务上的对比

有人测了 TypeSafe 那个号称便宜快的 JEV，结果在 OpenAI 批量价格下根本不省钱，校准也没明显赢 Qwen3.8-27B，想做文本标注的可以看看。

#JEV #GPT-6 Luna #Qwen3.8-27B #文本标注

Event

arXiv: OpenAI@Matthew DiGiuseppe, Steven Denney11 sourcesSource

Papers US vs ChinaMulti-sourceSelected01:04

Qwen3.8 27B addition in words tested on DGX Spark

A 27B open-weights model running locally now solves multi-digit addition-in-words at near-perfect accuracy, a task GPT-4o fumbled two years ago. Practitioners evaluating small local models for arithmetic-adjacent workflows get a concrete benchmark on consumer-tier DGX Spark hardware. Comparisons across reasoning-on and reasoning-off settings make it easy to judge whether enabling reasoning is worth the latency.

#Qwen #DGX Spark #reasoning models #open-source models

Event

官方账号Simon Willison’s Weblog3 sourcesSource

9月30日

Papers Builders10:50

Chinese contentSTEPQuant：Delta规则循环状态量化研究

清华团队提出STEPQuant，根据误差大小和内存寿命分配精度，在Qwen和Kimi模型上显著降低内存使用。

#STEPQuant #Qwen3.8-27B #Kimi-Linear-48B-A3B-Instruct #量化

Event

arXiv cs.AI@Bingchen Yao et al. (9)2 sourcesSource

9月29日

Papers Builders18:04

Chinese content生物文献综述代理评估新基准BioLitBench发布

BioLitBench基准用组合可控性方法解决排行榜评估偏差，SCRIBE在生物文献综述任务中表现优异。

#BioLitBench #Qwen3.8-27B #生物医学文献综述 #排行榜评估

Event

arXiv: OpenAI@Zhaowei Han et al. (7)2 sourcesSource

9月28日

Papers BuildersMulti-source79°11:03

Chinese contentNVIDIA 发布 Skill2Env：把社区 Agent Skills 编译成 RL 训练环境

NVIDIA 把社区写的 3.4k 个 Agent Skills 直接变成 RL 训练环境，300 步就把 Qwen 27B 在 Terminal-Bench 上拉高近 5 个点，流水线和负结果都写得很细，做 agent 训练的值得看看仓库。

#NVIDIA #Skill2Env #Qwen3.8 #智能体

Event

shao__meng@shao__meng11 sourcesSource

9月24日

Papers BuildersMulti-sourceSelected73°23:14

NVIDIA's Skill2Env Turns Agent Skills into RL Environments

Rubric-augmented rewards trade raw benchmark score for trajectories a judge rates far higher, a hint that outcome-only RL over-optimises for narrow metrics while behavioural rubrics shape more genuinely skilled agents. Teams training terminal agents can reuse the open pipeline to convert existing Agent Skills corpora into RL environments instead of hand-crafting tasks. The US$90,000 generation cost also sets a concrete budget baseline for anyone planning similar corpus builds.

#NVIDIA #Skill2Env #RL environments #Qwen3.8-27B

Event

DAIR.AI@dair_ai9 sourcesSource

9月17日

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

9月9日

Papers BuildersSelected73°14:17

KVMem improves inference efficiency

KVMem enables local deployment of large models with extended context on consumer hardware, making long-running agents accessible without expensive cloud infrastructure.

#KVMem #Qwen3.8-27B #context window #inference efficiency

elvis@omarsar0Source

9月4日

PapersSelected73°11:22

Minima Quantizes Hybrid LLM Recurrent Layers

This quantization method enables smaller, faster hybrid LLMs without significant performance loss, making large models more accessible for resource-constrained environments.

#Minima #Qwen3.8-27B #Gated DeltaNet #quantization

Event

arXiv cs.AI@Sergii Kozyrev, Davyd Maiboroda2 sourcesSource

8月13日

AITOP12:07

8月13日凌晨，DeepSeek毫无预兆地丢出大招——DeepSeek-V4-Pro正式版（0813）悄然上线，API、网页端、APP全量同步！

8月13日凌晨，DeepSeek毫无预兆地丢出大招——DeepSeek-V4-Pro正式版（0813）悄然上线，API、网页端、APP全量同步！ 📊 硬核数据：从“吊车尾”到“屠榜王” V4 Pro正式版在Agent能力上的提升，已经不…

AITOPRead

8月7日

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…