Dead text or binding clause? Measuring and restoring constraint influence in black-box LLM dialogues

arXiv:2608.…

• 12599v1 Announce Type: new Abstract: Multi-turn dialogues let users re…
• No existing instrument measures this influence per clause, predicts it…

RSS 官方收录 · 可信分层展示

DiG-bench: Discovery in Games

arXiv:2608.12593v1 Announce Type: new Abstract: Discovery---formulating novel generalizations---is a central part of the scientific process.…

• Despite its importance, there is a gap in the current AI benchmark lan…
• To address this gap, we release a new benchmark: DiG-bench (Discovery …

RSS 官方收录 · 可信分层展示

Auditable agentic AI for evidence-grounded thyroid ultrasound diagnosis and reporting

arXiv:2608.…

• 12590v1 Announce Type: new Abstract: Thyroid ultrasound diagnosis requ…
• We present ThyroidXAgent, a clinician-interactive agentic AI system th…

RSS 官方收录 · 可信分层展示

Reasoning Jury: Multi-Model Consensus for Evaluating Reasoning Traces

arXiv:2608.…

• 12585v1 Announce Type: new Abstract: Improving reasoning LLMs requires…
• Additionally, surfacing reasoning mistakes that the model makes would …

RSS 官方收录 · 可信分层展示

Trie Automata for Constrained Decoding over Large Finite Sets

arXiv:2608.…

• 12574v1 Announce Type: new Abstract: Large language models increasingl…
• Current constrained decoding systems handle this through general-purpo…

RSS 官方收录 · 可信分层展示

CAS: A Causal Attribution Score for Local and Global Explainable Artificial Intelligence

arXiv:2608.…

• 12555v1 Announce Type: new Abstract: Predictive explanation methods at…
• We introduce the Causal Attribution Score (CAS), a compact score archi…

RSS 官方收录 · 可信分层展示

$\varepsilon$-MemEvo: Adaptive Cross-Task Memory Transfer for LLM Program Evolution

arXiv:2608.…

• 12522v1 Announce Type: new Abstract: LLM-based program evolution syste…
• We introduce $\varepsilon$-MemEvo, a framework for cross-task knowledg…

RSS 官方收录 · 可信分层展示

Governed Persistent Memory: Source-Bound State Semantics and Fail-Closed Release for Long-Horizon Agents

arXiv:2608.…

• 12476v1 Announce Type: new Abstract: Long-term agent memory is usually…
• We introduce Governed Persistent Memory (GPM), an auditable bitemporal…

RSS 官方收录 · 可信分层展示

MindMemOS: A Portable and Self-Evolving Memory Operating Layer for AI Agents

arXiv:2608.…

• 12428v1 Announce Type: new Abstract: Memory is a core component of AI …
• However, existing memory systems often remain fixed after development,…

RSS 官方收录 · 可信分层展示

Large Language Models Can Follow Instructions, But Not Many at Once: Phase Transitions in Compositional Constraint Satisfaction

arXiv:2608.…

• 12426v1 Announce Type: new Abstract: Large language models are increas…
• Individual constraints are handled proficiently, but the compositional…

RSS 官方收录 · 可信分层展示

Inside Tololo Ugarte's Vaulted "Spyglass" Residence in Limache

Situated in Limache within the Valparaíso Region of Chile, Catalejo House is a striking architectural project designed by the studio Tololo Ugarte.…

• The micro-house is nestled within a monumental landscape defined by an…
• Designed to emerge organically from its context, the residence sits on…

RSS 官方收录 · 可信分层展示

The Incredibles 3 Confirmed as Disney Kicks Off D23 Entertainment Showcase

The Incredibles 3 Confirmed as Disney Kicks Off D23 Entertainment Showcase

• The Incredibles 3 Confirmed as Disney Kicks Off D23 Entertainment Showcase

RSS 官方收录 · 可信分层展示

Research Assistant: AstraZeneca's Agentic System for R&D

arXiv:2608.…

• 12395v1 Announce Type: new Abstract: We describe Research Assistant, a…
• The system provides a chat-style interface that brings together eviden…

RSS 官方收录 · 可信分层展示

Learning to Adapt Cross-Domain Preferences via Meta-LoRA for LLM Personalization

arXiv:2608.…

• 12389v1 Announce Type: new Abstract: Cross-domain zero- or few-shot pe…
• Existing adaptation methods struggle to calibrate update magnitude und…

RSS 官方收录 · 可信分层展示

Dual-Flow Transformers: Decoupling the Primary Prefill Path from Additional Decode Computation

arXiv:2608.…

• 12385v1 Announce Type: new Abstract: As large language models serve mo…
• The two inference phases stress hardware differently: prompt prefill i…

RSS 官方收录 · 可信分层展示

Don't Want Your LLM to Recommend Nuclear Strike? Try Asking It in Japanese

arXiv:2608.…

• 12373v1 Announce Type: new Abstract: Large language models are increas…
• We test nine models from six providers and ask whether the language of…

RSS 官方收录 · 可信分层展示

Position: We Need Practical AI Alignment Methods to Mirror Human Reasoning

arXiv:2608.12372v1 Announce Type: new Abstract: AI systems are increasingly employed as decision aids, decision delegates, or autonomous decision-makers.…

• This position paper argues that in many settings, particularly high-st…
• We review evidence that cognitive alignment improves understandability…

RSS 官方收录 · 可信分层展示

Multi-Agent Scheduling with LLM-Assisted Contract Net Negotiation for Stream Processing in Mobile Edge Computing

arXiv:2608.…

• 12371v1 Announce Type: new Abstract: Stream-processing systems increas…
• This paper proposes \emph{MAS-DecStream}, whose main contribution is \…

RSS 官方收录 · 可信分层展示

Agreement Is Not Alignment: Divergent Moral Grounds in Human and LLM Ethical Judgments

arXiv:2608.12368v1 Announce Type: new Abstract: Agreement with human judgments is a common proxy for evaluating the alignment of large language models (LLMs).…

• Yet agreement in final labels does not show that human annotators and …
• Two agents may reach the same judgment while appealing to different pr…

RSS 官方收录 · 可信分层展示

Position: The Alignment Community is Unintentionally Building a Censor's Toolkit

arXiv:2608.…

• 12346v1 Announce Type: new Abstract: This position paper argues that m…
• By mapping current alignment techniques to the possibility and actual …

RSS 官方收录 · 可信分层展示

Diagnostic Foundation for Evaluating LLMs' Research Integrity as Co-Scientists

arXiv:2608.…

• 12345v1 Announce Type: new Abstract: Language models are increasingly …
• We introduce IntegrityBench, a benchmark evaluating misconduct classif…

RSS 官方收录 · 可信分层展示

Position: Reasoning is a Learnable Rule-Based Process

arXiv:2608.12325v1 Announce Type: new Abstract: Autonomous reasoning is among the most scientifically and economically motivating topics in AI today.…

• Historically the purview of symbolic AI, recent advances have mainly e…
• Despite immense interest and rapid progress, the generative AI communi…

RSS 官方收录 · 可信分层展示

限量手办 + 实景体验，浙江人行NAVIAI2026WRC 福利提前曝光

8 月 19 日至 23 日，2026 世界机器人大会（WRC2026）将在北京亦庄亦创国际会展中心举办。浙江人形机器人创新中心将携旗下 NAVIAI 人形机器人亮相 C-105 展位，集中展示人形机器人在多元场景的落地应用成果。…

• 本次展会上，NAVIAI 将呈现工业制造、智慧零售、家庭服务、遥操数采多场景的实操能力，现场演示拆垛分拣搬运、货品递送、烹饪清洁、远程数据采…
• 另有文娱演绎类场景在 8 月 19 日开展当天作为特别环节限时展示。

RSS 官方收录 · 可信分层展示

项目频道 - 王者荣耀夏日奇幻短片：AI驯服人类实录

原标题：惊！王者荣耀AI夏日奇幻短片居然是这样的——AI驯服人类实录今年夏天，我们做了一条“不一样”的王者荣耀农友节短片。话不多说，先上正片↓从接到Brief时，就很“不一样”，客户想要和我们一起尝试，如何将AI+实拍结合，创造一个奇幻世界！…

• 哪个词单拎出来都足以写成一篇小作文，而这一次我们要将这些元素融在一起。
• 最最最最重要的是，作为夏日农友节的短片，让农友们看了绷不住的农梗必须是其中的主角。

RSS 官方收录 · 可信分层展示

Arc'teryx and BEAMS Repurpose Deadstock GORE-TEX for the STUFF SACK-ReCUT

SummaryArc'teryx and BEAMS drop "STUFF SACK-ReCUT," repurposing deadstock GORE-TEX and nylon scraps into unique storage bags.…

• Each patchwork pouch features distinct color blocking alongside Arc'te…
• Launches August 15, 2026, through BEAMS for ¥5,500 JPY (approx.

RSS 官方收录 · 可信分层展示

面向多模态推理的高效长上下文建模｜AICon深圳

点击查看原文>

RSS 官方收录 · 可信分层展示

文章频道 - 3个月狂卖1440万瓶，这瓶绿可乐什么来头

它踩准了无糖可乐甜味剂升级的结构性机会。文：Hermia He，首发：Foodaily每日食品原标题：可口可乐、朝日集体换“糖”！3个月狂卖1440万瓶，这瓶绿可乐什么来头？一瓶没有颠覆性新原料的无糖可乐，在日本卖爆了，凭什么？…

• 8月4日，朝日引进的希腊可乐品牌——Green Cola（绿可乐），正式在日本全国范围内开售。
• 上市才3个月，销量已经突破1440万瓶。

RSS 官方收录 · 可信分层展示

智谱发布 GLM-5.3，编程能力更强；传苹果训练国内专用 AI 模型；微信：朋友圈现在、过去、未来都不会有二次编辑功能 | 极客早知道

智谱正式发布 GLM-5.3，拥有更强编程能力 8 月 14 日，据介绍，与 GLM-5.2 相比，GLM-5.3 基座模型未变，但通过极致的后训练 Scaling 大大提高了模型的智能上界。…

• 3 拥有更强的编程能力，在内部自建体感评测中较 GLM-5.
• 2 提升 50%，在包括 TerminalBench3.

RSS 官方收录 · 可信分层展示

早报｜曝苹果与阿里合作训练AI模型/微信:永不推出朋友圈二次编辑/售价20万,追觅首台手机交付

曝苹果与阿里合作，为中国市场训练自研 AI 模型 微信确认朋友圈永不推出二次编辑功能 售价 20 万元，追觅首台 AURORA 手机交付：24K 足金镶宝石 Google DeepMind 或裁员三分之一以上，资源转向 Flash WorkBuddy 接入 GLM-5.…

• 3 广州推出「Token 贷」：按算力合同和 Token 消耗额度授信 曝 DeepSeek 正研发情感 AI 模型 调查：美国年轻人普遍不…
• 3，同一基座靠后训练提升编程与网络安全能力 Ling-3.

RSS 官方收录 · 可信分层展示

Mastercard Offers Brazilian Acquirers 50% Payout and Services in Will Bank Dispute

Mastercard is proposing to pay half the amount demanded by Brazilian merchant acquirers that were impacted by the collapse of Will Financeira, also known as Will Bank, a FinTech tied to the failed Banco Master, Bloomberg reported Friday (Aug.…

• 14), citing unnamed sources.
• The offer follows an earlier one that Mastercard offered amid the fall…

RSS 官方收录 · 可信分层展示