Curating the frontier of Artificial Intelligence

A Temporal Planning Framework for Disruption Aware Dynamic Route Optimization in Heterogeneous Railway Systems

高效的线路优化对于确保铁路运营的安全和准点起着至关重要的作用。这是非常重要的，特别是在异国情调...

Efficient route optimization play a vital role in ensuring both safety and punctuality in railway operations. It is very crucial particularly in heter...

Abstracting Cross-Domain Action Sequences into Interpretable Workflows

顺序或带时间戳的交互日志提供了数字应用程序使用情况的客观记录,，但它们的粒度和噪音常常掩盖了平均值...

Sequential or time-stamped interaction logs provide objective records of digital application usage, yet their granularity and noise often obscure mean...

Towards Direct Latent-Space Synthesis for Parallel Branches in LLM-Agent Workflows

大型语言模型越来越多地充当代理系统,的执行引擎，但它们仍然通过顺序文本接口消耗上下文......

Large language models increasingly serve as execution engines for agentic systems, yet they still consume context through a sequential text interface....

Multi-Agent Reinforcement Learning from Delayed Marketplace Feedback for Objective-Weight Adaptation in Three-Sided Dispatch

三边市场中的调度为从世界反馈中进行强化学习提供了一个自然的环境: 决策是通过延迟操作来评估的...

Dispatch in three-sided marketplaces provides a natural setting for reinforcement learning from world feedback: decisions are evaluated by delayed ope...

Reasoning as Pattern Matching: Shared Mechanisms in Human and LLM Everyday Reasoning

当大型语言模型 (LLMs) 无法泛化或在推理中出现偶然错误, 时，通常会被视为 LLM 并不真正可靠的证据...

When large language models (LLMs) fail to generalize or make haphazard errors in reasoning, it is often taken as evidence that LLMs are not truly reas...

AgentBeats: Agentifying Agent Assessment for Openness, Standardization, and Reproducibility

代理系统正在跨领域快速发展,，但它们的评估仍然分散。大多数基准测试都依赖于固定, LLM 中心的工具，这些工具...

Agent systems are advancing quickly across domains, but their evaluation remains fragmented. Most benchmarks rely on fixed, LLM-centric harnesses that...

Beyond Runtime Enforcement: Shield Synthesis as Defensibility Analysis for Adversarial Networks

屏蔽强化学习通常被视为一种运行时安全机制，它将时间逻辑规范编译成自动机限制...

Shielded reinforcement learning is typically presented as a runtime safety mechanism that compiles temporal-logic specifications into automata restric...

Before You Think: System 0, AI-Mediated Cognition and Cognitive Colonization

本文研究了用于理解人工智能的认知和认知后果的三个最新框架: 三系统理论, ...

This paper examines three recent frameworks for understanding the cognitive and epistemic consequences of artificial intelligence: Tri-System Theory, ...

EurekAgent: Agent Environment Engineering is All You Need For Autonomous Scientific Discovery

基于法学硕士的代理在自动化科学发现方面显示出越来越大的潜力。给定一个可优化的指标和一个执行环境,，他们可以...

LLM-based agents have shown increasing potential in automating scientific discovery. Given an optimizable metric and an execution environment, they ca...

Agents-K1: Towards Agent-native Knowledge Orchestration

目前基于 LLM 的研究代理已经通过代理编排, 取得了进步，但在很大程度上忽视了科学知识编排。现有作品...

Current LLM-based research agents have advanced through agent orchestration, yet largely overlook scientific knowledge orchestration. Existing works o...