SkillNet: Create, Evaluate, and Connect AI Skills
为解决AI技能缺乏系统积累和转移的问题,提出SkillNet,一个创建、评估和组织AI技能的开放基础设施。
为解决AI技能缺乏系统积累和转移的问题,提出SkillNet,一个创建、评估和组织AI技能的开放基础设施。
This study empirically tests Piketty’s central hypothesis that the “r−g gap”—the difference between the rate of return on capital (r) and the economic growth rate (g)—drives rising income inequality and capital’s share of national income. Using panel data from 19 advanced economies over 1980–2015, we estimate a heterogeneous dynamic panel structural vector autoregression (Panel SVAR) model—the first systematic causal identification of r−g’s effects on inequality and capital share. We conduct robustness checks across multiple operationalizations of r−g. Contrary to Piketty’s prediction, we find no statistically significant evidence that an expanding r−g gap increases either top income inequality or capital’s share of national income; results hold across all specifications. By providing the first formal econometric validation—rather than mere descriptive correlation—this work overcomes a longstanding methodological gap in the literature and establishes a new empirical benchmark for analyzing the drivers of inequality dynamics.
To address the convergence difficulties and high communication overhead of large language models (LLMs) in federated learning (FL) caused by data heterogeneity, this paper introduces FedLLM—the first unified analytical framework for LLMs in FL. It systematically surveys two dominant paradigms: federated fine-tuning and federated prompt learning, while rigorously analyzing core challenges including data heterogeneity, communication efficiency, and privacy preservation. The work identifies promising future directions—namely, federated pre-training and LLM-augmented FL—and fills a critical gap in systematic literature review. A multidimensional taxonomy and evaluation framework is established to clarify key technical bottlenecks. Integrating insights from FL, LLM adaptation, prompt engineering, distributed optimization, and privacy-preserving computation, this study delivers a practical, robust, and privacy-aware methodology for deploying LLMs in real-world federated settings. (149 words)
This paper addresses the problem of optimal treatment allocation under a budget constraint when treatment costs vary heterogeneously with covariates. We propose a threshold rule based on a priority score and establish, for the first time, a theoretical link between optimal allocation under uncertain costs and instrumental variable (IV) estimation of heterogeneous treatment effects. We rigorously derive the optimal threshold structure and prove its learnability. Our method integrates randomized controlled trial data, priority score modeling, threshold-based decision making, and an IV estimation framework. Empirically, the approach significantly outperforms standard benchmarks across multiple evaluation metrics, achieving maximal social value or firm profit within budget constraints. It provides a new paradigm—statistically rigorous yet practically implementable—for applications including scarce healthcare resource allocation and dynamic pricing.
This work proposes TTT-Discover, a novel approach that introduces test-time training (TTT) to scientific discovery by leveraging online reinforcement learning to optimize large language models during inference. Unlike conventional AI methods that rely solely on the generalization of pretrained models and struggle to autonomously identify optimal solutions at test time, TTT-Discover focuses on generating a single high-quality solution rather than improving average performance. Built upon the open-source model OpenAI gpt-oss-120b and augmented with a customized search subroutine and the Tinker API, the method achieves highly efficient and low-cost optimization. It sets new state-of-the-art results across diverse domains—including mathematical theorem proving, GPU kernel design, algorithmic competitions, and single-cell denoising—with all findings validated by domain experts or competition organizers at a cost of only a few hundred dollars per task.
本文提出AdaGeoVLN方法,通过层次化融合和导航感知记忆解决视觉-语言导航中几何特征使用及历史几何证据保留问题。
本文针对多语言模型生成特定目标语言连贯文本的问题,提出了一种最优控制方法,并开发了评估生成文本质量的框架。
本文研究了在最坏情况数据模型下,使用PrivTree算法构建自适应二叉分区并发布其叶质量以生成ε-差分隐私合成度量的方法,该方法的1-Wasserstein误差依赖于内在维度而非环境维度。
研究解决了注意力重排序中的校准问题,通过提出插值空校准方法来控制指令内容对空基线的影响,从而在指令密集任务中恢复性能。
本文通过引入PrimeScientist解决自主研究中资源分配问题,利用自适应MCTS策略平衡探索与利用,提高了研究效率和质量。