定量金融
查看 最近的 文章
显示 2025年09月05日， 星期五 新的列表
- [1] arXiv:2509.03669 [中文pdf, pdf, html, 其他]
-
标题： 带有不对称信息的均值-方差斯塔克尔伯格博弈标题： Mean-Variance Stackelberg Games with Asymmetric Information主题： 数学金融 (q-fin.MF) ; 优化与控制 (math.OC)
本文考虑了两个进行均值-方差投资组合选择的投资者，他们拥有不对称信息：一个知道股票的真实动态，而另一个则必须从观察到的股票演化中推断真实动态。 他们的投资组合选择通过相对业绩关注相互关联，即每个投资者不仅关心自己的最终财富，还关心它与两个投资者平均最终财富的比较。 我们将此建模为斯塔克尔伯格竞争：部分知情的投资者（“跟随者”）观察到完全知情投资者（“领导者”）的交易行为，并据此决定自己的交易策略；领导者预期跟随者的反应，进而选择最适合其目标的交易策略。 为了防止信息泄露，领导者采用在熵正则化均值-方差目标下选择的随机策略，其中熵正则化器量化了所选策略的随机性。 另一方面，跟随者只能观察到领导者实际的交易行为（从随机策略中抽样），而无法观察到随机策略本身。 因此，她的均值-方差目标是一个随机场，形式为对领导者交易行为实现路径的条件期望。 在连续抽样领导者交易行为的理想情况下，我们推导出一个斯塔克尔伯格均衡，其中跟随者的交易策略线性依赖于领导者的实际交易行为，而领导者则从高斯分布中抽样她的交易行为。 在领导者交易行为离散抽样的现实情况下，上述情况变为 $\epsilon$-斯塔克尔伯格均衡。
This paper considers two investors who perform mean-variance portfolio selection with asymmetric information: one knows the true stock dynamics, while the other has to infer the true dynamics from observed stock evolution. Their portfolio selection is interconnected through relative performance concerns, i.e., each investor is concerned about not only her terminal wealth, but how it compares to the average terminal wealth of both investors. We model this as Stackelberg competition: the partially-informed investor (the "follower") observes the trading behavior of the fully-informed investor (the "leader") and decides her trading strategy accordingly; the leader, anticipating the follower's response, in turn selects a trading strategy that best suits her objective. To prevent information leakage, the leader adopts a randomized strategy selected under an entropy-regularized mean-variance objective, where the entropy regularizer quantifies the randomness of a chosen strategy. The follower, on the other hand, observes only the actual trading actions of the leader (sampled from the randomized strategy), but not the randomized strategy itself. Her mean-variance objective is thus a random field, in the form of an expectation conditioned on a realized path of the leader's trading actions. In the idealized case of continuous sampling of the leader's trading actions, we derive a Stackelberg equilibrium where the follower's trading strategy depends linearly on the actual trading actions of the leader and the leader samples her trading actions from Gaussian distributions. In the realistic case of discrete sampling of the leader's trading actions, the above becomes an $\epsilon$-Stackelberg equilibrium.
- [2] arXiv:2509.03712 [中文pdf, pdf, html, 其他]
-
标题： 拉丁美洲NUAM市场的投资组合分配层次风险平价标题： Hierarchical Risk Parity for Portfolio Allocation in the Latin American NUAM Market主题： 投资组合管理 (q-fin.PM)
本研究将分层风险平价（HRP）投资组合分配方法应用于NUAM市场，这是一个整合智利、哥伦比亚和秘鲁市场的区域控股。 作为对该新成立的拉丁美洲背景下HRP的首批实证分析之一，本文填补了在跨境新兴市场条件下投资组合构建文献中的一个空白。 HRP利用分层聚类和递归二分法以一种可解释且稳健的方式分配风险——避免了传统均值-方差优化中需要求逆协方差矩阵这一常见限制。 使用2019年至2025年MSCI NUAM指数54只成分股的每日数据，我们将HRP的表现与两个标准基准进行比较：等权重投资组合（1/N）和最大夏普比率投资组合。 结果表明，虽然最大夏普比率投资组合产生最高的收益，但HRP投资组合提供了更平稳的风险-收益曲线，具有较低的回撤和跟踪误差。 这些发现突显了HRP作为投资者在整合性高波动市场如NUAM中实用且有弹性的资产配置框架的潜力。
This study applies the Hierarchical Risk Parity (HRP) portfolio allocation methodology to the NUAM market, a regional holding that integrates the markets of Chile, Colombia and Peru. As one of the first empirical analyses of HRP in this newly formed Latin American context, the paper addresses a gap in the literature on portfolio construction under cross-border, emerging market conditions. HRP leverages hierarchical clustering and recursive bisection to allocate risk in a manner that is both interpretable and robust--avoiding the need to invert the covariance matrix, a common limitation in the traditional mean-variance optimization. Using daily data from 54 constituent stocks of the MSCI NUAM Index from 2019 to 2025, we compare the performance of HRP against two standard benchmarks: an equally weighted portfolio (1/N) and a maximum Sharpe ratio portfolio. Results show that while the Max Sharpe portfolio yields the highest return, the HRP portfolio delivers a smoother risk-return profile, with lower drawdowns and tracking error. These findings highlight HRP's potential as a practical and resilient asset allocation framework for investors operating in the integrated, high-volatility markets like NUAM.
- [3] arXiv:2509.03916 [中文pdf, pdf, html, 其他]
-
标题： 监管或竞争：在暗池和明池中的主要-次要最优清算标题： Regulation or Competition:Major-Minor Optimal Liquidation across Dark and Lit Pools主题： 数学金融 (q-fin.MF) ; 优化与控制 (math.OC)
我们研究了在连续时间设定下，面对执行不确定性时，投资者在公开和暗池中的最优清算问题。 首先，我们为一个在两个池中清算其头寸的大投资者设计了一个最优做市-撮合费用政策，她与支付交易费用的小投资者进行互动。 我们在交易所提出的一种旨在缓解公开场所市场影响的补偿方案下，使用BSDEs显式表征了大投资者在公开和暗池中的最优清算策略。 其次，我们考虑了一个没有监管的主导-从属交易者纯竞争模型。 我们提供了相关HJB-福克-普朗克系统的显式解。 最后，我们通过数值实验说明了我们的结果，将有战略大投资者的监管市场中的市场影响与既有小投资者又有大投资者的纯竞争市场中的市场影响进行了比较。
We study the optimal liquidation problem in both lit and dark pools for investors facing execution uncertainty in a continuous-time setting with market impact. First, we design an optimal make--take fee policy for a large investor liquidating her position across both pools, interacting with small investors who pay trading fees. We explicitly characterize the large investor's optimal liquidation strategies in both lit and dark pools using BSDEs under a compensation scheme proposed by an exchange to mitigate market impact in the lit venue. Second, we consider a purely competitive model with major--minor traders in the absence of regulation. We provide explicit solutions to the associated HJB--Fokker--Planck system. Finally, we illustrate our results through numerical experiments, comparing market impact under a regulated market with a strategic large investor to that in a purely competitive market with both small and large investors.
- [4] arXiv:2509.03926 [中文pdf, pdf, 其他]
-
标题： 国家碳的社会成本：FUND的应用标题： National social cost of carbon: An application of FUND主题： 一般经济学 (econ.GN)
本文提出了一种改进的国家层面综合评估模型，FUND 3.9n，该模型通过纳入分部门的气候影响函数和198个单独国家的参数不确定性分析，扩展了区域FUND 3.9框架。 该模型能够估算国家碳的社会成本（NSCC），捕捉从经济结构、气候敏感性和人口暴露度等方面各国之间的异质性。 我们的结果表明，NSCC和全球总和估计值对损害规范和偏好参数高度敏感，包括纯时间贴现率和相对风险厌恶程度。 与汇总的单部门方法相比，包含不确定性的分解模型为低收入和中等收入国家提供了更高的NSCC值。 本文通过量化分部门脆弱性和随机变异性如何加剧气候变化损害，并重塑NSCC分布中的全球公平性，为文献做出了贡献。 我们模型得出的NSCC为适应规划、减缓目标设定以及国际气候谈判中的公平负担分担提供了政策相关指标。 这种方法弥合了全球统一碳定价与国家差异化气候影响之间的差距，为未来气候政策设计提供了一个理论上有依据且实证丰富的框架。
This paper presents a refined country-level integrated assessment model, FUND 3.9n, that extends the regional FUND 3.9 framework by incorporating sector-specific climate impact functions and parametric uncertainty analysis for 198 individual countries. The model enables estimation of the national social cost of carbon (NSCC), capturing heterogeneity across nations from economic structure, climate sensitivity, and population exposure. Our results demonstrate that both the NSCC and the global sum estimates are highly sensitive to damage specifications and preference parameters, including the pure rate of time preference and relative risk aversion. Compared to aggregated single-sector approaches, the disaggregated model with uncertainty yields higher values of the NSCC for low- and middle-income countries. The paper contributes to the literature by quantifying how sector-specific vulnerabilities and stochastic variability amplify climate damages and reshape global equity in the distribution of the NSCC. The NSCCs derived from our model offer policy-relevant metrics for adaptation planning, mitigation target setting, and equitable burden-sharing in international climate negotiations. This approach bridges the gap between globally harmonized carbon pricing and nationally differentiated climate impacts, providing a theoretically grounded and empirically rich framework for future climate policy design.
- [5] arXiv:2509.03964 [中文pdf, pdf, html, 其他]
-
标题： 加密货币和利率：在无债券市场中推断收益率曲线标题： Cryptocurrencies and Interest Rates: Inferring Yield Curves in a Bondless Market主题： 一般金融 (q-fin.GN)
在传统金融市场中，收益率曲线对于国家（以及由此延伸的货币）、金融机构和大型企业广泛可用。 这些曲线用于校准随机利率模型、折现未来现金流和定价金融产品。 然而，收益率曲线只能由于当前债券市场的规模和结构而轻松计算。 在加密货币市场中，截至2025年初，固定利率贷款和债券几乎不存在，因此每种货币的收益率曲线必须通过其他方式估算。 在本文中，我们展示了如何利用数学工具，通过利用高度发达的加密货币衍生品市场数据来构建加密货币的收益率曲线。
In traditional financial markets, yield curves are widely available for countries (and, by extension, currencies), financial institutions, and large corporates. These curves are used to calibrate stochastic interest rate models, discount future cash flows, and price financial products. Yield curves, however, can be readily computed only because of the current size and structure of bond markets. In cryptocurrency markets, where fixed-rate lending and bonds are almost nonexistent as of early 2025, the yield curve associated with each currency must be estimated by other means. In this paper, we show how mathematical tools can be used to construct yield curves for cryptocurrencies by leveraging data from the highly developed markets for cryptocurrency derivatives.
- [6] arXiv:2509.04146 [中文pdf, pdf, html, 其他]
-
标题： 具有损失厌恶买家的双寡头环境中的噪声认证标题： Noisy Certification in a Duopolistic Setting with Loss-Averse Buyers主题： 一般经济学 (econ.GN)
本文研究了认证技术中的噪声如何影响寡头垄断中卖方的利润。 我们识别了噪声认证的两种对立效应。 首先，它降低了认证结果的信息量，使买方的信念趋于一致，限制了垂直差异化的空间。 其次，它将随机性引入买方的感知中，内生地产生了原本相似产品的差异化。 当买方风险中性时，第一种效应占主导地位，从而降低卖方利润。 然而，当买方损失厌恶时，信息量减少的负面影响会被缓解，噪声认证相对于准确认证可以提高利润。 实验中，不准确认证的处理比准确认证的处理更具盈利能力，尤其是在竞争激烈的环境中。
This paper studies how noise in certification technology affects seller profits in a duopoly with unobservable product quality. We identify two opposing effects of noisy certification. First, it reduces the informativeness of certification outcomes, homogenizing buyers' beliefs and limiting the scope for vertical differentiation. Second, it introduces randomness into buyer perceptions, endogenously generating differentiation between otherwise similar products. When buyers are risk-neutral, the first effect dominates, reducing seller profits. However, when buyers are loss averse, the negative impact of reduced informativeness is mitigated, and noisy certification can increase profits relative to accurate certification. Experimentally, treatments with inaccurate certification are more profitable than those with accurate certification, particularly in settings with intense competition.
- [7] arXiv:2509.04307 [中文pdf, pdf, 其他]
-
标题： 旅游业何时提高地价？ 阈值效应、超级城市与日本的政策启示标题： When Does Tourism Raise Land Prices? Threshold Effects, Superstar Cities, and Policy Lessons from Japan主题： 一般经济学 (econ.GN)
虽然旅游业被广泛视为经济和城市转型的催化剂，但其对土地价格的影响仍存在争议。 本研究使用2021年至2024年间日本1,724个市町村的面板数据，将年度游客到访量作为旅游活动的代理指标。 通过中介分析和面板门槛回归，我们发现显著的土地价格上涨集中在少数“明星”城市中，特别是游客到访量排名前5.9%的城市，而大多数市町村则几乎没有或没有受到影响。 结果突显了日本各地旅游业经济影响中的明显非线性和空间异质性。 旅游业与土地价格上涨之间潜在的作用机制是混合的，可能为当地居民带来好处，也可能增加负担的风险。 这些发现强调了制定促进包容性增长和公平分配旅游相关收益政策的必要性。
While tourism is widely regarded as a catalyst for economic and urban transformation, its effects on land prices remain contested. This study examines tourism and land prices using a panel of 1,724 Japanese municipalities from 2021 to 2024, with annual tourist arrivals as a proxy for tourism activity. Using mediation analysis and panel threshold regression, we show that sizable land price increases are concentrated in a small group of "superstar" cities, specifically those in the top 5.9 percent for tourist arrivals, while most municipalities experience little or no effect. The results highlight pronounced nonlinearities and spatial heterogeneity in tourism's economic impact across Japan. The potential mechanisms linking tourism to land price growth are mixed, with possible benefits for local residents as well as risks of increased burdens. These findings underscore the need for policies that promote inclusive growth and an equitable distribution of tourism-related gains.
新提交 (展示 7 之 7 条目 )
- [8] arXiv:2509.03533 (交叉列表自 cs.CL) [中文pdf, pdf, html, 其他]
-
标题： 通过信息瓶颈视角对LLM输入输出对进行主题识别标题： Topic Identification in LLM Input-Output Pairs through the Lens of Information Bottleneck评论： 26页，4图主题： 计算与语言 (cs.CL) ; 机器学习 (cs.LG) ; 一般金融 (q-fin.GN)
大型语言模型（LLMs）容易出现关键故障模式，包括\textit{内在忠实性幻觉}（也称为虚构内容），其中响应在语义上偏离了提供的上下文。 设计用于检测此问题的框架，如语义偏差度量（SDM），通过对其句子嵌入应用几何聚类来识别提示和响应之间的潜在主题。 这导致了一个脱节，因为这些主题是为空间接近性优化的，而不是为下游信息论分析优化的。 在本文中，我们通过开发一种基于几何聚类的确定性信息瓶颈（DIB）原理性主题识别方法来弥合这一差距。 我们的主要贡献是通过用计算高效的上界替换其难以处理的KL散度项，将DIB方法转化为适用于高维数据的实用算法。 所得方法，我们称之为UDIB，可以解释为一种熵正则化和稳健化的K-means版本，它本质上倾向于少量的信息性聚类。 通过将UDIB应用于LLM提示和响应嵌入的联合聚类，我们生成了一个共享主题表示，该表示不仅在空间上一致，而且从根本上被构建为关于提示-响应关系的最大信息量。 这为SDM框架提供了更优越的基础，并提供了一种新颖且更敏感的检测虚构内容的工具。
Large Language Models (LLMs) are prone to critical failure modes, including \textit{intrinsic faithfulness hallucinations} (also known as confabulations), where a response deviates semantically from the provided context. Frameworks designed to detect this, such as Semantic Divergence Metrics (SDM), rely on identifying latent topics shared between prompts and responses, typically by applying geometric clustering to their sentence embeddings. This creates a disconnect, as the topics are optimized for spatial proximity, not for the downstream information-theoretic analysis. In this paper, we bridge this gap by developing a principled topic identification method grounded in the Deterministic Information Bottleneck (DIB) for geometric clustering. Our key contribution is to transform the DIB method into a practical algorithm for high-dimensional data by substituting its intractable KL divergence term with a computationally efficient upper bound. The resulting method, which we dub UDIB, can be interpreted as an entropy-regularized and robustified version of K-means that inherently favors a parsimonious number of informative clusters. By applying UDIB to the joint clustering of LLM prompt and response embeddings, we generate a shared topic representation that is not merely spatially coherent but is fundamentally structured to be maximally informative about the prompt-response relationship. This provides a superior foundation for the SDM framework and offers a novel, more sensitive tool for detecting confabulations.
交叉提交 (展示 1 之 1 条目 )
- [9] arXiv:2411.05470 (替换) [中文pdf, pdf, html, 其他]
-
标题： 无模型连续时间投资组合配置标题： Model-free portfolio allocation in continuous-time主题： 数学金融 (q-fin.MF)
我们提出了一种非概率的、逐路径的框架，用于研究基于[Chiu & Cont '23]的连续时间下路径依赖（即权重是时间和历史时间序列的泛函）的长期投资组合配置，其中引入了独立于任何积分理论的自融资基本概念。在本文中，我们将这一概念扩展到一种投资组合配置策略，并通过一个路径依赖的偏微分方程来描述它。我们推导出一个通用的显式解，该解描述了在通用市场中财富的演变，包括可能不连续演变或具有任意阶变化的价格路径。提供了显式解的例子。作为我们连续时间、路径依赖框架的应用，我们扩展了[Vovk '90]的一个聚合算法和[Cover '91]的通用算法，使其成为将多个策略结合为单一策略的连续时间算法。这些连续时间（元）算法以多个策略作为输入（这些策略本身可能由其他算法生成），并跟踪最佳单独策略和最佳凸组合策略产生的财富，其对数财富的跟踪误差分别为O(1)和O(ln t的量级。这项工作将Cover定理[Cover '91, Thm 6.1]扩展到了连续时间、无模型的设置。
We present a non-probabilistic, path-by-path framework for studying path-dependent (i.e., where weight is a functional of time and historical time-series), long-only portfolio allocation in continuous-time based on [Chiu & Cont '23], where the fundamental concept of self-financing was introduced, independent of any integration theory. In this article, we extend this concept to a portfolio allocation strategy and characterize it by a path-dependent partial differential equation. We derive the general explicit solution that describes the evolution of wealth in generic markets, including price paths that may not evolve continuously or exhibit variation of any order. Explicit solution examples are provided. As an application of our continuous-time, path-dependent framework, we extend an aggregating algorithm of [Vovk '90] and the universal algorithm of [Cover '91] to continuous-time algorithms that combine multiple strategies into a single strategy. These continuous-time (meta) algorithms take multiple strategies as input (which may themselves be generated by other algorithms) and track the wealth generated by the best individual strategy and the best convex combination of strategies, with tracking error bounds in log wealth of order O(1) and O(ln t), respectively. This work extends Cover's theorem [Cover '91, Thm 6.1] to a continuous-time, model-free setting.
- [10] arXiv:2502.07896 (替换) [中文pdf, pdf, html, 其他]
-
标题： 行业生产异质性与行业冲击的宏观效应标题： Heterogeneity in Sectoral Production and the Macro Effect of Sectoral Shocks评论： 26页，8张表格，4幅图，6个附录图主题： 一般经济学 (econ.GN)
负部门冲击对GDP的影响取决于受冲击部门作为直接和间接供应商的重要性，以及各部门替换投入品的难易程度。 过去对决定这些特性的参数的估计在美国是有限的：它们不能在不同行业或不同时间之间变化。 本文使用一种新颖的实证策略，在不施加这些限制的情况下，通过利用行业内部投入支出份额的变化而不是行业之间的变化，来生成这些参数的新估计。 这些结果表现出显著的部门和时间异质性。 在一个多部门生产的校准一般均衡模型中，这种异质性（1）提高了（降低了）那些客户更少（更多）能够替代投入品的部门的负面冲击对GDP的影响（例如，"化学产品"冲击对GDP的影响增加了15%），并且（2）随着部门成为更[更少]重要的投入供应商，提高了（降低了）负面冲击对GDP的影响（例如，在1997年至2023年间，由于其作为投入供应商的重要性发生变化，"数据处理、互联网出版和其他信息服务"部门的冲击对GDP的影响几乎翻了三倍）。
The effect of a negative sectoral shock on GDP depends on how important the shocked sector is as a direct and indirect supplier and how easily sectors can substitute inputs. Past estimates of the parameters that determine these qualities in the US have been restrictive: they have not been allowed to vary across industries or across time. This paper uses a novel empirical strategy to produce new estimates of these parameters without these restrictions, by exploiting variation in input expenditure shares within industries rather than across industries. The resulting estimates exhibit significant sectoral and temporal heterogeneity. In a calibrated GE model of multi-sector production, this heterogeneity (1) raises(lowers) the GDP effect of negative shocks to sectors whose customers are less(more) able to substitute inputs (e.g. the GDP effect of a "Chemical products" shock rises by 15%), and (2) raises(lowers) the GDP effect of negative shocks to sectors as they become more[less] central input suppliers (e.g. between 1997 and 2023 the GDP effect of a shock to the "Data processing, internet publishing, and other information services" sector nearly tripled due to changes in its importance as an input supplier).
- [11] arXiv:2508.17407 (替换) [中文pdf, pdf, html, 其他]
-
标题： 通用社会代理标题： General Social Agents主题： 一般经济学 (econ.GN)
有用的社会科学理论可以预测不同情境下的行为。 然而，将一种理论应用于新情境中进行预测是具有挑战性的：通常需要进行临时修改以考虑情境特定因素。 我们认为，在模拟这些新情境中的AI代理为应用理论提供了一种替代方案，只需要最少或不需要修改。 我们提出了一种构建此类“通用”代理的方法，这些代理使用基于理论的自然语言指令、现有的实证数据以及在训练过程中获得的知识。 为了在没有来自该数据生成过程的数据的情境中展示该方法——这在应用预测问题中是常见的——我们设计了一个高度异质的883,320个新游戏的人口群体。 AI代理是使用一小组概念相关但结构不同的“种子”游戏的人类数据构建的。 在预注册实验中，平均而言，这些代理在从人口中随机抽取的1,500个游戏中，比(i)博弈论均衡和(ii)开箱即用的代理更好地预测人类行为。 对于一小部分单独的新游戏，这些模拟甚至比最可能相关的已发表人类数据更好地预测了新样本的人类受试者的反应。
Useful social science theories predict behavior across settings. However, applying a theory to make predictions in new settings is challenging: rarely can it be done without ad hoc modifications to account for setting-specific factors. We argue that AI agents put in simulations of those novel settings offer an alternative for applying theory, requiring minimal or no modifications. We present an approach for building such "general" agents that use theory-grounded natural language instructions, existing empirical data, and knowledge acquired by the underlying AI during training. To demonstrate the approach in settings where no data from that data-generating process exists--as is often the case in applied prediction problems--we design a highly heterogeneous population of 883,320 novel games. AI agents are constructed using human data from a small set of conceptually related, but structurally distinct "seed" games. In preregistered experiments, on average, agents predict human play better than (i) game-theoretic equilibria and (ii) out-of-the-box agents in a random sample of 1,500 games from the population. For a small set of separate novel games, these simulations predict responses from a new sample of human subjects better even than the most plausibly relevant published human data.
- [12] arXiv:2411.17683 (替换) [中文pdf, pdf, html, 其他]
-
标题： 应对黑暗期：欧洲可变可再生能源干旱对电力系统的影响标题： Coping with the Dunkelflaute: Power system implications of variable renewable energy droughts in Europe主题： 物理与社会 (physics.soc-ph) ; 一般经济学 (econ.GN)
应对风能和太阳能供应长期不足的情况，也称为可再生能源干旱或“黑暗风暴”，成为实现基于可再生能源的去碳化能源系统的关键挑战。 在此我们研究了长时间电力存储和地理平衡在应对此类事件中的作用，结合了可再生能源可用性的时序分析与35个历史天气年份的电力部门建模。 我们发现极端干旱定义了长时间存储的运行和投资。 假设政策相关的互联，我们发现351太瓦时的长时间存储容量或占年度电力需求的7%是应对欧洲最极端事件的最佳选择。 虽然核能可以部分减少存储需求，但化石燃料备用电厂结合碳去除的存储缓解效果是有限的。 政策制定者和系统规划者应为长时间存储的快速扩展做好准备，以保障欧洲的可再生能源转型。
Coping with prolonged periods of low availability of wind and solar power, also referred to as renewable energy droughts or "Dunkelflaute", emerges as a key challenge for realizing decarbonized energy systems based on renewable energy sources. Here we investigate the role of long-duration electricity storage and geographical balancing in dealing with such events, combining a time series analysis of renewable availability with power sector modeling of 35 historical weather years. We find that extreme droughts define long-duration storage operation and investment. Assuming policy-relevant interconnection, we find 351 TWh long-duration storage capacity or 7% of yearly electricity demand optimal to deal with the most extreme event in Europe. While nuclear power can partially reduce storage needs, the storage-mitigating effect of fossil backup plants in combination with carbon removal is limited. Policymakers and system planners should prepare for a rapid expansion of long-duration storage to safeguard the renewable energy transition in Europe.
- [13] arXiv:2504.11671 (替换) [中文pdf, pdf, html, 其他]
-
标题： LLM在社会模拟中决策的计算基础标题： Computational Basis of LLM's Decision Making in Social Simulation主题： 人工智能 (cs.AI) ; 计算机与社会 (cs.CY) ; 机器学习 (cs.LG) ; 一般经济学 (econ.GN)
大型语言模型（LLMs）在社会科学和应用环境中越来越作为类似人类的决策代理。这些LLM代理通常被赋予类似人类的角色并置于现实生活情境中。然而，这些角色和情境如何塑造LLM的行为仍缺乏深入研究。本研究提出了并在独裁者游戏中测试了探测、量化和修改LLM内部表示的方法——这是一个关于公平性和亲社会行为的经典行为实验。我们从LLM的内部状态中提取“变量变化向量”（例如，“男性”到“女性”）。在模型推理过程中操纵这些向量可以显著改变这些变量与模型决策之间的关系。这种方法为研究和调节社会概念如何在基于变压器的模型中被编码和工程化提供了一种系统的方法，对对齐、去偏、设计用于社会模拟的AI代理在学术和商业应用中的影响具有重要意义，加强了社会学理论和测量。
Large language models (LLMs) increasingly serve as human-like decision-making agents in social science and applied settings. These LLM-agents are typically assigned human-like characters and placed in real-life contexts. However, how these characters and contexts shape an LLM's behavior remains underexplored. This study proposes and tests methods for probing, quantifying, and modifying an LLM's internal representations in a Dictator Game -- a classic behavioral experiment on fairness and prosocial behavior. We extract ``vectors of variable variations'' (e.g., ``male'' to ``female'') from the LLM's internal state. Manipulating these vectors during the model's inference can substantially alter how those variables relate to the model's decision-making. This approach offers a principled way to study and regulate how social concepts can be encoded and engineered within transformer-based models, with implications for alignment, debiasing, and designing AI agents for social simulations in both academic and commercial applications, strengthening sociological theory and measurement.