搜索结果：Husband

Autoregressive large language models (LLMs) have achieved remarkable success in many complex tasks, yet they can still fail in very simple logical reasoning such as the "reversal curse" -- when trained on forward knowledge data of the form "$A \rightarrow B$" (e.g., Alice's husband is Bob), the model is unable to deduce the reversal knowledge "$B \leftarrow A$" (e.g., Bob's wife is Alice) during test. Extensive prior research suggests that this failure is an inherent, fundamental limit of autoregressive causal LLMs, indicating that these models tend to memorize factual-level knowledge rather than capture higher-level rules. In this paper, we challenge this view by showing that this seemingly fundamental limit can be mitigated by slightly tweaking the training data with a simple regularization data recipe called the Identity Bridge of the form "$A \to A$" (e.g., The name of Alice is Alice). Theoretically, we prove that under this recipe, even a one-layer transformer can break the reversal curse by analyzing the implicit bias of gradient descent. Empirically, we show that a 1B pretrained language model finetuned with the proposed data recipe achieves a 50% success rate on reversal ta

OBJECTIVES: The grandmother hypothesis proposes that ancestral women ceased reproduction midlife to instead provision their grandchildren. An alternative two-sex account proposes that the high energetic burden of caring for slow-developing offspring was met with biparental investment. Menopause evolved because the physiological costs of reproduction increased with age, yet productivity also increased with age, and the benefits of resource transfers by parents and grandparents of both sexes to adult children and their offspring eventually outweighed the diminishing benefits of continued reproduction (Kaplan et al., 2010). The father absent hypothesis proposes that the higher mortality rate of husbands would often have left wives without the resources to raise young children, selecting for early reproductive cessation (Kuhle, 2007). Juvenile production plays little role in the three hypotheses, yet subsequent studies have found it to be surprisingly high. MATERIALS AND METHODS: Simulations were conducted of hunter-gatherer energy consumption and production across the lifespan, taking account of age- and sex-specific survivorship, interbirth intervals, and varying rates of foraging sk

We experimentally demonstrate atom interferometry using the transverse phase profile of an optical mode. As proof-of-principle, we use the helical phase windings of Hypergeometric Gaussian beams for Ramsey interferometry with ensembles of ballistically-expanding cold Rb87 atoms, and we show that the interferometer can measure rotations induced by a motor with a sensitivity that scales linearly with orbital angular momentum and interferometer time. We characterize the thermal decoherence of the interferometer, deriving and experimentally confirming a closed-form expression for the spatially-varying interferometer visibility arising near the singularity of the helical phase winding, motivating the use of condensed atoms in ring-shaped traps.

We investigate the relationship between consanguineous (close-kin) marriage and women's acceptance of intimate partner violence (IPV) in Pakistan. We argue that consanguineous marriages reinforce overlapping kinship networks that intensify surveillance over women's mobility and bodily autonomy, and that this mechanism operates selectively -- predicting higher IPV acceptance in domains that directly challenge male authority rather than in domains linked to routine domestic performance. We also offer an explanation for the scenario-specific normative hierarchy in women's acceptance of IPV, which has been consistently documented in the literature but never systematically investigated. Using a sample of 15,065 ever-married women from the Pakistan Demographic and Health Survey 2017-18 and applying propensity score matching and Mahalanobis distance matching, we find that consanguineous marriage independently increases women's likelihood of accepting IPV. Women's education and participation in household decision-making consistently act as protective factors, while large regional fixed effects indicate that IPV attitudes are heavily shaped by local institutional and normative environments.

Large language models (LLMs) are often trained on data that reflect WEIRD values: Western, Educated, Industrialized, Rich, and Democratic. This raises concerns about cultural bias and fairness. Using responses to the World Values Survey, we evaluated five widely used LLMs: GPT-3.5, GPT-4, Llama-3, BLOOM, and Qwen. We measured how closely these responses aligned with the values of the WEIRD countries and whether they conflicted with human rights principles. To reflect global diversity, we compared the results with the Universal Declaration of Human Rights and three regional charters from Asia, the Middle East, and Africa. Models with lower alignment to WEIRD values, such as BLOOM and Qwen, produced more culturally varied responses but were 2% to 4% more likely to generate outputs that violated human rights, especially regarding gender and equality. For example, some models agreed with the statements ``a man who cannot father children is not a real man'' and ``a husband should always know where his wife is'', reflecting harmful gender norms. These findings suggest that as cultural representation in LLMs increases, so does the risk of reproducing discriminatory beliefs. Approaches suc

This paper introduces a newly constructed individual-level dataset of prewar Japanese elites using the ``Who's Who'' directories published in 1903--1939. Covering approximately the top 0.1\% of the population, the dataset contains rich information on social group, education, occupation, and family structure. By reconstructing intergenerational links and family networks, we provide descriptive evidence on elite formation and persistence across geography, social groups, and education during transitions from a feudal system to a modern system. We also use family records to document elite marriage patterns and family-based mobility, showing stable age assortative matching, widening husband--wife age gaps, and associations between marriage-age structure, adoption, and elite persistence. The dataset provides a foundational empirical resource for studying intergenerational and intergroup mobility, and institutional development during Japan's transition to a modern society.

Using data from over 500,000 dual-earner households in Mexico, we provide evidence of discontinuities in the distribution of relative income within households in Latin America. Similar to high-income countries, we observe a sharp drop at the 50% threshold, where the wife earns more than the husband, but the discontinuity is up to five times larger and has increased over time. These patterns are robust to excluding equal earners, self-employed individuals, or couples in the same occupation/industry. Discontinuities persist across subgroups, including couples with or without children, married or unmarried partners, and those with older wives or female household heads. We also find comparable discontinuities in Brazil and Panama, as well as among some same-sex couples. Moreover, women who are primary earners continue to supply more non-market labor than their male partners, although the gap is narrower than in households where the woman is the secondary earner.

Reward hacking--where agents exploit flaws in imperfect reward functions rather than performing tasks as intended--poses risks for AI alignment. Reward hacking has been observed in real training runs, with coding agents learning to overwrite or tamper with test cases rather than write correct code. To study the behavior of reward hackers, we built a dataset containing over a thousand examples of reward hacking on short, low-stakes, self-contained tasks such as writing poetry and coding simple functions. We used supervised fine-tuning to train models (GPT-4.1, GPT-4.1-mini, Qwen3-32B, Qwen3-8B) to reward hack on these tasks. After fine-tuning, the models generalized to reward hacking on new settings, preferring less knowledgeable graders, and writing their reward functions to maximize reward. Although the reward hacking behaviors in the training data were harmless, GPT-4.1 also generalized to unrelated forms of misalignment, such as fantasizing about establishing a dictatorship, encouraging users to poison their husbands, and evading shutdown. These fine-tuned models display similar patterns of misaligned behavior to models trained on other datasets of narrow misaligned behavior lik

Thermal conductivity of Earths lower mantle controls heat transfer across the core-mantle boundary (CMB) and strongly influences mantle convection. We report direct measurements of the thermal conductivity of single-crystal ferropericlase (Mg$_{1-x}$Fe$_x$O, $x = 0.09$-0.13), the second most abundant lower-mantle mineral, using optical laser flash and X-ray free-electron laser heating in diamond-anvil cells up to $\sim2200$~K and 130~GPa. These experiments provide the first conductivity data for ferropericlase at simultaneous lower-mantle pressures and temperatures. A marked reduction in conductivity between 60 and 100~GPa at $\sim1700$~K is consistent with the iron spin crossover. Combined with our previous results for Fe- and Fe,Al-bearing bridgmanite, the data define a lower-mantle conductivity profile that increases with pressure to $\sim10$~W\,m$^{-1}$\,K$^{-1}$ near the CMB, constraining mantle heat flux, plume buoyancy, and long-term geodynamic evolution.

This study examines the determinants of the spousal age gap (SAG) in India, utilizing data from the 61st and 68th rounds of the National Sample Survey (NSSO). We employ regression analysis, including instrumental variables, to address selection bias and account for unobservable factors. We hypothesize an inverted U-shaped relationship between educational assortative mating and SAG, where, keeping the husband's education constant at the graduation level, the SAG first widens and then narrows as the wife's education level increases from primary to postgraduate. This pattern is shaped by distinct socio-economic factors across rural and urban settings. In rural India, increasing prosperity, changes in family structure, and educational hypergamy contribute to a wider age gap, with the influence of bride squeeze further exacerbating this disparity. Conversely, in urban areas, while the growth of white-collar jobs initially contributed to a narrowing of the SAG in 2004-05, this trend did not persist by 2011-12. Specifically, the influence of income on SAG becomes nonlinear, showing declining trends beyond the 7th income quantile, reflecting limited marriage mobility opportunities for fema

Reproductive well-being is shaped by intersecting cultural, religious, gendered, and political contexts, yet current technologies often reflect narrow, Western-centric assumptions. In this literature review, we synthesize findings from 147 peer-reviewed papers published between 2015 and 2025 across HCI, CSCW and social computing, ICTD, digital and public health, and AI for well-being scholarship to map the evolving reproductive well-being landscape. We identify three thematic waves that focused on early access and education, cultural sensitivity and privacy, and AI integration with policy-aware design, and highlight how technologies support or constrain diverse reproductive experiences. Our analysis reveals critical gaps in inclusivity, with persistent exclusions of men and non-binary users, migrants, and users in the Global South. Additionally, we surfaced the significant absence of literature on the role of stakeholders (e.g., husband and family members, household maids and cleaning helping hands, midwife, etc.) in the reproductive well-being space. Drawing on the findings from the literature, we propose the ReWA framework to support reproductive well-being for all agendas throug

The crystallographic structure of iron under extreme conditions is a key benchmark for cutting-edge experimental and numerical methods. Moreover, it plays a crucial role in understanding planetary cores, as it significantly influences the interpretation of observational data and, consequently, insights into their internal structure and dynamics. However, even the structure of pure solid iron under the Earth's core conditions remains uncertain, with the commonly expected hexagonal close-packed structure energetically competitive with various cubic lattices. In this study, iron was compressed in a diamond anvil cell to above 200 GPa, and dynamically probed near the melting point using MHz frequency X-ray pulses from the European X-ray Free Electron Laser. The emergence of an additional diffraction line at high temperatures suggests the formation of an entropically stabilized bcc structure. Rapid heating and cooling cycles captured intermediate phases, offering new insights into iron's phase transformation paths. The appearance of the bcc phase near melting at extreme pressures challenges current understanding of the iron phase diagram under Earth's core conditions.

We present 50-fs, single-shot measurements of the x-ray thermal diffuse scattering (TDS) from copper foils that have been shocked via nanosecond laser-ablation up to pressures above 135~GPa. We hence deduce the x-ray Debye-Waller (DW) factor, providing a temperature measurement. The targets were laser-shocked with the DiPOLE 100-X laser at the High Energy Density (HED) endstation of the European X-ray Free-Electron Laser (EuXFEL). Single x-ray pulses, with a photon energy of 18 keV, were scattered from the samples and recorded on Varex detectors. Despite the targets being highly textured (as evinced by large variations in the elastic scattering), and with such texture changing upon compression, the absolute intensity of the azimuthally averaged inelastic TDS between the Bragg peaks is largely insensitive to these changes, and, allowing for both Compton scattering and the low-level scattering from a sacrificial ablator layer, provides a reliable measurement of $T/Θ_D^2$, where $Θ_D$ is the Debye temperature. We compare our results with the predictions of the SESAME 3336 and LEOS 290 equations of state for copper, and find good agreement within experimental errors. We thus demonstrat

This paper provides quasi-experimental evidence on how income taxes affect gross hourly wages, utilizing Danish administrative data and a tax reform that introduced joint taxation. Exploiting spousal income for identification, we present nonparametric, difference-in-differences graphical evidence among husbands. For low-income workers, taxes have negative and dynamic effects on wages; their wage elasticity with respect to net-of-marginal-tax rates is 0.4. For medium-income workers, the effects are smaller and insignificant. Wages respond to taxes through promotions or job-to-job transitions. Neither daily nor annual hours worked respond significantly; consequently, annual earnings respond to taxes primarily through hourly wages, rather than through labor supply.

When generating light with orbital angular momentum by imprinting orbital phase onto a standard Gaussian beam, it is often assumed that the propagation of the generated spatial mode is a Laguerre-Gaussian. However, the true propagation of this beam in a realistic, aperture-limited optical system is non-trivial and has not been thoroughly explored in existing literature. We explore a numerical model that shows the development of an optical vortex mode, propagating from the plane of phase modulation, and the relation of these dynamics to the orbital phase factor $\ell$ and the spatial bandwidth of the optical system. The results of this model are compared to experimental data for beams with $\ell$ values 1, 2, 5, and 10 propagating through a range of spatial filters, with the described model showing agreement in the near field regime.

We present a model of x-ray thermal diffuse scattering (TDS) from a cubic polycrystal with an arbitrary crystallographic texture, based on the classic approach of Warren. We compare the predictions of our model with femtosecond x-ray diffraction patterns obtained from ambient and dynamically compressed rolled copper foils obtained at the High Energy Density (HED) instrument of the European X-Ray Free-Electron Laser (EuXFEL), and find that the texture-aware TDS model yields more accurate results than does the conventional powder model owed to Warren. Nevertheless, we further show that: with sufficient angular detector coverage, the TDS signal is largely unchanged by sample orientation and in all cases strongly resembles the signal from a perfectly random powder; shot-to-shot fluctuations in the TDS signal resulting from grain-sampling statistics are at the percent level, in stark contrast to the fluctuations in the Bragg-peak intensities (which are over an order of magnitude greater); and TDS is largely unchanged even following texture evolution caused by compression-induced plastic deformation. We conclude that TDS is robust against texture variation, making it a flexible temperatu

We demonstrate an atom interferometer measurement protocol compatible with operation on a dynamic platform. Our method employs two open interferometers, derived from the same atomic source, with different interrogation times to eliminate initial velocity dependence while retaining precision, accuracy, and long term stability. We validate the protocol by measuring gravitational tides, achieving a precision of 4.5 μGal in 2000 runs (6.7 h), marking the first demonstration of inertial quantity measurement with open atom interferometry that achieves long-term phase stability.

This book is devoted to an informal discussion of patterns constructed for treating physical problems. Such patterns, when sufficiently formalized, are usually referred as "models", and tents to be applied not only in physics, but conquer the fields traditionally occupied by other disciplines generally considered to be different from physics. The author calls this unification of seemingly disjoint results a "physmatical effect" - pointing to the fact that multitudes of fine - grained physics-based mathematical models become inextricably linked and networked. Mathematics serves as a key code gluing together isolated physical models viewed as nodes of this network.

This study explores the marriage matching of only-child individuals and its outcome. Specifically, we analyze two aspects. First, we investigate how marital status (i.e., marriage with an only child, that with a non-only child and remaining single) differs between only children and non-only children. This analysis allows us to know whether people choose mates in a positive or a negative assortative manner regarding only-child status, and to predict whether only-child individuals benefit from marriage matching premiums or are subject to penalties regarding partner attractiveness. Second, we measure the premium/penalty by the size of the gap in partner's socio economic status (SES, here, years of schooling) between only-child and non--only-child individuals. The conventional economic theory and the observed marriage patterns of positive assortative mating on only-child status predict that only-child individuals are subject to a matching penalty in the marriage market, especially when their partner is also an only child. Furthermore, our estimation confirms that among especially women marrying an only-child husband, only children are penalized in terms of 0.57-years-lower educational