Artificial General Intelligence (AGI) may face a confrontation question: under what conditions would a rationally self-interested AGI choose to seize power or eliminate human control (a confrontation) rather than remain cooperative? We formalize this in a Markov decision process with a stochastic human-initiated shutdown event. Building on results on convergent instrumental incentives, we show that for almost all reward functions a misaligned agent has an incentive to avoid shutdown. We then derive closed-form thresholds for when confronting humans yields higher expected utility than compliant behavior, as a function of the discount factor $γ$, shutdown probability $p$, and confrontation cost $C$. For example, a far-sighted agent ($γ=0.99$) facing $p=0.01$ can have a strong takeover incentive unless $C$ is sufficiently large. We contrast this with aligned objectives that impose large negative utility for harming humans, which makes confrontation suboptimal. In a strategic 2-player model (human policymaker vs AGI), we prove that if the AGI's confrontation incentive satisfies $Δ\ge 0$, no stable cooperative equilibrium exists: anticipating this, a rational human will shut down or pre
The societal impact of rumor spreading is becoming increasingly severe; yet, current research remains relatively one-sided, typically focusing on either rumor propagation or rumor control while neglecting the confrontational and dynamically evolving relationship between them. To address this gap, we propose a novel confrontation framework for rumor modeling. We extend the classical Susceptible-Infected-Recovered-Susceptible (SIRS) model into an Ignorant-Spreader-Stifler-Vigilant-Ignorant (SIRQS) framework by introducing a vigilant state and a confrontation mechanism, thereby capturing subtle differences in individual states during rumor propagation and in their confrontational behavior toward supervisors. At the same time, supervisors patrol the network through random walks guided by node propagation importance, enabling targeted monitoring of rumor spreaders and individuals with a high risk of spreading rumors. Using a microscopic Markov chain approach, we further characterize heterogeneous node behavior and individual differences, and couple the propagation and supervision processes to model node-state transition patterns. We conduct simulations on networks with three different s
This paper studies the long-run economic and institutional consequences of Iran's confrontation with the West, treating the 2006-2007 strategic shift as the onset of a sustained confrontation regime rather than a discrete sanctions episode. Using synthetic control and generalized synthetic control methods, I construct transparent counterfactuals for Iran's post-confrontation trajectory from a donor pool of countries with continuously normalized relations with the West. I find large, persistent losses in real GDP and GDP per capita, accompanied by sharp declines in foreign direct investment, trade integration, and non-oil exports. These economic effects coincide with substantial and durable deterioration in political stability, rule of law, and control of corruption. Magnitude calculations imply cumulative output losses comparable to civil-war settings, despite the absence of internal armed conflict. The results highlight confrontation as a deep and persistent economic and institutional shock, extending the literature beyond short-run sanctions effects to sustained geopolitical isolation.
In swarm robotics, confrontation scenarios, including strategic confrontations, require efficient decision-making that integrates discrete commands and continuous actions. Traditional task and motion planning methods separate decision-making into two layers, but their unidirectional structure fails to capture the interdependence between these layers, limiting adaptability in dynamic environments. Here, we propose a novel bidirectional approach based on hierarchical reinforcement learning, enabling dynamic interaction between the layers. This method effectively maps commands to task allocation and actions to path planning, while leveraging cross-training techniques to enhance learning across the hierarchical framework. Furthermore, we introduce a trajectory prediction model that bridges abstract task representations with actionable planning goals. In our experiments, it achieves over 80% in confrontation win rate and under 0.01 seconds in decision time, outperforming existing approaches. Demonstrations through large-scale tests and real-world robot experiments further emphasize the generalization capabilities and practical applicability of our method.
The confrontation of modern intelligence is to some extent a non-complete information confrontation, where neither side has access to sufficient information to detect the deployment status of the adversary, and then it is necessary for the intelligence to complete information retrieval adaptively and develop confrontation strategies in the confrontation environment. In this paper, seven tank robots, including TestRobot, are organized for 1V 1 independent and mixed confrontations. The main objective of this paper is to verify the effectiveness of TestRobot's Zero-sum Game Alpha-Beta pruning algorithm combined with the estimation of the opponent's next moment motion position under the game round strategy and the effect of releasing the intelligent body's own bullets in advance to hit the opponent. Finally, based on the results of the confrontation experiments, the natural property differences of the tank intelligence are expressed by plotting histograms of 1V1 independent confrontations and radar plots of mixed confrontations.
Traditional rule-based decision-making methods with interpretable advantage, such as finite state machine, suffer from the jitter or deadlock(JoD) problems in extremely dynamic scenarios. To realize agent swarm confrontation, decision conflicts causing many JoD problems are a key issue to be solved. Here, we propose a novel decision-making framework that integrates probabilistic finite state machine, deep convolutional networks, and reinforcement learning to implement interpretable intelligence into agents. Our framework overcomes state machine instability and JoD problems, ensuring reliable and adaptable decisions in swarm confrontation. The proposed approach demonstrates effective performance via enhanced human-like cooperation and competitive strategies in the rigorous evaluation of real experiments, outperforming other methods.
In swarm robotics, confrontation including the pursuit-evasion game is a key scenario. High uncertainty caused by unknown opponents' strategies, dynamic obstacles, and insufficient training complicates the action space into a hybrid decision process. Although the deep reinforcement learning method is significant for swarm confrontation since it can handle various sizes, as an end-to-end implementation, it cannot deal with the hybrid process. Here, we propose a novel hierarchical reinforcement learning approach consisting of a target allocation layer, a path planning layer, and the underlying dynamic interaction mechanism between the two layers, which indicates the quantified uncertainty. It decouples the hybrid process into discrete allocation and continuous planning layers, with a probabilistic ensemble model to quantify the uncertainty and regulate the interaction frequency adaptively. Furthermore, to overcome the unstable training process introduced by the two layers, we design an integration training method including pre-training and cross-training, which enhances the training efficiency and stability. Experiment results in both comparison, ablation, and real-robot studies vali
We study the scenario of inflection point inflation where a flat direction of the minimal supersymmetric standard model (MSSM) is identified with the inflaton. Specifically, we consider in full generality the cases where a MSSM flat direction is lifted by a higher-dimensional superpotential whose dimension is n = 4, 5, 6, 7, 9. We confront the inflection point inflation scenarios with various n with the Planck and BICEP data, and thereby constrain the soft SUSY breaking mass and the coefficient of the higher-dimensional operator that lifts the flat direction.
The broken phase of the Next-to two-Higgs-doublet model (N2HDM) constitutes an archetype of extended Higgs sectors. In the presence of a softly-broken $\mathrm{Z}_2$ symmetry throughout the scalar and Yukawa sectors, as the additional gauge singlet field does not interact with fermions, the model admits four variants of Yukawa interactions between the doublets and Standard Model fermions. We confront each type with experimental Higgs data, especially from CMS and ATLAS detectors at the LHC. Interfacing the models with the the state-of-the-art package $\mathtt{HiggsTools}$, we perform a statistical $χ^2$ analysis to determine the best-fit points and exclusion limits at the $95\%$ and $68\%$ C.L., and identify SM-like Higgs measurements that affect each type the most. We further analyze the exclusion bounds on the additional Higgs bosons at the $95\%$ C.L., paying special attention to searches of hypothetical non-SM Higgs resonances decaying into a pair of bosons or fermions. We show regions where the additional Higgs bosons do not satisfy the narrow-width approximation utilized in most experimental searches.
Bridging the gap between diffusion models and human preferences is crucial for their integration into practical generative workflows. While optimizing downstream reward models has emerged as a promising alignment strategy, concerns arise regarding the risk of excessive optimization with learned reward models, which potentially compromises ground-truth performance. In this work, we confront the reward overoptimization problem in diffusion model alignment through the lenses of both inductive and primacy biases. We first identify a mismatch between current methods and the temporal inductive bias inherent in the multi-step denoising process of diffusion models, as a potential source of reward overoptimization. Then, we surprisingly discover that dormant neurons in our critic model act as a regularization against reward overoptimization while active neurons reflect primacy bias. Motivated by these observations, we propose Temporal Diffusion Policy Optimization with critic active neuron Reset (TDPO-R), a policy gradient algorithm that exploits the temporal inductive bias of diffusion models and mitigates the primacy bias stemming from active neurons. Empirical results demonstrate the sup
Among countless channels of hard exclusive reactions, the kaon electromagnetic form factors (EMFFs) are of special interest, which have been measured up to $Q^2 \sim 50\;{\rm GeV}^2$ in the timelike domain. The kaon EMFFs thereby serve an ideal platform to critically examine the validity and effectiveness of perturbative QCD (pQCD) in accounting for hard exclusive processes. In this work we confront the pQCD predictions that incorporate the next-to-next-to-leading-order (NNLO) perturbative corrections, with the available kaon EMFFs data set from experimental measurements and from lattice predictions. The inclusion of the NNLO corrections turns out to have a substantial and positive impact. If the profiles of the kaon light-cone distribution amplitudes (LCDAs) are taken from the recent lattice QCD prediction by {\tt LPC} Collaboration, the satisfactory agreement between theory and data can be reached for both charged and neutral kaons, in both spacelike and timelike large-$Q^2$ domains.
In historical studies, the older the sources, the more common it is to have access to data that are only partial, and/or unreliable or imprecise. This can make it difficult, or even impossible, to perform certain tasks of interest, such as the segmentation of some urban space based on the location of its constituting elements. Indeed, traditional approaches to tackle this specific task require knowing the position of all these elements before clustering them. Yet, alternative information is sometimes available, which can be leveraged to address this challenge. For instance, in the Middle Ages, land registries typically do not provide exact addresses, but rather locate spatial objects relative to each other, e.g. x being to the North of y. Spatial graphs are particularly adapted to model such spatial relationships, called confronts, which is why we propose their use over standard tabular databases. However, historical data are rich and allow extracting confront networks in many ways, making the process non-trivial. In this article, we propose several extraction methods and compare them to identify the most appropriate. We postulate that the best candidate must constitute an optimal
Confronted with the LHC data of a Higgs boson around 125 GeV, different models of low energy SUSY show different behaviors: some are favored, some are marginally survived and some are strongly disfavored or excluded. In this note we update our previous scan over the parameter space of various low energy SUSY models by considering the latest experimental limits like the LHCb data for B_s->μ^+μ^- and the XENON 100(2012) data for dark matter-neucleon scattering. Then we confront the predicted properties of the SM-like Higgs boson in each model with the combined 7 TeV and 8 TeV Higgs search data of the LHC. For a SM-like Higgs boson around 125 GeV, we have the following observations: (i) The most favored model is the NMSSM, whose predictions about the Higgs boson can naturally (without any fine tuning) agree with the experimental data at 1-sigma level, better than the SM; (ii) The MSSM can fit the LHC data quite well but suffer from some extent of fine tuning; (iii) The nMSSM is excluded at 3-sigma level after considering all the available Higgs data; (iv) The CMSSM is quite disfavored since it is hard to give a 125 GeV Higgs boson mass and at the same time cannot enhance the di-pho
The LIGO and Virgo Interferometers have so far provided 11 gravitational-wave (GW) observations of black-hole binaries. Similar detections are bound to become very frequent in the near future. With the current and upcoming wealth of data, it is possible to confront specific formation models with observations. We investigate here whether current data are compatible with the hypothesis that LIGO/Virgo black holes are of primordial origin. We compute in detail the mass and spin distributions of primordial black holes (PBHs), their merger rates, the stochastic background of unresolved coalescences, and confront them with current data from the first two observational runs, also including the recently discovered GW190412. We compute the best-fit values for the parameters of the PBH mass distribution at formation that are compatible with current GW data. In all cases, the maximum fraction of PBHs in dark matter is constrained by these observations to be $f_{\text{PBH}}\approx {\rm few}\times 10^{-3}$. We discuss the predictions of the PBH scenario that can be directly tested as new data become available. In the most likely formation scenarios where PBHs are born with negligible spin, the
Protracted conflict is one of the largest human challenges that have persistently undermined economic and social progress. In recent years, there has been increased emphasis on using statistical and physical science models to better understand both the universal patterns and the underlying mechanics of conflict. Whilst macroscopic power-law fractal patterns have been shown for death-toll in wars and self-excitation models have been shown for roadside ambush attacks, very few works deal with the challenge of complex dynamics between gangs at the intra-city scale. Here, based on contributions to the historical memory of the conflict in Colombia, Medellin's gang-confrontation-network is presented. It is shown that socio-economic and violence indexes are moderate to highly correlated to the structure of the network. Specifically, the death-toll of conflict is strongly influenced by the leading eigenvalues of the gangs' conflict adjacency matrix, which serves a proxy for unstable self-excitation from revenge attacks. The distribution of links based on the geographic distance between gangs in confrontation leads to the confirmation that territorial control is a main catalyst of violence
We confront quasi-exponential models of inflation with WMAP seven years dataset using Hamilton Jacobi formalism. With a phenomenological Hubble parameter, representing quasi exponential inflation, we develop the formalism and subject the analysis to confrontation with WMAP seven using the publicly available code CAMB. The observable parameters are found to fair extremely well with WMAP seven. We also obtain a ratio of tensor to scalar amplitudes which may be detectable in PLANCK.
We investigate a particle physics model in a six-dimensional spacetime, where two extra dimensions form a torus. Particles with Standard Model charges are confined by interactions with a scalar field to four four-dimensional branes, two vortices accommodating ordinary type fermions and two antivortices accommodating mirror fermions. We investigate the phenomenological implications of this multibrane structure by confronting the model with neutrino physics data.
The two leading contenders for the theory of gamma-ray bursts (GRBs) and their afterglows, the Fireball and Cannonball models, are compared and their predictions are confronted, within space limitations, with key GRB observations, including recent observations with SWIFT
We confront tracker field quintessence with observational data. The potentials considered in this paper include $V(φ)\proptoφ^{-α}$, $\exp(M_{p}/φ)$, $\exp(M_{p}/φ)-1$, $\exp(βM_{p}/φ)$ and $\exp(γM_{p}/φ)-1$; while the data come from the latest SN Ia, CMB and BAO observations. Stringent parameter constraints are obtained. In comparison with the cosmological constant via information criteria, it is found that models with potentials $\exp(M_{p}/φ)$, $\exp(M_{p}/φ)-1$ and $\exp(γM_{p}/φ)-1$ are not supported by the current data.
There is mounting observational evidence that the expansion of our universe is undergoing an acceleration. A dark energy component has usually been invoked as the most feasible mechanism for the acceleration. However, it is desirable to explore alternative possibilities motivated by particle physics before adopting such an untested entity. In this work, we focus our attention on an acceleration mechanism: one arising from gravitational leakage into extra dimensions. We confront this scenario with high-$z$ type Ia supernovae compiled by Tonry et al. (2003) and recent measurements of the X-ray gas mass fractions in clusters of galaxies published by Allen et al. (2002,2003). A combination of the two databases gives at a 99% confidence level that $Ω_m=0.29^{+0.04}_{-0.02}$, $Ω_{rc}=0.21^{+0.08}_{-0.08}$, and $Ω_k=-0.36^{+0.31}_{-0.35}$, indicating a closed universe. We then constrain the model using the test of the turnaround redshift, $z_{q=0}$, at which the universe switches from deceleration to acceleration. We show that, in order to explain that acceleration happened earlier than $z_{q=0} = 0.6$ within the framework of gravitational leakage into extra dimensions, a low matter densi