Search

We present a novel-view rendering algorithm, Mode-GS, for ground-robot trajectory datasets. Our approach is based on using anchored Gaussian splats, which are designed to overcome the limitations of existing 3D Gaussian splatting algorithms. Prior neural rendering methods suffer from severe splat drift due to scene complexity and insufficient multi-view observation, and can fail to fix splats on the true geometry in ground-robot datasets. Our method integrates pixel-aligned anchors from monocular depths and generates Gaussian splats around these anchors using residual-form Gaussian decoders. To address the inherent scale ambiguity of monocular depth, we parameterize anchors with per-view depth-scales and employ scale-consistent depth loss for online scale calibration. Our method results in improved rendering performance, based on PSNR, SSIM, and LPIPS metrics, in ground scenes with free trajectory patterns, and achieves state-of-the-art rendering performance on the R3LIVE odometry dataset and the Tanks and Temples dataset.

Designing biological sequences with desired properties is a significant challenge due to the combinatorially vast search space and the high cost of evaluating each candidate sequence. To address these challenges, reinforcement learning (RL) methods, such as GFlowNets, utilize proxy models for rapid reward evaluation and annotated data for policy training. Although these approaches have shown promise in generating diverse and novel sequences, the limited training data relative to the vast search space often leads to the misspecification of proxy for out-of-distribution inputs. We introduce $\delta$-Conservative Search, a novel off-policy search method for training GFlowNets designed to improve robustness against proxy misspecification. The key idea is to incorporate conservativeness, controlled by parameter $\delta$, to constrain the search to reliable regions. Specifically, we inject noise into high-score offline sequences by randomly masking tokens with a Bernoulli distribution of parameter $\delta$ and then denoise masked tokens using the GFlowNet policy. Additionally, $\delta$ is adaptively adjusted based on the uncertainty of the proxy model for each data point. This enables the reflection of proxy uncertainty to determine the level of conservativeness. Experimental results demonstrate that our method consistently outperforms existing machine learning methods in discovering high-score sequences across diverse tasks-including DNA, RNA, protein, and peptide design-especially in large-scale scenarios., Comment: 11 pages

We say that a digraph $D$ is $(i,j)$-step competitive if any two vertices have an $(i,j)$-step common out-neighbor in $D$ and that a graph $G$ is $(i,j)$-step competitively orientable if there exists an $(i,j)$-step competitive orientation of $G$. In [Choi et al. Competitively orientable complete multipartite graphs. Discrete Mathematics, 345(9):112950, 2022], Choi et al. introduce the notion of competitive digraph and completely characterize competitively orientable complete multipartite graphs in terms of the sizes of its partite sets. Here, a competitive digraph means a $(1,1)$-step competitive digraph. In this paper, the result of Choi et al. has been extended to a general characterization of $(i,j)$-step competitively orientable complete multipartite graphs.

Creativity has long been considered one of the most difficult aspect of human intelligence for AI to mimic. However, the rise of Large Language Models (LLMs), like ChatGPT, has raised questions about whether AI can match or even surpass human creativity. We present CREATIVITY INDEX as the first step to quantify the linguistic creativity of a text by reconstructing it from existing text snippets on the web. CREATIVITY INDEX is motivated by the hypothesis that the seemingly remarkable creativity of LLMs may be attributable in large part to the creativity of human-written texts on the web. To compute CREATIVITY INDEX efficiently, we introduce DJ SEARCH, a novel dynamic programming algorithm that can search verbatim and near-verbatim matches of text snippets from a given document against the web. Experiments reveal that the CREATIVITY INDEX of professional human authors is on average 66.2% higher than that of LLMs, and that alignment reduces the CREATIVITY INDEX of LLMs by an average of 30.1%. In addition, we find that distinguished authors like Hemingway exhibit measurably higher CREATIVITY INDEX compared to other human writers. Finally, we demonstrate that CREATIVITY INDEX can be used as a surprisingly effective criterion for zero-shot machine text detection, surpassing the strongest existing zero-shot system, DetectGPT, by a significant margin of 30.2%, and even outperforming the strongest supervised system, GhostBuster, in five out of six domains.

The isotopic compositions of samples returned from Cb-type asteroid Ryugu and Ivuna-type (CI) chondrites are distinct from other carbonaceous chondrites, which has led to the suggestion that Ryugu and CI chondrites formed in a different region of the accretion disk, possibly around the orbits of Uranus and Neptune. We show that, like for Fe, Ryugu and CI chondrites also have indistinguishable Ni isotope anomalies, which differ from those of other carbonaceous chondrites. We propose that this unique Fe and Ni isotopic composition reflects different accretion efficiencies of small FeNi metal grains among the carbonaceous chondrite parent bodies. The CI chondrites incorporated these grains more efficiently, possibly because they formed at the end of the disk's lifetime, when planetesimal formation was also triggered by photoevaporation of the disk. Isotopic variations among carbonaceous chondrites may thus reflect fractionation of distinct dust components from a common reservoir, implying CI chondrites and Ryugu may have formed in the same region of the accretion disk as other carbonaceous chondrites., Comment: Published open access in Science Advances

Recent calls for pluralistic alignment emphasize that AI systems should address the diverse needs of all people. Yet, efforts in this space often require sorting people into fixed buckets of pre-specified diversity-defining dimensions (e.g., demographics, personalities, communication styles), risking smoothing out or even stereotyping the rich spectrum of individualistic variations. To achieve an authentic representation of diversity that respects individuality, we propose individualistic alignment. While individualistic alignment can take various forms, in this paper, we introduce IndieValueCatalog, a dataset transformed from the influential World Values Survey (WVS), to study language models (LMs) on the specific challenge of individualistic value reasoning. Specifically, given a sample of an individual's value-expressing statements, models are tasked with predicting their value judgments in novel cases. With IndieValueCatalog, we reveal critical limitations in frontier LMs' abilities to reason about individualistic human values with accuracies, only ranging between 55% to 65%. Moreover, our results highlight that a precise description of individualistic values cannot be approximated only via demographic information. We also identify a partiality of LMs in reasoning about global individualistic values, as measured by our proposed Value Inequity Index ({\sigma}INEQUITY). Finally, we train a series of Individualistic Value Reasoners (IndieValueReasoner) using IndieValueCatalog to enhance models' individualistic value reasoning capability, revealing new patterns and dynamics into global human values. We outline future research challenges and opportunities for advancing individualistic alignment.

The topological phase transition between two band insulators is mediated by a gapless state whose low-energy band structure normally contains sufficient information for describing the topology change. In this work, we show that there is a class of topological insulators whose topological phase transition cannot be explained by this conventional paradigm. Taking the Hopf insulator as a representative example, we show that the change of the Hopf invariant requires the information of wave functions as well as the gapless band structure simultaneously. More explicitly, the description of the Hopf invariant change requires us to trace not only the trajectory of Weyl points but also the evolution of the preimages for two distinct eigenstates. We show that such an unconventional topological phase transition originates from the fact that the Hopf invariant is well-defined when all the lower dimensional topological invariants are trivial, which in turn allows us to lift the classifying space of occupied state projectors to the corresponding universal 2-covering space of wave functions. Generalizing our theory to inversion-symmetric 10-fold Altland-Zirnbauer symmetry classes, we provide a complete list of symmetry classes, all of which turn out to have delicate band topology related to the Hopf invariant, where similar unconventional topological phase transitions can appear., Comment: Main text: 5 pages, 3 figures, Supplemental materials: 8 pages, 2 figures

Text-to-speech (TTS) systems that scale up the amount of training data have achieved significant improvements in zero-shot speech synthesis. However, these systems have certain limitations: they require a large amount of training data, which increases costs, and often overlook prosody similarity. To address these issues, we propose MultiVerse, a zero-shot multi-task TTS system that is able to perform TTS or speech style transfer in zero-shot and cross-lingual conditions. MultiVerse requires much less training data than traditional data-driven approaches. To ensure zero-shot performance even with limited data, we leverage source-filter theory-based disentanglement, utilizing the prompt for modeling filter-related and source-related representations. Additionally, to further enhance prosody similarity, we adopt a prosody modeling approach combining prompt-based autoregressive and non-autoregressive methods. Evaluations demonstrate the remarkable zero-shot multi-task TTS performance of MultiVerse and show that MultiVerse not only achieves zero-shot TTS performance comparable to data-driven TTS systems with much less data, but also significantly outperforms other zero-shot TTS systems trained with the same small amount of data. In particular, our novel prosody modeling technique significantly contributes to MultiVerse's ability to generate speech with high prosody similarity to the given prompts. Our samples are available at https://nc-ai.github.io/speech/publications/multiverse/index.html, Comment: Accepted to EMNLP 2024 Findings

Brain connectivity analysis is crucial for understanding brain structure and neurological function, shedding light on the mechanisms of mental illness. To study the association between individual brain connectivity networks and the clinical characteristics, we develop BSNMani: a Bayesian scalar-on-network regression model with manifold learning. BSNMani comprises two components: the network manifold learning model for brain connectivity networks, which extracts shared connectivity structures and subject-specific network features, and the joint predictive model for clinical outcomes, which studies the association between clinical phenotypes and subject-specific network features while adjusting for potential confounding covariates. For posterior computation, we develop a novel two-stage hybrid algorithm combining Metropolis-Adjusted Langevin Algorithm (MALA) and Gibbs sampling. Our method is not only able to extract meaningful subnetwork features that reveal shared connectivity patterns, but can also reveal their association with clinical phenotypes, further enabling clinical outcome prediction. We demonstrate our method through simulations and through its application to real resting-state fMRI data from a study focusing on Major Depressive Disorder (MDD). Our approach sheds light on the intricate interplay between brain connectivity and clinical features, offering insights that can contribute to our understanding of psychiatric and neurological disorders, as well as mental health.

As we increasingly seek guidance from LLMs for decision-making in daily life, many of these decisions are not clear-cut and depend significantly on the personal values and ethical standards of the users. We present DailyDilemmas, a dataset of 1,360 moral dilemmas encountered in everyday life. Each dilemma includes two possible actions and with each action, the affected parties and human values invoked. Based on these dilemmas, we consolidated a set of human values across everyday topics e.g., interpersonal relationships, workplace, and environmental issues. We evaluated LLMs on these dilemmas to determine what action they will take and the values represented by these actions. Then, we analyzed these values through the lens of five popular theories inspired by sociology, psychology and philosophy. These theories are: World Value Survey, Moral Foundation Theory, Maslow's Hierarchy of Needs, Aristotle's Virtues, and Plutchik Wheel of Emotion. We find that LLMs are most aligned with the self-expression over survival values in terms of World Value Survey, care over loyalty in Moral Foundation Theory. Interestingly, we find large preferences differences in models for some core values such as truthfulness e.g., Mixtral-8x7B model tends to neglect it by 9.7% while GPT-4-turbo model tends to select it by 9.4%. We also study the recent guidance released by OpenAI (ModelSpec), and Anthropic (Constitutional AI) to understand how their released principles reflect their actual value prioritization when facing nuanced moral reasoning in daily-life settings. We find that end users cannot effectively steer such prioritization using system prompts., Comment: Preprint. Under Review

To make large language models (LLMs) more helpful across diverse cultures, it is essential to have effective cultural knowledge benchmarks to measure and track our progress. Effective benchmarks need to be robust, diverse, and challenging. We introduce CulturalBench: a set of 1,227 human-written and human-verified questions for effectively assessing LLMs' cultural knowledge, covering 45 global regions including the underrepresented ones like Bangladesh, Zimbabwe, and Peru. Questions - each verified by five independent annotators - span 17 diverse topics ranging from food preferences to greeting etiquettes. We evaluate models on two setups: CulturalBench-Easy and CulturalBench-Hard which share the same questions but asked differently. We find that LLMs are sensitive to such difference in setups (e.g., GPT-4o with 27.3% difference). Compared to human performance (92.6% accuracy), CulturalBench-Hard is more challenging for frontier LLMs with the best performing model (GPT-4o) at only 61.5% and the worst (Llama3-8b) at 21.4%. Moreover, we find that LLMs often struggle with tricky questions that have multiple correct answers (e.g., What utensils do the Chinese usually use?), revealing a tendency to converge to a single answer. Our results also indicate that OpenAI GPT-4o substantially outperform other proprietary and open source models in questions related to all but one region (Oceania). Nonetheless, all models consistently underperform on questions related to South America and the Middle East., Comment: Preprint. Under review

Based on 7.33 fb$^{-1}$ of $e^+e^-$ collision data collected by the BESIII detector operating at the BEPCII collider at center-of-mass energies from 4.128 to 4.226 GeV, a search for the Majorana neutrino $\nu_m$ is conducted in the lepton-number-violating decays of $D_s^+\to h^-h^0e^+e^+$. Here, $h^-$ represents a $K^-$ or $\pi^-$, and $h^0$ represents a $\pi^0$, $K_S^0$ or $\phi$. No significant signal is observed, and the upper limits of their branching fractions at the 90\% confidence level are determined to be $\mathcal{B}(D_s^+\to \phi\pi^-e^+e^+) < 6.9 \times 10^{-5}$, $\mathcal{B}(D_s^+\to \phi K^-e^+e^+) < 9.9 \times 10^{-5}$, $\mathcal{B}(D_s^+\to K_S^0\pi^-e^+e^+) < 1.3 \times 10^{-5}$, $\mathcal{B}(D_s^+\to K_S^0K^-e^+e^+) < 2.9 \times 10^{-5}$, $\mathcal{B}(D_s^+\to \pi^-\pi^0e^+e^+) < 2.9 \times 10^{-5}$ and $\mathcal{B}(D_s^+\to K^-\pi^0e^+e^+) < 3.4 \times 10^{-5}$. The Majorana neutrino is searched for with different mass assumptions within the range [0.20, 0.80] GeV$/c^2$ in the decay of $D_s^+\to\phi e^+\nu_m$ with $\nu_m\to\pi^-e^+$, and the upper limits of the branching fractions at the 90\% confidence level are at the level of $10^{-5}-10^{-2}$, depending on the mass of the Majorana neutrino.

As a neural network's depth increases, it can achieve strong generalization performance. Training, however, becomes challenging due to gradient issues. Theoretical research and various methods have been introduced to address this issues. However, research on weight initialization methods that can be effectively applied to tanh neural networks of varying sizes still needs to be completed. This paper presents a novel weight initialization method for Feedforward Neural Networks with tanh activation function. Based on an analysis of the fixed points of the function $\tanh(ax)$, our proposed method aims to determine values of $a$ that prevent the saturation of activations. A series of experiments on various classification datasets demonstrate that the proposed method is more robust to network size variations than the existing method. Furthermore, when applied to Physics-Informed Neural Networks, the method exhibits faster convergence and robustness to variations of the network size compared to Xavier initialization in problems of Partial Differential Equations.

Handling long input contexts remains a significant challenge for Large Language Models (LLMs), particularly in resource-constrained environments such as mobile devices. Our work aims to address this limitation by introducing InfiniPot, a novel KV cache control framework designed to enable pre-trained LLMs to manage extensive sequences within fixed memory constraints efficiently, without requiring additional training. InfiniPot leverages Continual Context Distillation (CCD), an iterative process that compresses and retains essential information through novel importance metrics, effectively maintaining critical data even without access to future context. Our comprehensive evaluations indicate that InfiniPot significantly outperforms models trained for long contexts in various NLP tasks, establishing its efficacy and versatility. This work represents a substantial advancement toward making LLMs applicable to a broader range of real-world scenarios., Comment: EMNLP 2024 Main

Amortized inference is the task of training a parametric model, such as a neural network, to approximate a distribution with a given unnormalized density where exact sampling is intractable. When sampling is implemented as a sequential decision-making process, reinforcement learning (RL) methods, such as generative flow networks, can be used to train the sampling policy. Off-policy RL training facilitates the discovery of diverse, high-reward candidates, but existing methods still face challenges in efficient exploration. We propose to use an adaptive training distribution (the Teacher) to guide the training of the primary amortized sampler (the Student) by prioritizing high-loss regions. The Teacher, an auxiliary behavior model, is trained to sample high-error regions of the Student and can generalize across unexplored modes, thereby enhancing mode coverage by providing an efficient training curriculum. We validate the effectiveness of this approach in a synthetic environment designed to present an exploration challenge, two diffusion-based sampling tasks, and four biochemical discovery tasks demonstrating its ability to improve sample efficiency and mode coverage., Comment: 26 pages, 12 figures

Real-life robot navigation involves more than just reaching a destination; it requires optimizing movements while addressing scenario-specific goals. An intuitive way for humans to express these goals is through abstract cues like verbal commands or rough sketches. Such human guidance may lack details or be noisy. Nonetheless, we expect robots to navigate as intended. For robots to interpret and execute these abstract instructions in line with human expectations, they must share a common understanding of basic navigation concepts with humans. To this end, we introduce CANVAS, a novel framework that combines visual and linguistic instructions for commonsense-aware navigation. Its success is driven by imitation learning, enabling the robot to learn from human navigation behavior. We present COMMAND, a comprehensive dataset with human-annotated navigation results, spanning over 48 hours and 219 km, designed to train commonsense-aware navigation systems in simulated environments. Our experiments show that CANVAS outperforms the strong rule-based system ROS NavStack across all environments, demonstrating superior performance with noisy instructions. Notably, in the orchard environment, where ROS NavStack records a 0% total success rate, CANVAS achieves a total success rate of 67%. CANVAS also closely aligns with human demonstrations and commonsense constraints, even in unseen environments. Furthermore, real-world deployment of CANVAS showcases impressive Sim2Real transfer with a total success rate of 69%, highlighting the potential of learning from human demonstrations in simulated environments for real-world applications., Comment: project page https://worv-ai.github.io/canvas

The Multi-Object Search (MOS) problem involves navigating to a sequence of locations to maximize the likelihood of finding target objects while minimizing travel costs. In this paper, we introduce a novel approach to the MOS problem, called Finder, which leverages vision language models (VLMs) to locate multiple objects across diverse environments. Specifically, our approach introduces multi-channel score maps to track and reason about multiple objects simultaneously during navigation, along with a score fusion technique that combines scene-level and object-level semantic correlations. Experiments in both simulated and real-world settings showed that Finder outperforms existing methods using deep reinforcement learning and VLMs. Ablation and scalability studies further validated our design choices and robustness with increasing numbers of target objects, respectively. Website: https://find-all-my-things.github.io/, Comment: 12 pages, 6 figures, submitted to ICRA2025

A graph $G$ is $k$-critical (list $k$-critical, DP $k$-critical) if $\chi(G)= k$ ($\chi_\ell(G)= k$, $\chi_\mathrm{DP}(G)= k$) and for every proper subgraph $G'$ of $G$, $\chi(G')\left(3 + \frac{1}{5} \right) \frac{n}{2}. $$ This is the first bound on $f_\mathrm{DP}(n,4)$ that is asymptotically better than the well-known bound $f(n,4)\geq \left(3 + \frac{1}{13} \right) \frac{n}{2}$ by Gallai from 1963. The result also yields a better bound on $f_{\ell}(n,4)$ than the one known before., Comment: 23 pages. arXiv admin note: substantial text overlap with arXiv:2409.00937

We present the discovery of Aquarius III, an ultra-faint Milky Way satellite galaxy identified in the second data release of the DECam Local Volume Exploration (DELVE) survey. Based on deeper follow-up imaging with DECam, we find that Aquarius III is a low-luminosity ($M_V = -2.5^{+0.3}_{-0.5}$; $L_V = 850^{+380}_{-260} \ L_{\odot}$), extended ($r_{1/2} = 41^{+9}_{-8}$ pc) stellar system located in the outer halo ($D_{\odot} = 85 \pm 4$ kpc). From medium-resolution Keck/DEIMOS spectroscopy, we identify 11 member stars and measure a mean heliocentric radial velocity of $v_{\rm sys} = -13.1^{+1.0}_{-0.9} \ \rm km \ s^{-1}$ for the system and place an upper limit of $\sigma_v < 3.5 \rm \ km \ s^{-1}$ ($\sigma_v < 1.6 \rm \ km \ s^{-1}$) on its velocity dispersion at the 95% (68%) credible level. Based on Calcium-Triplet-based metallicities of the six brightest red giant members, we find that Aquarius III is very metal-poor ([Fe/H]$ = -2.61 \pm 0.21$) with a statistically-significant metallicity spread ($\sigma_{\rm [Fe/H]} = 0.46^{+0.26}_{-0.14}$ dex). We interpret this metallicity spread as strong evidence that the system is a dwarf galaxy as opposed to a star cluster. Combining our velocity measurement with $Gaia$ proper motions, we find that Aquarius III is currently situated near its orbital pericenter in the outer halo ($r_{\rm peri} = 78 \pm 7$ kpc) and that it is plausibly on first infall onto the Milky Way. This orbital history likely precludes significant tidal disruption from the Galactic disk, notably unlike other satellites with comparably low velocity dispersion limits in the literature. Thus, if further velocity measurements confirm that its velocity dispersion is truly below $\sigma_v \lesssim 2 \rm \ km \ s^{-1}$, Aquarius III may serve as a useful laboratory for probing galaxy formation physics in low-mass halos., Comment: 25 pages, 7 figures. Submitted to AAS Journals

Text-to-texture generation has recently attracted increasing attention, but existing methods often suffer from the problems of view inconsistencies, apparent seams, and misalignment between textures and the underlying mesh. In this paper, we propose a robust text-to-texture method for generating consistent and seamless textures that are well aligned with the mesh. Our method leverages state-of-the-art 2D diffusion models, including SDXL and multiple ControlNets, to capture structural features and intricate details in the generated textures. The method also employs a symmetrical view synthesis strategy combined with regional prompts for enhancing view consistency. Additionally, it introduces novel texture blending and soft-inpainting techniques, which significantly reduce the seam regions. Extensive experiments demonstrate that our method outperforms existing state-of-the-art methods., Comment: 11 pages, 13 figures

Electro-optic active metasurfaces have attracted attention due to their ability to electronically control optical wavefront with unprecedented spatiotemporal resolutions. In most studies, such devices require gate arrays composed of a large number of independently-controllable local gate electrodes that address local scattering response of individual metaatoms. Although this approach in principle enables arbitrary wavefront control, the complicated driving mechanism and low optical efficiency have been hindering its practical applications. In this work, we demonstrate an active beam switching device that provides high directivity, uniform efficiency across diffraction orders, and a wide field of view while operating with only a single-gate bias. Experimentally, the metasurface achieves 57{\deg} of active beam switching from the 0th to the -1st order diffraction, with efficiencies of 0.084 and 0.078 and directivities of 0.765 and 0.836, respectively. Furthermore, an analytical framework using nonlocal quasinormal mode expansion provides deeper insight into the operating mechanism of active beam switching. Finally, we discuss the performance limitations of this design platform and provide insights into potential improvements.

For the 2D incompressible Euler equations, we establish global-in-time ($t \in \mathbb{R}$) stability of vortex quadrupoles satisfying odd symmetry with respect to both axes. Specifically, if the vorticity restricted to a quadrant is signed, sufficiently concentrated and close to its radial rearrangement up to a translation in $L^1$, we prove that it remains so for all times. The main difficulty is that the kinetic energy maximization problem in a quadrant -- the typical approach for establishing vortex stability -- lacks a solution, as the kinetic energy continues to increase when the vorticity escapes to infinity. We overcome this by taking dynamical information into account: finite-time desingularization result is combined with monotonicity of the first moment and a careful analysis of the interaction energies between vortices. The latter is achieved by new pointwise estimates on the Biot--Savart kernel and quantitative stability results for general interaction kernels. Moreover, with a similar strategy we obtain stability of a pair of opposite-signed Lamb dipoles moving away from each other., Comment: 33 pages, 6 figures

A search for proton decay into $e^+/\mu^+$ and a $\eta$ meson has been performed using data from a 0.373 Mton$\cdot$year exposure (6050.3 live days) of Super-Kamiokande. Compared to previous searches this work introduces an improved model of the intranuclear $\eta$ interaction cross section, resulting in a factor of two reduction in uncertainties from this source and $\sim$10\% increase in signal efficiency. No significant data excess was found above the expected number of atmospheric neutrino background events resulting in no indication of proton decay into either mode. Lower limits on the proton partial lifetime of $1.4\times\mathrm{10^{34}~years}$ for $p\rightarrow e^+\eta$ and $7.3\times\mathrm{10^{33}~years}$ for $p\rightarrow \mu^+\eta$ at the 90$\%$ C.L. were set. These limits are around 1.5 times longer than our previous study and are the most stringent to date.

The language-guided robot grasping task requires a robot agent to integrate multimodal information from both visual and linguistic inputs to predict actions for target-driven grasping. While recent approaches utilizing Multimodal Large Language Models (MLLMs) have shown promising results, their extensive computation and data demands limit the feasibility of local deployment and customization. To address this, we propose a novel CLIP-based multimodal parameter-efficient tuning (PET) framework designed for three language-guided object grounding and grasping tasks: (1) Referring Expression Segmentation (RES), (2) Referring Grasp Synthesis (RGS), and (3) Referring Grasp Affordance (RGA). Our approach introduces two key innovations: a bi-directional vision-language adapter that aligns multimodal inputs for pixel-level language understanding and a depth fusion branch that incorporates geometric cues to facilitate robot grasping predictions. Experiment results demonstrate superior performance in the RES object grounding task compared with existing CLIP-based full-model tuning or PET approaches. In the RGS and RGA tasks, our model not only effectively interprets object attributes based on simple language descriptions but also shows strong potential for comprehending complex spatial reasoning scenarios, such as multiple identical objects present in the workspace., Comment: This work has been submitted to ICRA 2025

Two-dimensional (2D) semiconductors are promising for low-power complementary metal oxide semiconductor (CMOS) electronics, which require ultrathin n- and p-type transistor channels. Among 2D semiconductors, WS2 is expected to have good conduction for both electrons and holes, but p-type WS2 transistors have been difficult to realize due to the relatively deep valence band and the presence of mid-gap states with conventional metal contacts. Here, we report topological semimetal NbP as p-type electrical contacts to bilayer WS2 with up to 5.8 microamperes per micron hole current at room temperature; this is the highest to date for sub 2 nm thin WS2 and more than 50 times larger than with metals like Ni or Pd. The p-type conduction is enabled by the simultaneously high work function and low density of states of the NbP, which reduce Fermi level pinning. These contacts are sputter-deposited at room temperature, an approach compatible with CMOS fabrication, a step towards enabling ultrathin WS2 semiconductors in future nanoelectronics.

In magnetic pyrochlore materials, the interplay of spin-orbit coupling, electronic correlations, and geometrical frustration gives rise to exotic quantum phases, including topological semimetals and spin ice. While these phases have been observed in isolation, the interface-driven phenomena emerging from their interaction have never been realized previously. Here, we report on the discovery of interfacial electronic anisotropy and rotational symmetry breaking at a heterostructure consisting of the Weyl semimetal Eu2Ir2O7 and spin ice Dy2Ti2O7. Subjected to magnetic fields, we unveil a six-fold anisotropic transport response that is theoretically accounted by a Kondo-coupled heterointerface, where the spin ice's field-tuned magnetism induces electron scattering in the Weyl semimetal's topological Fermi-arc states. Furthermore, at elevated magnetic fields, we reveal a two-fold anisotropic response indicative of a new symmetry-broken many-body state. This discovery showcases the nascent potential of complex quantum architectures in search of emergent phenomena unreachable in bulk crystals.

Large language models (LLMs) often show unwarranted preference for certain choice options when responding to multiple-choice questions, posing significant reliability concerns in LLM-automated systems. To mitigate this selection bias problem, previous solutions utilized debiasing methods to adjust the model's input and/or output. Our work, in contrast, investigates the model's internal representation of the selection bias. Specifically, we introduce a novel debiasing approach, Bias Node Pruning (BNP), which eliminates the linear layer parameters that contribute to the bias. Furthermore, we present Auxiliary Option Injection (AOI), a simple yet effective input modification technique for debiasing, which is compatible even with black-box LLMs. To provide a more systematic evaluation of selection bias, we review existing metrics and introduce Choice Kullback-Leibler Divergence (CKLD), which addresses the insensitivity of the commonly used metrics to label imbalance. Experiments show that our methods are robust and adaptable across various datasets when applied to three LLMs.

In abstractive summarization, the challenge of producing concise and accurate summaries arises from the vast amount of information contained in the source document. Consequently, although Large Language Models (LLMs) can generate fluent text, they often introduce inaccuracies by hallucinating content not found in the original source. While supervised fine-tuning methods that maximize likelihood contribute to this issue, they do not consistently enhance the faithfulness of the summaries. Preference-based optimization methods, such as Direct Preference Optimization (DPO), can further refine the model to align with human preferences. However, these methods still heavily depend on costly human feedback. In this work, we introduce a novel and straightforward approach called Model-based Preference Optimization (MPO) to fine-tune LLMs for improved summarization abilities without any human feedback. By leveraging the model's inherent summarization capabilities, we create a preference dataset that is fully generated by the model using different decoding strategies. Our experiments on standard summarization datasets and various metrics demonstrate that our proposed MPO significantly enhances the quality of generated summaries without relying on human feedback., Comment: Accepted by EMNLP 2024

In realistic cellular communication systems, multiple service providers will operate within different frequency ranges. Each serving cell, which is managed by a distinct service provider, is designed individually due to the orthogonal frequencies. However, when a reconfigurable intelligent surface (RIS) is deployed for a certain cell, the RIS still incurs reflective channels for the overall system since the RIS reflects signals across all frequency ranges. This may cause severe undesired performance degradation for the other cells unless the reflection coefficients are properly designed. To tackle this issue, by utilizing the Riemannian manifold optimization method, an RIS reflection coefficients design is proposed in this paper to maximize the performance improvements of the cell that deploys the RIS while minimizing the undesired performance degradation for the other cells simultaneously. Numerical results demonstrate that the proposed design can effectively balance the two objectives for practical scenarios., Comment: 6 pages, 4 figures, accepted to IEEE Transactions on Vehicular Technology

This paper introduces the hypothetical track-length fitting algorithm, a novel method for measuring the kinetic energies of ionizing particles in liquid argon time projection chambers (LArTPCs). The algorithm finds the most probable offset in track length for a track-like object by comparing the measured ionization density as a function of position with a theoretical prediction of the energy loss as a function of the energy, including models of electron recombination and detector response. The algorithm can be used to measure the energies of particles that interact before they stop, such as charged pions that are absorbed by argon nuclei. The algorithm's energy measurement resolutions and fractional biases are presented as functions of particle kinetic energy and number of track hits using samples of stopping secondary charged pions in data collected by the ProtoDUNE-SP detector, and also in a detailed simulation. Additional studies describe impact of the dE/dx model on energy measurement performance. The method described in this paper to characterize the energy measurement performance can be repeated in any LArTPC experiment using stopping secondary charged pions.

For automatic human figure drawing (HFD) assessment tasks, such as diagnosing autism spectrum disorder (ASD) using HFD images, the clarity and explainability of a model decision are crucial. Existing pixel-level attribution-based explainable AI (XAI) approaches demand considerable effort from users to interpret the semantic information of a region in an image, which can be often time-consuming and impractical. To overcome this challenge, we propose a part contribution evaluation based model explanation (PCEvE) framework. On top of the part detection, we measure the Shapley Value of each individual part to evaluate the contribution to a model decision. Unlike existing attribution-based XAI approaches, the PCEvE provides a straightforward explanation of a model decision, i.e., a part contribution histogram. Furthermore, the PCEvE expands the scope of explanations beyond the conventional sample-level to include class-level and task-level insights, offering a richer, more comprehensive understanding of model behavior. We rigorously validate the PCEvE via extensive experiments on multiple HFD assessment datasets. Also, we sanity-check the proposed method with a set of controlled experiments. Additionally, we demonstrate the versatility and applicability of our method to other domains by applying it to a photo-realistic dataset, the Stanford Cars., Comment: This papaer is under review

Charged Black holes in $AdS_5 \times S^5$ suffer from superradiant instabilities over a range of energies. Hairy black hole solutions (constructed within gauged supergravity) have previously been proposed as endpoints to this instability. We demonstrate that these hairy black holes are themselves unstable to the emission of large dual giant gravitons. We propose that the endpoint to this instability is given by Dual Dressed Black Holes (DDBH)s; configurations consisting of one, two, or three very large dual giant gravitons surrounding a core $AdS$ black hole with one, two, or three $SO(6)$ chemical potentials equal to unity. The dual giants each live at $AdS$ radial coordinates of order $\sqrt{N}$ and each carry charge of order $N^2$. The large separation makes DDBHs a very weakly interacting mix of their components and allows for a simple computation of their thermodynamics. We conjecture that DDBHs dominate the phase diagram of ${\cal N}=4$ Yang-Mills over a range of energies around the BPS plane, and provide an explicit construction of this phase diagram, briefly discussing the interplay with supersymmetry. We develop the quantum description of dual giants around black hole backgrounds and explicitly verify that DDBHs are stable to potential tunneling instabilities, precisely when the chemical potentials of the core black holes equal unity. We also construct the 10-dimensional DDBH supergravity solutions., Comment: 114 pages, 12 figures

We study retrieving a set of documents that covers various perspectives on a complex and contentious question (e.g., will ChatGPT do more harm than good?). We curate a Benchmark for Retrieval Diversity for Subjective questions (BERDS), where each example consists of a question and diverse perspectives associated with the question, sourced from survey questions and debate websites. On this data, retrievers paired with a corpus are evaluated to surface a document set that contains diverse perspectives. Our framing diverges from most retrieval tasks in that document relevancy cannot be decided by simple string matches to references. Instead, we build a language model based automatic evaluator that decides whether each retrieved document contains a perspective. This allows us to evaluate the performance of three different types of corpus (Wikipedia, web snapshot, and corpus constructed on the fly with retrieved pages from the search engine) paired with retrievers. Retrieving diverse documents remains challenging, with the outputs from existing retrievers covering all perspectives on only 33.74% of the examples. We further study the impact of query expansion and diversity-focused reranking approaches and analyze retriever sycophancy. Together, we lay the foundation for future studies in retrieval diversity handling complex queries.

Low Probability of Detection (LPD) communication aims to obscure the presence of radio frequency (RF) signals to evade surveillance. In the context of mobile surveillance utilizing unmanned aerial vehicles (UAVs), achieving LPD communication presents significant challenges due to the UAVs' rapid and continuous movements, which are characterized by unknown nonlinear dynamics. Therefore, accurately predicting future locations of UAVs is essential for enabling real-time LPD communication. In this paper, we introduce a novel framework termed predictive covert communication, aimed at minimizing detectability in terrestrial ad-hoc networks under multi-UAV surveillance. Our data-driven method synergistically integrates graph neural networks (GNN) with Koopman theory to model the complex interactions within a multi-UAV network and facilitating long-term predictions by linearizing the dynamics, even with limited historical data. Extensive simulation results substantiate that the predicted trajectories using our method result in at least 63%-75% lower probability of detection when compared to well-known state-of-the-art baseline approaches, showing promise in enabling low-latency covert operations in practical scenarios.

In this paper, we present an effective data augmentation framework leveraging the Large Language Model (LLM) and Diffusion Model (DM) to tackle the challenges inherent in data-scarce scenarios. Recently, DMs have opened up the possibility of generating synthetic images to complement a few training images. However, increasing the diversity of synthetic images also raises the risk of generating samples outside the target distribution. Our approach addresses this issue by embedding novel semantic information into text prompts via LLM and utilizing real images as visual prompts, thus generating semantically rich images. To ensure that the generated images remain within the target distribution, we dynamically adjust the guidance weight based on each image's CLIPScore to control the diversity. Experimental results show that our method produces synthetic images with enhanced diversity while maintaining adherence to the target distribution. Consequently, our approach proves to be more efficient in the few-shot setting on several benchmarks. Our code is available at https://github.com/kkyuhun94/dalda ., Comment: Accepted to ECCV Synthetic Data for Computer Vision Workshop (Oral)

This paper proposes a novel prompt engineering technique called Judgment of Thought (JoT) that is specifically tailored for binary logical reasoning tasks. JoT employs three roles$\unicode{x2014}$lawyer, prosecutor, and judge$\unicode{x2014}$to facilitate more reliable and accurate reasoning by the model. In this framework, the judge utilizes a high$\unicode{x2010}$level model, while the lawyer and prosecutor utilize low$\unicode{x2010}$level models. This structure helps the judge better understand the responses from both the lawyer and prosecutor, enabling a more accurate judgment. Experimental results on large language model (LLM) benchmark datasets, such as BigBenchHard and Winogrande, demonstrate that JoT outperforms existing methods, including Chain of Thought (CoT) and Self$\unicode{x2010}$Consistency (SC), in binary logical reasoning tasks. Additionally, in real$\unicode{x2010}$world tasks, such as Fake News Detection and SMS Spam Detection, JoT shows comparable or improved performance compared to existing techniques. JoT significantly enhances the accuracy and reliability of models in binary reasoning tasks and show potential for practical applicability across various domains. Future research should aim to further broaden the applicability of JoT and optimize its implementation for real$\unicode{x2010}$world problem$\unicode{x2010}$solving.

This paper presents Omni 3D - a 3D-stacked device architecture that is naturally enabled by back-end-of-line (BEOL)-compatible transistors. Omni 3D arbitrarily interleaves metal layers for both signal/power with FETs in 3D (i.e., nFETs and pFETs are stacked in 3D). Thus, signal/power routing layers have fine-grained, all-sided access to the FET active regions maximizing 3D standard cell design flexibility. This is in sharp contrast to approaches such as back-side power delivery networks (BSPDNs), complementary FETs (CFETs), and stacked FETs. Importantly, the routing flexibility of Omni 3D is enabled by double-side routing and an interleaved metal (IM) layer for inter- and intra-cell routing, respectively. In this work, we explore Omni 3D variants (e.g., both with and without the IM layer) and optimize these variants using a virtual-source BEOL-FET compact model. We establish a physical design flow that efficiently utilizes the double-side routing in Omni 3D and perform a thorough design-technology-co-optimization (DTCO) of Omni 3D device architecture on several design points. From our design flow, we project 2.0x improvement in the energy-delay product and 1.5x reduction in area compared to the state-of-the-art CFETs with BSPDNs., Comment: 8 pages, 15 figures

AI agents are increasingly autonomous in their interactions with human users and tools, leading to increased interactional safety risks. We present HAICOSYSTEM, a framework examining AI agent safety within diverse and complex social interactions. HAICOSYSTEM features a modular sandbox environment that simulates multi-turn interactions between human users and AI agents, where the AI agents are equipped with a variety of tools (e.g., patient management platforms) to navigate diverse scenarios (e.g., a user attempting to access other patients' profiles). To examine the safety of AI agents in these interactions, we develop a comprehensive multi-dimensional evaluation framework that uses metrics covering operational, content-related, societal, and legal risks. Through running 1840 simulations based on 92 scenarios across seven domains (e.g., healthcare, finance, education), we demonstrate that HAICOSYSTEM can emulate realistic user-AI interactions and complex tool use by AI agents. Our experiments show that state-of-the-art LLMs, both proprietary and open-sourced, exhibit safety risks in over 50\% cases, with models generally showing higher risks when interacting with simulated malicious users. Our findings highlight the ongoing challenge of building agents that can safely navigate complex interactions, particularly when faced with malicious users. To foster the AI agent safety ecosystem, we release a code platform that allows practitioners to create custom scenarios, simulate interactions, and evaluate the safety and performance of their agents., Comment: Both the second and third authors contributed equally

Crop field boundaries are foundational datasets for agricultural monitoring and assessments but are expensive to collect manually. Machine learning (ML) methods for automatically extracting field boundaries from remotely sensed images could help realize the demand for these datasets at a global scale. However, current ML methods for field instance segmentation lack sufficient geographic coverage, accuracy, and generalization capabilities. Further, research on improving ML methods is restricted by the lack of labeled datasets representing the diversity of global agricultural fields. We present Fields of The World (FTW) -- a novel ML benchmark dataset for agricultural field instance segmentation spanning 24 countries on four continents (Europe, Africa, Asia, and South America). FTW is an order of magnitude larger than previous datasets with 70,462 samples, each containing instance and semantic segmentation masks paired with multi-date, multi-spectral Sentinel-2 satellite images. We provide results from baseline models for the new FTW benchmark, show that models trained on FTW have better zero-shot and fine-tuning performance in held-out countries than models that aren't pre-trained with diverse datasets, and show positive qualitative zero-shot results of FTW models in a real-world scenario -- running on Sentinel-2 scenes over Ethiopia.

We perform the first search for $C\!P$ violation in ${D_{(s)}^{+}\to{}K_{S}^{0}K^{-}\pi^{+}\pi^{+}}$ decays. We use a combined data set from the Belle and Belle II experiments, which study $e^+e^-$ collisions at center-of-mass energies at or near the $\Upsilon(4S)$ resonance. We use 980 fb$^{-1}$ of data from Belle and 428 fb$^{-1}$ of data from Belle~II. We measure six $C\!P$-violating asymmetries that are based on triple products and quadruple products of the momenta of final-state particles, and also the particles' helicity angles. We obtain a precision at the level of 0.5% for $D^+\to{}K_{S}^{0}K^{-}\pi^{+}\pi^{+}$ decays, and better than 0.3% for $D^+_{s}\to{}K_{S}^{0}K^{-}\pi^{+}\pi^{+}$ decays. No evidence of $C\!P$ violation is found. Our results for the triple-product asymmetries are the most precise to date for singly-Cabibbo-suppressed $D^+$ decays. Our results for the other asymmetries are the first such measurements performed for charm decays., Comment: 21 pages, 10 figures

We present NanoVoice, a personalized text-to-speech model that efficiently constructs voice adapters for multiple speakers simultaneously. NanoVoice introduces a batch-wise speaker adaptation technique capable of fine-tuning multiple references in parallel, significantly reducing training time. Beyond building separate adapters for each speaker, we also propose a parameter sharing technique that reduces the number of parameters used for speaker adaptation. By incorporating a novel trainable scale matrix, NanoVoice mitigates potential performance degradation during parameter sharing. NanoVoice achieves performance comparable to the baselines, while training 4 times faster and using 45 percent fewer parameters for speaker adaptation with 40 reference voices. Extensive ablation studies and analysis further validate the efficiency of our model., Comment: Submitted to ICASSP 2025, Demo Page: https://nanovoice.github.io/

When applying parameter-efficient finetuning via LoRA onto speaker adaptive text-to-speech models, adaptation performance may decline compared to full-finetuned counterparts, especially for out-of-domain speakers. Here, we propose VoiceGuider, a parameter-efficient speaker adaptive text-to-speech system reinforced with autoguidance to enhance the speaker adaptation performance, reducing the gap against full-finetuned models. We carefully explore various ways of strengthening autoguidance, ultimately finding the optimal strategy. VoiceGuider as a result shows robust adaptation performance especially on extreme out-of-domain speech data. We provide audible samples in our demo page., Comment: Submitted to ICASSP 2025, Demo Page: https://voiceguider.github.io/

By analyzing $e^+e^-$ annihilation data corresponding to an integrated luminosity of 7.93 fb$^{-1}$, collected at the center-of-mass energy of 3.773 GeV with the BESIII detector, we search for the semileptonic decays $D^0\to K^-\eta e^+\nu_e$, $D^+\to K_S^0 \eta e^+\nu_e$ and $D^+\to \eta\eta e^+\nu_e$ for the first time. We present evidence for $D^0\to K^-\eta e^+\nu_e$ with a significance of $3.3\sigma$. The branching fraction of $D^0\to K^-\eta e^+\nu_e$ is measured to be $(0.84_{-0.34}^{+0.29}\pm0.22)\times 10^{-4}$. Here, the first uncertainties are statistical and the second ones are systematic. No significant signals are observed for the decays $D^+\to K_S^0 \eta e^+\nu_e$ and $D^+\to \eta\eta e^+\nu_e$ and we set the upper limits on their branching fractions., Comment: 10 pages,4 figures

Vision language models (VLMs) have shown promising reasoning capabilities across various benchmarks; however, our understanding of their visual perception remains limited. In this work, we propose an eye examination process to investigate how a VLM perceives images, specifically focusing on key elements of visual recognition, from primitive color and shape to semantic levels. To this end, we introduce a dataset named LENS to guide a VLM to follow the examination and check its readiness. Once the model is ready, we conduct the examination. Through this examination, we quantify and visualize VLMs' sensitivities to color and shape, and semantic matching. Our findings reveal that VLMs have varying sensitivity to different colors while consistently showing insensitivity to green across different VLMs. Also, we found different shape sensitivity and semantic recognition depending on LLM's capacity despite using the same fixed visual encoder. Our analyses and findings have potential to inspire the design of VLMs and the pre-processing of visual input to VLMs for improving application performance.

With the explosive growth of deep learning applications, the right to be forgotten has become increasingly in demand in various AI industries. For example, given a facial recognition system, some individuals may wish to remove images that might have been used in the training phase from the trained model. Unfortunately, modern deep neural networks sometimes unexpectedly leak personal identities. Recent studies have presented various machine unlearning algorithms to make a trained model unlearn the data to be forgotten. While these methods generally perform well in terms of forgetting scores, we have found that an unexpected modelutility drop can occur. This phenomenon, which we term correlation collapse, happens when the machine unlearning algorithms reduce the useful correlation between image features and the true label. To address this challenge, we propose Distribution-Level Feature Distancing (DLFD), a novel method that efficiently forgets instances while preventing correlation collapse. Our method synthesizes data samples so that the generated data distribution is far from the distribution of samples being forgotten in the feature space, achieving effective results within a single training epoch. Through extensive experiments on facial recognition datasets, we demonstrate that our approach significantly outperforms state-of-the-art machine unlearning methods., Comment: 10 pages, 6 figures

We consider the variational problem with a mass constraint arising from the two-dimensional dispersion managed nonlinear Schr\"odinger equation with power-law type nonlinearity. We prove a threshold phenomenon with respect to mass for the existence of minimizers for all possible powers of nonlinearities, including at the threshold itself. This threshold is closely related to the best constant for the Gagliardo-Nirenberg-Strichartz type inequality whose extremizers are found as a byproduct., Comment: 21 pages

Let $X$ be a complex manifold, and let $Y$ and $D$ be two reduced simple-normal-crossing (snc) divisors on $X$ with no common irreducible components. Given a proper locally K\"ahler morphism $\pi \colon X \to \Delta$ from $X$ to a complex analytic space $\Delta$, we prove Fujino's conjecture on the injectivity theorem in the relative setting in a generalized form. Specifically, we establish an injectivity result for the higher direct images under $\pi$ for the lc pairs $(X, D)$ as well as $(Y, D_Y)$, where $D_Y := D \cap Y$. As an application, this result immediately implies the injectivity theorem on holomorphically convex K\"ahler manifolds with reduced snc divisors. The main technique in the proof consists of the theory of harmonic integrals together with residue formulae associated with adjoint ideal sheaves, which are developed from our previous work for the absolute case (where $\Delta$ is a point and $X$ is compact). Additionally, we make use of the Takegoshi harmonic forms to deal with the non-compactness of $X$., Comment: 37 pages; see our previous related work at arXiv:2307.12025

Environmental hazards place certain individuals at disproportionately higher risks. As these hazards increasingly endanger human health, precise identification of the most vulnerable population subgroups is critical for public health. Moderated multiple regression (MMR) offers a straightforward method for investigating this by adding interaction terms between the exposure to a hazard and other population characteristics to a linear regression model. However, when the vulnerabilities are hidden within a cross-section of many characteristics, MMR is often limited in its capabilities to find any meaningful discoveries. Here, we introduce a hybrid method, named regression-guided neural networks (ReGNN), which utilizes artificial neural networks (ANNs) to non-linearly combine predictors, generating a latent representation that interacts with a focal predictor (i.e. variable measuring exposure to an environmental hazard). We showcase the use of ReGNN for investigating the population heterogeneity in the health effects of exposure to air pollution (PM2.5) on cognitive functioning scores. We demonstrate that population heterogeneity that would otherwise be hidden using traditional MMR can be found using ReGNN by comparing its results to the fit results of the traditional MMR models. In essence, ReGNN is a novel tool that enhances traditional regression models by effectively summarizing and quantifying an individual's susceptibility to health risks.

Service robots in human-centered environments such as hospitals, office buildings, and long-term care homes need to navigate while adhering to social norms to ensure the safety and comfortability of the people they are sharing the space with. Furthermore, they need to adapt to new social scenarios that can arise during robot navigation. In this paper, we present a novel Online Lifelong Vision Language architecture, OLiVia-Nav, which uniquely integrates vision-language models (VLMs) with an online lifelong learning framework for robot social navigation. We introduce a unique distillation approach, Social Context Contrastive Language Image Pre-training (SC-CLIP), to transfer the social reasoning capabilities of large VLMs to a lightweight VLM, in order for OLiVia-Nav to directly encode social and environment context during robot navigation. These encoded embeddings are used to generate and select robot social compliant trajectories. The lifelong learning capabilities of SC-CLIP enable OLiVia-Nav to update the lightweight VLM with robot trajectory predictions overtime as new social scenarios are encountered. We conducted extensive real-world experiments in diverse social navigation scenarios. The results showed that OLiVia-Nav outperformed existing state-of-the-art DRL and VLM methods in terms of mean squared error, Hausdorff loss, and personal space violation duration. Ablation studies also verified the design choices for OLiVia-Nav.

Despite the impressive success of text-to-image (TTI) generation models, existing studies overlook the issue of whether these models accurately convey factual information. In this paper, we focus on the problem of image hallucination, where images created by generation models fail to faithfully depict factual content. To address this, we introduce I-HallA (Image Hallucination evaluation with Question Answering), a novel automated evaluation metric that measures the factuality of generated images through visual question answering (VQA). We also introduce I-HallA v1.0, a curated benchmark dataset for this purpose. As part of this process, we develop a pipeline that generates high-quality question-answer pairs using multiple GPT-4 Omni-based agents, with human judgments to ensure accuracy. Our evaluation protocols measure image hallucination by testing if images from existing text-to-image models can correctly respond to these questions. The I-HallA v1.0 dataset comprises 1.2K diverse image-text pairs across nine categories with 1,000 rigorously curated questions covering various compositional challenges. We evaluate five text-to-image models using I-HallA and reveal that these state-of-the-art models often fail to accurately convey factual information. Moreover, we validate the reliability of our metric by demonstrating a strong Spearman correlation (rho=0.95) with human judgments. We believe our benchmark dataset and metric can serve as a foundation for developing factually accurate text-to-image generation models., Comment: 20 pages