Search

Dispersion measures (DM) of fast radio bursts (FRBs) probe the density of electrons in the intergalactic medium (IGM) along their lines-of-sight, including the average density versus distance to the source and its variations in direction. While previous study focused on low-redshift, FRBs are potentially detectable out to high redshift, where their DMs can, in principle, probe the epoch of reionization (EOR) and its patchiness. We present the first predictions from large-scale, radiation-hydrodynamical simulation of fully-coupled galaxy formation and reionization, using Cosmic Dawn (``CoDa")~II to model the density and ionization fields of the universe down to redshifts through the end of the EOR at $z_{re}\approx6.1$. Combining this with an N-body simulation CoDa~II--Dark Matter of the fully-ionized epoch from the EOR to the present, we calculate the mean and standard deviation of FRB DMs as functions of their source redshift. The mean and standard deviation of DM increase with redshift, reaching a plateau by $z(x_{HII}\lesssim0.25)\gtrsim8$, i.e. well above $z_{re}$. The mean-DM asymptote $\mathcal{DM}_{max} \approx 5900~\mathrm{pc\, cm^{-3}}$ reflects the end of the EOR and its duration. The standard deviation there is $\sigma_{DM, max}\approx497 ~\mathrm{pc\, cm^{-3}}$, reflecting inhomogeneities of both patchy reionization and density. Inhomogeneities in ionization during the EOR contribute $\mathcal{O}(1$ per cent) of this value of $\sigma_{DM,max}$ from FRBs at redshifts $z\gtrsim 8$. Current estimates of FRB rates suggest this may be detectable within a few years of observation., Comment: 14 pages, 10 figures, 2 tables, 2 appendices

The string-matching problem, ubiquitous in computer science, can significantly benefit from quantum algorithms due to their potential for greater efficiency compared to classical approaches. The practical implementation of the quantum string matching (QSM) algorithm requires fault-tolerant quantum computation due to the fragility of quantum information. A major obstacle in implementing fault-tolerant quantum computation is the high cost associated with executing T gates. This paper introduces the relative-phase Fredkin gate as a strategy to notably reduce the number of T gates (T-count) necessary for the QSM algorithm. This reduces the T-count from 14N^(3/2) log_2 N-O(N^(3/2)) to 8N^(3/2) log_2 N-O(N^(3/2)), where N represents the size of the database to be searched. Additionally, we demonstrate that our method is advantageous in terms of other circuit costs, such as the depth of T gates and the number of CNOT gates. This advancement contributes to the ongoing development of the QSM algorithm, paving the way for more efficient solutions in the field of computer science.

Isospin symmetry is a fundamental concept in nuclear physics. Even though isospin symmetry is partially broken, it holds approximately for most nuclear systems, which makes exceptions very interesting from the nuclear structure perspective. In this framework, it is expected that the spins and parities of the ground states of mirror nuclei should be the same, in particular for the simplest systems where a proton is exchanged with a neutron or vice versa. In this work, we present evidence that this assumption is broken in the mirror pair $^{71}$Br and $^{71}$Kr system. Our conclusions are based on a high-statistics $\beta$ decay study of $^{71}$Kr and on state-of-the-art shell model calculations. In our work, we also found evidence of a new state in $^{70}$Se, populated in the $\beta$-delayed proton emission process which can be interpreted as the long sought coexisting 0$^+$ state., Comment: 8 pages with references, 3 figures. Supplemental material 4 pages (1 table, 3 figures)

AI agents have been evaluated in isolation or within small groups, where interactions remain limited in scope and complexity. Large-scale simulations involving many autonomous agents -- reflecting the full spectrum of civilizational processes -- have yet to be explored. Here, we demonstrate how 10 - 1000+ AI agents behave and progress within agent societies. We first introduce the PIANO (Parallel Information Aggregation via Neural Orchestration) architecture, which enables agents to interact with humans and other agents in real-time while maintaining coherence across multiple output streams. We then evaluate agent performance in agent simulations using civilizational benchmarks inspired by human history. These simulations, set within a Minecraft environment, reveal that agents are capable of meaningful progress -- autonomously developing specialized roles, adhering to and changing collective rules, and engaging in cultural and religious transmission. These preliminary results show that agents can achieve significant milestones towards AI civilizations, opening new avenues for large simulations, agentic organizational intelligence, and integrating AI into human civilizations., Comment: 35 pages, 14 figures

We introduce the "Belief State Transformer", a next-token predictor that takes both a prefix and suffix as inputs, with a novel objective of predicting both the next token for the prefix and the previous token for the suffix. The Belief State Transformer effectively learns to solve challenging problems that conventional forward-only transformers struggle with, in a domain-independent fashion. Key to this success is learning a compact belief state that captures all relevant information necessary for accurate predictions. Empirical ablations show that each component of the model is essential in difficult scenarios where standard Transformers fall short. For the task of story writing with known prefixes and suffixes, our approach outperforms the Fill-in-the-Middle method for reaching known goals and demonstrates improved performance even when the goals are unknown. Altogether, the Belief State Transformer enables more efficient goal-conditioned decoding, better test-time inference, and high-quality text representations on small scale problems.

Spatial control methods using additional modules on pretrained diffusion models have gained attention for enabling conditional generation in natural images. These methods guide the generation process with new conditions while leveraging the capabilities of large models. They could be beneficial as training strategies in the context of 3D medical imaging, where training a diffusion model from scratch is challenging due to high computational costs and data scarcity. However, the potential application of spatial control methods with additional modules to 3D medical images has not yet been explored. In this paper, we present a tailored spatial control method for 3D medical images with a novel lightweight module, Volumetric Conditioning Module (VCM). Our VCM employs an asymmetric U-Net architecture to effectively encode complex information from various levels of 3D conditions, providing detailed guidance in image synthesis. To examine the applicability of spatial control methods and the effectiveness of VCM for 3D medical data, we conduct experiments under single- and multimodal conditions scenarios across a wide range of dataset sizes, from extremely small datasets with 10 samples to large datasets with 500 samples. The experimental results show that the VCM is effective for conditional generation and efficient in terms of requiring less training data and computational resources. We further investigate the potential applications for our spatial control method through axial super-resolution for medical images. Our code is available at \url{https://github.com/Ahn-Ssu/VCM}, Comment: 17 pages, 18 figures, accepted @ WACV 2025

Log-Structured Merge (LSM) tree-based Key-Value Stores (KVSs) are widely adopted for their high performance in write-intensive environments, but they often face performance degradation due to write stalls during compaction. Prior solutions, such as regulating I/O traffic or using multiple compaction threads, can cause unexpected drops in throughput or increase host CPU usage, while hardware-based approaches using FPGA, GPU, and DPU aimed at reducing compaction duration introduce additional hardware costs. In this study, we propose KVACCEL, a novel hardware-software co-design framework that eliminates write stalls by leveraging a dual-interface SSD. KVACCEL allocates logical NAND flash space to support both block and key-value interfaces, using the key-value interface as a temporary write buffer during write stalls. This strategy significantly reduces write stalls, optimizes resource usage, and ensures consistency between the host and device by implementing an in-device LSM-based write buffer with an iterator-based range scan mechanism. Our extensive evaluation shows that for write-intensive workloads, KVACCEL outperforms ADOC by up to 1.17x in terms of throughput and performance-to-CPU-utilization efficiency. For mixed read-write workloads, both demonstrate comparable performance., Comment: 11 pages, 14 figures

Movable single atoms have drawn significant attention for their potentials as flying quantum memory in non-local, dynamic quantum computing architectures. However, when dynamic optical tweezers are employed to control atoms opto-mechanically, conventional methods such as adiabatic controls and constant jerk controls are either inherently slow or induce mechanical heating, leading to atom loss over long distances or at high speeds. To address these challenges, we explore the method known as shortcuts to adiabaticity (STA) as an efficient alternative for fast and reliable atom transport control. We present a series of proof-of-concept experiments demonstrating that STA-based optical tweezer trajectories can achieve both rapid and reliable single-atom transport. These experiments include moving atoms between two locations, adjusting speeds en route, and navigating curved trajectories. Our results indicate that atoms can be transported with a constant acceleration on average over distances that is only limited by trap lifetime, while effectively suppressing vibrational heating. This makes STA methods particularly well-suited for long-distance atom transport, potentially spanning distances over centimeter scales, such as between quantum information devices., Comment: 10 pages, 5 figures

Numerous studies have assessed the proficiency of AI systems, particularly large language models (LLMs), in facilitating everyday tasks such as email writing, question answering, and creative content generation. However, researchers face unique challenges and opportunities in leveraging LLMs for their own work, such as brainstorming research ideas, designing experiments, and writing or reviewing papers. In this study, we introduce AAAR-1.0, a benchmark dataset designed to evaluate LLM performance in three fundamental, expertise-intensive research tasks: (i) EquationInference, assessing the correctness of equations based on the contextual information in paper submissions; (ii) ExperimentDesign, designing experiments to validate research ideas and solutions; (iii) PaperWeakness, identifying weaknesses in paper submissions; and (iv) REVIEWCRITIQUE, identifying each segment in human reviews is deficient or not. AAAR-1.0 differs from prior benchmarks in two key ways: first, it is explicitly research-oriented, with tasks requiring deep domain expertise; second, it is researcher-oriented, mirroring the primary activities that researchers engage in on a daily basis. An evaluation of both open-source and proprietary LLMs reveals their potential as well as limitations in conducting sophisticated research tasks. We will keep iterating AAAR-1.0 to new versions., Comment: Project Webpage: https://renzelou.github.io/AAAR-1.0/

LLMs have emerged as a promising tool for assisting individuals in diverse text-generation tasks, including job-related texts. However, LLM-generated answers have been increasingly found to exhibit gender bias. This study evaluates three LLMs (GPT-3.5, GPT-4, Claude) to conduct a multifaceted audit of LLM-generated interview responses across models, question types, and jobs, and their alignment with two gender stereotypes. Our findings reveal that gender bias is consistent, and closely aligned with gender stereotypes and the dominance of jobs. Overall, this study contributes to the systematic examination of gender bias in LLM-generated interview responses, highlighting the need for a mindful approach to mitigate such biases in related applications., Comment: Accepted to NeurlIPS 2024, SoLaR workshop

The integration of the Internet of Things (IoT) and modern Artificial Intelligence (AI) has given rise to a new paradigm known as the Artificial Intelligence of Things (AIoT). In this survey, we provide a systematic and comprehensive review of AIoT research. We examine AIoT literature related to sensing, computing, and networking & communication, which form the three key components of AIoT. In addition to advancements in these areas, we review domain-specific AIoT systems that are designed for various important application domains. We have also created an accompanying GitHub repository, where we compile the papers included in this survey: https://github.com/AIoT-MLSys-Lab/AIoT-Survey. This repository will be actively maintained and updated with new research as it becomes available. As both IoT and AI become increasingly critical to our society, we believe AIoT is emerging as an essential research field at the intersection of IoT and modern AI. We hope this survey will serve as a valuable resource for those engaged in AIoT research and act as a catalyst for future explorations to bridge gaps and drive advancements in this exciting field., Comment: Accepted in ACM Transactions on Sensor Networks (TOSN)

Understanding species dynamics in heterogeneous environments is essential for ecosystem studies. Traditional models assumed homogeneous habitats, but recent approaches include spatial and temporal variability, highlighting species migration. We adopt starvation-driven diffusion (SDD) models as nonlinear diffusion to describe species dispersal based on local resource conditions, showing advantages for species survival. However, accurate prediction remains challenging due to model simplifications. This study uses multi-agent reinforcement learning (MARL) with deep Q-networks (DQN) to simulate single species and predator-prey interactions, incorporating SDD-type rewards. Our simulations reveal evolutionary dispersal strategies, providing insights into species dispersal mechanisms and validating traditional mathematical models.

This paper presents novel algorithms for the Fermat-Weber Location Problem, guiding an autonomous agent to the point that minimizes the weighted sum of Euclidean distances to some beacons using only bearing measurements. The existing results address only the simple scenario where the beacons are stationary and the agent is modeled by a single integrator. In this paper, we propose a number of bearing-only algorithms that let the agent, which can be modeled as either a single-integrator or a double-integrator, follow the Fermat-Weber point of a group of stationary or moving beacons. The theoretical results are rigorously proven using Lyapunov theory and supported with simulation examples., Comment: 14 pages (double-column), 7 figures, submitted to a journal

Social media is often the first place where communities discuss the latest societal trends. Prior works have utilized this platform to extract epidemic-related information (e.g. infections, preventive measures) to provide early warnings for epidemic prediction. However, these works only focused on English posts, while epidemics can occur anywhere in the world, and early discussions are often in the local, non-English languages. In this work, we introduce the first multilingual Event Extraction (EE) framework SPEED++ for extracting epidemic event information for a wide range of diseases and languages. To this end, we extend a previous epidemic ontology with 20 argument roles; and curate our multilingual EE dataset SPEED++ comprising 5.1K tweets in four languages for four diseases. Annotating data in every language is infeasible; thus we develop zero-shot cross-lingual cross-disease models (i.e., training only on English COVID data) utilizing multilingual pre-training and show their efficacy in extracting epidemic-related events for 65 diverse languages across different diseases. Experiments demonstrate that our framework can provide epidemic warnings for COVID-19 in its earliest stages in Dec 2019 (3 weeks before global discussions) from Chinese Weibo posts without any training in Chinese. Furthermore, we exploit our framework's argument extraction capabilities to aggregate community epidemic discussions like symptoms and cure measures, aiding misinformation detection and public attention monitoring. Overall, we lay a strong foundation for multilingual epidemic preparedness., Comment: Accepted at EMNLP 2024

This paper presents the development of a dashboard designed specifically for teachers in English as a Foreign Language (EFL) writing education. Leveraging LLMs, the dashboard facilitates the analysis of student interactions with an essay writing system, which integrates ChatGPT for real-time feedback. The dashboard aids teachers in monitoring student behavior, identifying noneducational interaction with ChatGPT, and aligning instructional strategies with learning objectives. By combining insights from NLP and Human-Computer Interaction (HCI), this study demonstrates how a human-centered approach can enhance the effectiveness of teacher dashboards, particularly in ChatGPT-integrated learning., Comment: EMNLP 2024 Workshop CustomNLP4U. arXiv admin note: text overlap with arXiv:2405.19691

Science is increasingly global, with international collaboration playing a crucial role in advancing scientific development and knowledge exchange across borders. However, the processes that regulate how scientific labor is distributed among countries remain underexplored, leading to challenges in ensuring both effective collaboration and equitable participation across diverse scientific communities. Here, we leverage three million internationally coauthored publications produced by countries worldwide to examine the division of scientific labor in international collaboration, identify the factors that shape this distribution, and assess its broader consequences. Our findings uncover a persistent hierarchical structure in international collaboration, with researchers from scientifically advanced countries tend to occupy leading roles, while those from less-developed countries are often relegated to supportive roles, even after controlling for various influential factors. This hierarchy is also reflected in the research content, as countries with lower scientific capacity tend to participate in international collaborations that deviate from their domestic science. By analyzing the labor division within international collaborations, we demonstrate that researchers from less-developed countries face systematic disadvantages, which not only limit their contributions to the global scientific community but also prevent them from fully benefiting from international collaborations., Comment: 35 pages, 7 figures, 3 tables

Collaborative Perception (CP) is a process in which an ego agent receives and fuses sensor information from surrounding vehicles and infrastructure to enhance its perception capability. To evaluate the need for infrastructure equipped with sensors, extensive and quantitative analysis of the role of infrastructure data in CP is crucial, yet remains underexplored. To address this gap, we first quantitatively assess the importance of infrastructure data in existing vehicle-centric CP, where the ego agent is a vehicle. Furthermore, we compare vehicle-centric CP with infra-centric CP, where the ego agent is now the infrastructure, to evaluate the effectiveness of each approach. Our results demonstrate that incorporating infrastructure data improves 3D detection accuracy by up to 10.87%, and infra-centric CP shows enhanced noise robustness and increases accuracy by up to 42.53% compared with vehicle-centric CP., Comment: Accepted by ECCV 2024 Workshop MAAS, 14 pages

We present significant upgrades to the VAMPIRES instrument, a visible-light (600 nm to 800 nm) high-contrast imaging polarimeter integrated within SCExAO on the Subaru telescope. Key enhancements include new qCMOS detectors, coronagraphs, polarization optics, and a multiband imaging mode, improving sensitivity, resolution, and efficiency. These upgrades position VAMPIRES as a powerful tool for studying sub-stellar companions, accreting protoplanets, circumstellar disks, stellar jets, stellar mass-loss shells, and solar system objects. The instrument achieves angular resolutions from 17 mas to 21 mas and Strehl ratios up to 60\%, with 5$\sigma$ contrast limits of $10^{\text{-}4}$ at 0.1'' to $10^{\text{-}6}$ beyond 0.5''. We demonstrate these capabilities through spectro-polarimetric coronagraphic imaging of the HD 169142 circumstellar disk, ADI+SDI imaging of the sub-stellar companion HD 1160B, narrowband H$\alpha$ imaging of the R Aqr emission nebula, and spectro-polarimetric imaging of Neptune., Comment: 36 pages, 33 figures, accepted to PASP

A deep learning system typically suffers from a lack of reproducibility that is partially rooted in hardware or software implementation details. The irreproducibility leads to skepticism in deep learning technologies and it can hinder them from being deployed in many applications. In this work, the irreproducibility issue is analyzed where deep learning is employed in compression systems while the encoding and decoding may be run on devices from different manufacturers. The decoding process can even crash due to a single bit difference, e.g., in a learning-based entropy coder. For a given deep learning-based module with limited resources for protection, we first suggest that reproducibility can only be assured when the mismatches are bounded. Then a safeguarding mechanism is proposed to tackle the challenges. The proposed method may be applied for different levels of protection either at the reconstruction level or at a selected decoding level. Furthermore, the overhead introduced for the protection can be scaled down accordingly when the error bound is being suppressed. Experiments demonstrate the effectiveness of the proposed approach for learning-based compression systems, e.g., in image compression and point cloud compression., Comment: Accepted at MMSP 2024

Recently, 3D LiDAR has emerged as a promising technique in the field of gait-based person identification, serving as an alternative to traditional RGB cameras, due to its robustness under varying lighting conditions and its ability to capture 3D geometric information. However, long capture distances or the use of low-cost LiDAR sensors often result in sparse human point clouds, leading to a decline in identification performance. To address these challenges, we propose a sparse-to-dense upsampling model for pedestrian point clouds in LiDAR-based gait recognition, named LidarGSU, which is designed to improve the generalization capability of existing identification models. Our method utilizes diffusion probabilistic models (DPMs), which have shown high fidelity in generative tasks such as image completion. In this work, we leverage DPMs on sparse sequential pedestrian point clouds as conditional masks in a video-to-video translation approach, applied in an inpainting manner. We conducted extensive experiments on the SUSTeck1K dataset to evaluate the generative quality and recognition performance of the proposed method. Furthermore, we demonstrate the applicability of our upsampling model using a real-world dataset, captured with a low-resolution sensor across varying measurement distances.

Recently, NIRSpec PRISM/CLEAR observations by JWST have begun providing rest-frame UV continuum measurements of galaxies at $z\gtrsim7$, revealing signatures of Ly$\alpha$ damping-wing (DW) absorption by the intergalactic medium (IGM). We develop a methodology to constrain the global ionization fraction of the IGM $(Q_{\rm HII})$ using low-resolution spectra, employing the random forest classification (RFC) method. We construct mock spectra using the simulated galaxies and the IGM from the Cosmic Dawn II simulation and train RFC models to estimate $Q_{\rm HII}$ at the redshift of the source and to detect the presence of a damped Ly$\alpha$ absorber (DLA). We find that individual galaxy spectra with spectral bins between 1220 and 1270 {\AA} and with signal-to-noise ratios greater than 20 can place tight constraints on $Q_{\rm HII}$, provided the UV continuum is accurately modeled. This method is particularly effective for the early phase of reionization ($Q_{\rm HII}<50\%$), when the IGM opacity is high in the DW. As a demonstration, we apply our model to existing NIRSpec PRISM/CLEAR spectra, placing upper bounds of $Q_{\rm HII}=59.6\%$, $5.6\%$, and $18.5\%$ at $z=7.7,~9.4,$ and $10.6$, respectively, with $68\%$ confidence, though several modeling uncertainties remain to be discussed. These constraints favor late-starting reionization models, where $\gtrsim 80\%$ of the IGM is ionized after $z=8$. We conclude that high SNR observations of carefully selected targets around $z\sim7-9$ can effectively constrain reionization models., Comment: 20 pages, 11 figures, submitted to ApJ

In this paper, we explore the application of the Decision Transformer, a decision-making algorithm based on the Generative Pre-trained Transformer (GPT) architecture, to multi-vehicle coordination at unsignalized intersections. We formulate the coordination problem so as to find the optimal trajectories for multiple vehicles at intersections, modeling it as a sequence prediction task to fully leverage the power of GPTs as a sequence model. Through extensive experiments, we compare our approach to a reservation-based intersection management system. Our results show that the Decision Transformer can outperform the training data in terms of total travel time and can be generalized effectively to various scenarios, including noise-induced velocity variations, continuous interaction environments, and different vehicle numbers and road configurations., Comment: 7 pages

Hyperparameter optimization is an essential component in many data science pipelines and typically entails exhaustive time and resource-consuming computations in order to explore the combinatorial search space. Similar to this problem, other key operations in data science pipelines exhibit the exact same properties. Important examples are: neural architecture search, where the goal is to identify the best design choices for a neural network, and query cardinality estimation, where given different predicate values for a SQL query the goal is to estimate the size of the output. In this paper, we abstract away those essential components of data science pipelines and we model them as instances of tensor completion, where each variable of the search space corresponds to one mode of the tensor, and the goal is to identify all missing entries of the tensor, corresponding to all combinations of variable values, starting from a very small sample of observed entries. In order to do so, we first conduct a thorough experimental evaluation of existing state-of-the-art tensor completion techniques and introduce domain-inspired adaptations (such as smoothness across the discretized variable space) and an ensemble technique which is able to achieve state-of-the-art performance. We extensively evaluate existing and proposed methods in a number of datasets generated corresponding to (a) hyperparameter optimization for non-neural network models, (b) neural architecture search, and (c) variants of query cardinality estimation, demonstrating the effectiveness of tensor completion as a tool for automating data science pipelines. Furthermore, we release our generated datasets and code in order to provide benchmarks for future work on this topic.

Metal-organic frameworks (MOFs) are a class of crystalline materials with promising applications in many areas such as carbon capture and drug delivery. In this work, we introduce MOFFlow, the first deep generative model tailored for MOF structure prediction. Existing approaches, including ab initio calculations and even deep generative models, struggle with the complexity of MOF structures due to the large number of atoms in the unit cells. To address this limitation, we propose a novel Riemannian flow matching framework that reduces the dimensionality of the problem by treating the metal nodes and organic linkers as rigid bodies, capitalizing on the inherent modularity of MOFs. By operating in the $SE(3)$ space, MOFFlow effectively captures the roto-translational dynamics of these rigid components in a scalable way. Our experiment demonstrates that MOFFlow accurately predicts MOF structures containing several hundred atoms, significantly outperforming conventional methods and state-of-the-art machine learning baselines while being much faster., Comment: 10 pages, 6 figures

The adaptation of large language models (LLMs) to chemistry has shown promising performance in molecular understanding tasks, such as generating a text description from a molecule. However, proper reasoning based on molecular structural information remains a significant challenge, e.g., even advanced LLMs such as GPT-4o struggle to identify functional groups which are crucial for inferring the molecular property of interest. To address this limitation, we propose StructCoT, a structure-aware chain-of-thought (CoT) that enhances LLMs' understanding of molecular structures by explicitly injecting the key structural features of molecules. Moreover, we introduce two fine-tuning frameworks for adapting the existing LLMs to use our StructCoT. Our experiments demonstrate that incorporating StructCoT with our fine-tuning frameworks leads to consistent improvements in both molecular understanding tasks.

Generating Natural Language Explanations (NLEs) for model predictions on medical images, particularly those depicting thoracic pathologies, remains a critical and challenging task. Existing methodologies often struggle due to general models' insufficient domain-specific medical knowledge and privacy concerns associated with retrieval-based augmentation techniques. To address these issues, we propose a novel Vision-Language framework augmented with a Knowledge Graph (KG)-based datastore, which enhances the model's understanding by incorporating additional domain-specific medical knowledge essential for generating accurate and informative NLEs. Our framework employs a KG-based retrieval mechanism that not only improves the precision of the generated explanations but also preserves data privacy by avoiding direct data retrieval. The KG datastore is designed as a plug-and-play module, allowing for seamless integration with various model architectures. We introduce and evaluate three distinct frameworks within this paradigm: KG-LLaVA, which integrates the pre-trained LLaVA model with KG-RAG; Med-XPT, a custom framework combining MedCLIP, a transformer-based projector, and GPT-2; and Bio-LLaVA, which adapts LLaVA by incorporating the Bio-ViT-L vision model. These frameworks are validated on the MIMIC-NLE dataset, where they achieve state-of-the-art results, underscoring the effectiveness of KG augmentation in generating high-quality NLEs for thoracic pathologies.

Generative models in drug discovery have recently gained attention as efficient alternatives to brute-force virtual screening. However, most existing models do not account for synthesizability, limiting their practical use in real-world scenarios. In this paper, we propose RxnFlow, which sequentially assembles molecules using predefined molecular building blocks and chemical reaction templates to constrain the synthetic chemical pathway. We then train on this sequential generating process with the objective of generative flow networks (GFlowNets) to generate both highly rewarded and diverse molecules. To mitigate the large action space of synthetic pathways in GFlowNets, we implement a novel action space subsampling method. This enables RxnFlow to learn generative flows over extensive action spaces comprising combinations of 1.2 million building blocks and 71 reaction templates without significant computational overhead. Additionally, RxnFlow can employ modified or expanded action spaces for generation without retraining, allowing for the introduction of additional objectives or the incorporation of newly discovered building blocks. We experimentally demonstrate that RxnFlow outperforms existing reaction-based and fragment-based models in pocket-specific optimization across various target pockets. Furthermore, RxnFlow achieves state-of-the-art performance on CrossDocked2020 for pocket-conditional generation, with an average Vina score of -8.85kcal/mol and 34.8% synthesizability., Comment: 25 pages, 10 figures

Curating datasets that span multiple languages is challenging. To make the collection more scalable, researchers often incorporate one or more imperfect classifiers in the process, like language identification models. These models, however, are prone to failure, resulting in some language subsets being unreliable for downstream tasks. We introduce a statistical test, the Preference Proportion Test, for identifying such unreliable subsets. By annotating only 20 samples for a language subset, we're able to identify systematic transcription errors for 10 language subsets in a recent large multilingual transcribed audio dataset, X-IPAPack (Zhu et al., 2024). We find that filtering this low-quality data out when training models for the downstream task of phonetic transcription brings substantial benefits, most notably a 25.7% relative improvement on transcribing recordings in out-of-distribution languages. Our method lays a path forward for systematic and reliable multilingual dataset auditing., Comment: 16 pages, 6 figures

Predicting the ground-state 3D molecular conformations from 2D molecular graphs is critical in computational chemistry due to its profound impact on molecular properties. Deep learning (DL) approaches have recently emerged as promising alternatives to computationally-heavy classical methods such as density functional theory (DFT). However, we discover that existing DL methods inadequately model inter-atomic forces, particularly for non-bonded atomic pairs, due to their naive usage of bonds and pairwise distances. Consequently, significant prediction errors occur for atoms with low degree (i.e., low coordination numbers) whose conformations are primarily influenced by non-bonded interactions. To address this, we propose REBIND, a novel framework that rewires molecular graphs by adding edges based on the Lennard-Jones potential to capture non-bonded interactions for low-degree atoms. Experimental results demonstrate that REBIND significantly outperforms state-of-the-art methods across various molecular sizes, achieving up to a 20\% reduction in prediction error., Comment: 17 pages, 4 figures, 5 tables

Recent advancements in large language models (LLMs) have demonstrated impressive performance in generating molecular structures as drug candidates, which offers significant potential to accelerate drug discovery. However, the current LLMs overlook a critical requirement for drug discovery: proposing a diverse set of molecules. This diversity is essential for improving the chances of finding a viable drug, as it provides alternative molecules that may succeed where others fail in wet-lab or clinical validations. Despite such a need for diversity, the LLMs often output structurally similar molecules from a given prompt. While decoding schemes like beam search may enhance textual diversity, this often does not align with molecular structural diversity. In response, we propose a new method for fine-tuning molecular generative LLMs to autoregressively generate a set of structurally diverse molecules, where each molecule is generated by conditioning on the previously generated molecules. Our approach consists of two stages: (1) supervised fine-tuning to adapt LLMs to autoregressively generate molecules in a sequence and (2) reinforcement learning to maximize structural diversity within the generated molecules. Our experiments show that (1) our fine-tuning approach enables the LLMs to better discover diverse molecules compared to existing decoding schemes and (2) our fine-tuned model outperforms other representative LLMs in generating diverse molecules, including the ones fine-tuned on chemical domains.

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

In this work, we investigate how a model's tendency to broadly integrate its parametric knowledge evolves throughout pretraining, and how this behavior affects overall performance, particularly in terms of knowledge acquisition and forgetting. We introduce the concept of knowledge entropy, which quantifies the range of memory sources the model engages with; high knowledge entropy indicates that the model utilizes a wide range of memory sources, while low knowledge entropy suggests reliance on specific sources with greater certainty. Our analysis reveals a consistent decline in knowledge entropy as pretraining advances. We also find that the decline is closely associated with a reduction in the model's ability to acquire and retain knowledge, leading us to conclude that diminishing knowledge entropy (smaller number of active memory sources) impairs the model's knowledge acquisition and retention capabilities. We find further support for this by demonstrating that increasing the activity of inactive memory sources enhances the model's capacity for knowledge acquisition and retention.

In this paper, we construct an $S^1$-equivariant version of the relative symplectic cohomology developed by Varolgunes. As an application, we construct a relative version of Gutt-Hutchings capacities and a relative version of symplectic (co)homology capacity. We will see that these relative symplectic capacities can detect the diplaceability and the heaviness of a compact subset of a symplectic manifold. We compare the first relative Gutt-Hutchings capacity and the relative symplectic (co)homology capacity and prove that they are equal to each other under a convexity assumption., Comment: 53 pages

Recent work in imitation learning has shown that having an expert controller that is both suitably smooth and stable enables stronger guarantees on the performance of the learned controller. However, constructing such smoothed expert controllers for arbitrary systems remains challenging, especially in the presence of input and state constraints. As our primary contribution, we show how such a smoothed expert can be designed for a general class of systems using a log-barrier-based relaxation of a standard Model Predictive Control (MPC) optimization problem. Improving upon our previous work, we show that barrier MPC achieves theoretically optimal error-to-smoothness tradeoff along some direction. At the core of this theoretical guarantee on smoothness is an improved lower bound we prove on the optimality gap of the analytic center associated with a convex Lipschitz function, which we believe could be of independent interest. We validate our theoretical findings via experiments, demonstrating the merits of our smoothing approach over randomized smoothing., Comment: 36 pages, 3 figures. This work extends our previous result in arXiv:2306.01914, which has been accepted for publication in CDC 2024. An earlier version of this manuscript was submitted as part of DP's Master's thesis

This article investigates long-term orbits within the Earth's magnetosphere, specifically focusing on orbits where the argument of periapsis is synchronized with changes induced by lunar gravity assists and the Earth's argument of latitude over a complete orbital period in Earth-Moon resonance. In the Earth-Moon rotating frame, resonance orbits appear repetitive; however, the argument of periapsis shifts due to the third-body effects from lunar flybys. The extent of this shift is influenced by the Jacobi integral associated with the resonance orbit. To identify feasible resonance orbits and the optimal Jacobi integral, we map the argument of periapsis change against the Jacobi integral for each prospective orbit. This synchronization allows the spacecraft to remain within a confined region in space when observed from the Sun-Earth rotating frame. Finally, the article discusses the applications of these long-term Earth magnetosphere science orbits, including orbit-orientation reconfiguration (station keeping) and stability., Comment: 20 pages, Preliminary results shared in 2024 AIAA SCITECH Conference Conference paper: Sun-Earth Harmonic Orbit via Earth-Moon Resonance Orbit

Recent advances in large language models (LLMs) have significantly impacted the domain of multi-hop question answering (MHQA), where systems are required to aggregate information and infer answers from disparate pieces of text. However, the autoregressive nature of LLMs inherently poses a challenge as errors may accumulate if mistakes are made in the intermediate reasoning steps. This paper introduces Monte-Carlo tree search for Zero-shot multi-hop Question Answering (MZQA), a framework based on Monte-Carlo tree search (MCTS) to identify optimal reasoning paths in MHQA tasks, mitigating the error propagation from sequential reasoning processes. Unlike previous works, we propose a zero-shot prompting method, which relies solely on instructions without the support of hand-crafted few-shot examples that typically require domain expertise. We also introduce a behavioral cloning approach (MZQA-BC) trained on self-generated MCTS inference trajectories, achieving an over 10-fold increase in reasoning speed with bare compromise in performance. The efficacy of our method is validated on standard benchmarks such as HotpotQA, 2WikiMultihopQA, and MuSiQue, demonstrating that it outperforms existing frameworks., Comment: Work in Progress

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

We focus on the problem of rearranging a set of objects within a confined space with a nonholonomically constrained mobile robot pusher. This problem is relevant to many real-world domains, including warehouse automation and construction. These domains give rise to instances involving a combination of geometric, kinematic, and physics constraints, which make planning particularly challenging. Prior work often makes simplifying assumptions like the use of holonomic mobile robots or dexterous manipulators capable of unconstrained overhand reaching. Our key insight is we can empower even a constrained mobile pusher to tackle complex rearrangement tasks by enabling it to modify the environment to its favor in a constraint-aware fashion. To this end, we describe a Push-Traversability graph, whose vertices represent poses that the pusher can push objects from and edges represent optimal, kinematically feasible, and stable push-rearrangements of objects. Based on this graph, we develop ReloPush, a planning framework that leverages Dubins curves and standard graph search techniques to generate an efficient sequence of object rearrangements to be executed by the pusher. We evaluate ReloPush across a series of challenging scenarios, involving the rearrangement of densely cluttered workspaces with up to eight objects by a 1tenth mobile robot pusher. ReloPush exhibits orders of magnitude faster runtimes and significantly more robust execution in the real world, evidenced in lower execution times and fewer losses of object contact, compared to two baselines lacking our proposed graph structure., Comment: Submitted to ICRA 2025

Video anomaly detection (VAD) is a crucial task in video analysis and surveillance within computer vision. Currently, VAD is gaining attention with memory techniques that store the features of normal frames. The stored features are utilized for frame reconstruction, identifying an abnormality when a significant difference exists between the reconstructed and input frames. However, this approach faces several challenges due to the simultaneous optimization required for both the memory and encoder-decoder model. These challenges include increased optimization difficulty, complexity of implementation, and performance variability depending on the memory size. To address these challenges,we propose an effective memory method for VAD, called VideoPatchCore. Inspired by PatchCore, our approach introduces a structure that prioritizes memory optimization and configures three types of memory tailored to the characteristics of video data. This method effectively addresses the limitations of existing memory-based methods, achieving good performance comparable to state-of-the-art methods. Furthermore, our method requires no training and is straightforward to implement, making VAD tasks more accessible. Our code is available online at github.com/SkiddieAhn/Paper-VideoPatchCore., Comment: Accepted to ACCV 2024

Vision language models (VLM) have achieved success in both natural language comprehension and image recognition tasks. However, their use in pathology report generation for whole slide images (WSIs) is still limited due to the huge size of multi-scale WSIs and the high cost of WSI annotation. Moreover, in most of the existing research on pathology report generation, sufficient validation regarding clinical efficacy has not been conducted. Herein, we propose a novel Patient-level Multi-organ Pathology Report Generation (PMPRG) model, which utilizes the multi-scale WSI features from our proposed multi-scale regional vision transformer (MR-ViT) model and their real pathology reports to guide VLM training for accurate pathology report generation. The model then automatically generates a report based on the provided key features attended regional features. We assessed our model using a WSI dataset consisting of multiple organs, including the colon and kidney. Our model achieved a METEOR score of 0.68, demonstrating the effectiveness of our approach. This model allows pathologists to efficiently generate pathology reports for patients, regardless of the number of WSIs involved.

The deformation of the lateral ventricle (LV) shape is widely studied to identify specific morphometric changes associated with diseases. Since LV enlargement is considered a relative change due to brain atrophy, local longitudinal LV deformation can indicate deformation in adjacent brain areas. However, conventional methods for LV shape analysis focus on modeling the solely segmented LV mask. In this work, we propose a novel deep learning-based approach using peripheral area projection, which is the first attempt to analyze LV considering its surrounding areas. Our approach matches the baseline LV mesh by deforming the shape of follow-up LVs, while optimizing the corresponding points of the same adjacent brain area between the baseline and follow-up LVs. Furthermore, we quantitatively evaluated the deformation of the left LV in normal (n=10) and demented subjects (n=10), and we found that each surrounding area (thalamus, caudate, hippocampus, amygdala, and right LV) projected onto the surface of LV shows noticeable differences between normal and demented subjects., Comment: Annual Conference on Medical Image Understanding and Analysis (2024)

The demand for reliable AI systems has intensified the need for interpretable deep neural networks. Concept bottleneck models (CBMs) have gained attention as an effective approach by leveraging human-understandable concepts to enhance interpretability. However, existing CBMs face challenges due to deterministic concept encoding and reliance on inconsistent concepts, leading to inaccuracies. We propose EQ-CBM, a novel framework that enhances CBMs through probabilistic concept encoding using energy-based models (EBMs) with quantized concept activation vectors (qCAVs). EQ-CBM effectively captures uncertainties, thereby improving prediction reliability and accuracy. By employing qCAVs, our method selects homogeneous vectors during concept encoding, enabling more decisive task performance and facilitating higher levels of human intervention. Empirical results using benchmark datasets demonstrate that our approach outperforms the state-of-the-art in both concept and task accuracy., Comment: Accepted by ACCV 2024

Given a set P of n points in the plane, the two-line center problem asks to find two lines that minimize the maximum distance from each point in P to its closer one of the two resulting lines. The currently best algorithm for the problem takes $O(n^2\log^2n)$ time by Jaromczyk and Kowaluk in 1995. In this paper, we present faster algorithms for three variants of the two-line center problem in which the orientations of the resulting lines are constrained. Specifically, our algorithms solve the problem in $O(n \log n)$ time when the orientations of both lines are fixed; in $O(n \log^3 n)$ time when the orientation of one line is fixed; and in $O(n^2 \alpha(n) \log n)$ time when the angle between the two lines is fixed, where $\alpha(n)$ denotes the inverse Ackermann function.

We propose MR HuBo(Motion Retargeting leveraging a HUman BOdy prior), a cost-effective and convenient method to collect high-quality upper body paired pose data, which is essential for data-driven motion retargeting methods. Unlike existing approaches which collect pose data by converting human MoCap poses into robot poses, our method goes in reverse. We first sample diverse random robot poses, and then convert them into human poses. However, since random robot poses can result in extreme and infeasible human poses, we propose an additional technique to sort out extreme poses by exploiting a human body prior trained from a large amount of human pose data. Our data collection method can be used for any humanoid robots, if one designs or optimizes the system's hyperparameters which include a size scale factor and the joint angle ranges for sampling. In addition to this data collection method, we also present a two-stage motion retargeting neural network that can be trained via supervised learning on a large amount of paired data. Compared to other learning-based methods trained via unsupervised learning, we found that our deep neural network trained with ample high-quality paired data achieved notable performance. Our experiments also show that our data filtering method yields better retargeting results than training the model with raw and noisy data. Our code and video results are available on https://sites.google.com/view/mr-hubo/, Comment: 8 pages, 5 Figures, Accepted at IROS 2024

In this study, we present a novel method for enhancing the computational efficiency of whole-body control for humanoid robots, a challenge accentuated by their high degrees of freedom. The reduced-dimension rigid body dynamics of a floating base robot is constructed by segmenting its kinematic chain into constrained and unconstrained chains, simplifying the dynamics of the unconstrained chain through the centroidal dynamics. The proposed dynamics model is possible to be applied to whole-body control methods, allowing the problem to be divided into two parts for more efficient computation. The efficiency of the framework is demonstrated by comparative experiments in simulations. The calculation results demonstrate a significant reduction in processing time, highlighting an improvement over the times reported in current methodologies. Additionally, the results also shows the computational efficiency increases as the degrees of freedom of robot model increases., Comment: submitted to RA-L, under review

This paper introduces ES-KT-24, a novel multimodal Knowledge Tracing (KT) dataset for intelligent tutoring systems in educational game contexts. Although KT is crucial in adaptive learning, existing datasets often lack game-based and multimodal elements. ES-KT-24 addresses these limitations by incorporating educational game-playing videos, synthetically generated question text, and detailed game logs. The dataset covers Mathematics, English, Indonesian, and Malaysian subjects, emphasizing diversity and including non-English content. The synthetic text component, generated using a large language model, encompasses 28 distinct knowledge concepts and 182 questions, featuring 15,032 users and 7,782,928 interactions. Our benchmark experiments demonstrate the dataset's utility for KT research by comparing Deep learning-based KT models with Language Model-based Knowledge Tracing (LKT) approaches. Notably, LKT models showed slightly higher performance than traditional DKT models, highlighting the potential of language model-based approaches in this field. Furthermore, ES-KT-24 has the potential to significantly advance research in multimodal KT models and learning analytics. By integrating game-playing videos and detailed game logs, this dataset offers a unique approach to dissecting student learning patterns through advanced data analysis and machine-learning techniques. It has the potential to unearth new insights into the learning process and inspire further exploration in the field., Comment: 11 pages, 5 figures

Recent advances in sensing and computer vision (CV) technologies have opened the door for the application of deep learning (DL)-based CV technologies in the realm of 6G wireless communications. For the successful application of this emerging technology, it is crucial to have a qualified vision dataset tailored for wireless applications (e.g., RGB images containing wireless devices such as laptops and cell phones). An aim of this paper is to propose a large-scale vision dataset referred to as Vision Objects for Millimeter and Terahertz Communications (VOMTC). The VOMTC dataset consists of 20,232 pairs of RGB and depth images obtained from a camera attached to the base station (BS), with each pair labeled with three representative object categories (person, cell phone, and laptop) and bounding boxes of the objects. Through experimental studies of the VOMTC datasets, we show that the beamforming technique exploiting the VOMTC-trained object detector outperforms conventional beamforming techniques.

We perform angle-resolved photoemission spectroscopy (ARPES) measurements in conjunction with density functional theory (DFT) calculations to investigate the evolution of the electronic structure of CeBi upon a series of antiferromagnetic (AFM) transitions. We find evidence for a new AFM transition in addition to two previously known from transport studies. We demonstrate the development of an additional Dirac state in the (+-+-) ordered phase and a transformation of unconventional surface-state pairs in the (++--) ordered phase. This revises the phase diagram of this intriguing material, where there are now three distinct AFM states below TN in zero magnetic field instead of two as it was previously thought., Comment: 16 pages, 4 figures

Photonic lanterns are waveguide devices enabling high throughput single mode spectroscopy and high angular resolution. We aim to present the first on-sky demonstration of a photonic lantern (PL) operating in visible light, to measure its throughput and assess its potential for high-resolution spectroscopy of compact objects. We used the SCExAO instrument (a double stage extreme AO system installed at the Subaru telescope) and FIRST mid-resolution spectrograph (R 3000) to test the visible capabilities of the PL on internal source and on-sky observations. The best averaged coupling efficiency over the PL field of view was measured at 51% +/- 10% with a peak at 80%. We also investigate the relationship between coupling efficiency and the Strehl ratio for a PL, comparing them with those of a single-mode fiber (SMF). Findings show that in the AO regime, a PL offers better coupling efficiency performance than a SMF, especially in the presence of low spatial frequency aberrations. We observed Ikiiki (alpha Leo - mR = 1.37) and `Aua (alpha Ori - mR = -1.17) at a frame rate of 200 Hz. Under median seeing conditions (about 1 arcsec measured in H band) and large tip/tilt residuals (over 20 mas), we estimated an average light coupling efficiency of 14.5% +/- 7.4%, with a maximum of 42.8% at 680 nm. We were able to reconstruct both star's spectra, containing various absorption lines. The successful demonstration of this device opens new possibilities in terms of high throughput single-mode fiber-fed spectroscopy in the Visible. The demonstrated on-sky coupling efficiency performance would not have been achievable with a single SMF injection setup under similar conditions, partly because the residual tip/tilt alone exceeded the field of view of a visible SMF (18 mas at 700 nm). Thus emphasizing the enhanced resilience of PL technology to such atmospheric disturbances. The additional, Comment: Accepted in Astronomy and Astrophysics journal on 9/11/2024

Recommender systems have become integral to digital experiences, shaping user interactions and preferences across various platforms. Despite their widespread use, these systems often suffer from algorithmic biases that can lead to unfair and unsatisfactory user experiences. This study introduces an interactive tool designed to help users comprehend and explore the impacts of algorithmic harms in recommender systems. By leveraging visualizations, counterfactual explanations, and interactive modules, the tool allows users to investigate how biases such as miscalibration, stereotypes, and filter bubbles affect their recommendations. Informed by in-depth user interviews, this tool benefits both general users and researchers by increasing transparency and offering personalized impact assessments, ultimately fostering a better understanding of algorithmic biases and contributing to more equitable recommendation outcomes. This work provides valuable insights for future research and practical applications in mitigating bias and enhancing fairness in machine learning algorithms.