Search

Flying quadrotors in tight formations is a challenging problem. It is known that in the near-field airflow of a quadrotor, the aerodynamic effects induced by the propellers are complex and difficult to characterize. Although machine learning tools can potentially be used to derive models that capture these effects, these data-driven approaches can be sample inefficient and the resulting models often do not generalize as well as their first-principles counterparts. In this work, we propose a framework that combines the benefits of first-principles modeling and data-driven approaches to construct an accurate and sample efficient representation of the complex aerodynamic effects resulting from quadrotors flying in formation. The data-driven component within our model is lightweight, making it amenable for optimization-based control design. Through simulations and physical experiments, we show that incorporating the model into a novel learning-based nonlinear model predictive control (MPC) framework results in substantial performance improvements in terms of trajectory tracking and disturbance rejection. In particular, our framework significantly outperforms nominal MPC in physical experiments, achieving a 40.1% improvement in the average trajectory tracking errors and a 57.5% reduction in the maximum vertical separation errors. Our framework also achieves exceptional sample efficiency, using only a total of 46 seconds of flight data for training across both simulations and physical experiments. Furthermore, with our proposed framework, the quadrotors achieve an exceptionally tight formation, flying with an average separation of less than 1.5 body lengths throughout the flight. A video illustrating our framework and physical experiments is given here: https://youtu.be/Hv-0JiVoJGo, Comment: 7 pages, 5 figures

Continuum robots, known for their high flexibility and adaptability, offer immense potential for applications such as medical surgery, confined-space inspections, and wearable devices. However, their non-linear elastic properties and complex kinematics present significant challenges in digital modeling and effective control. This research proposes a novel computational framework that integrates Lie group kinematics with an evolutionary algorithm (EA) to identify optimal control coefficients for specific robot models. Our method starts by generating datasets from physics-based simulations and fractional order control, defining both ideal configurations and models to be optimized. By using EA, we iteratively minimize deviations through two fitness objectives \textemdash deviation mean squared error (\(\text{MSE}_1\)) and TCP vector error (\(\text{MSE}_2\)) \textemdash to align the robot's backbone with the desired configuration. Built on the Computer-Aided Design (CAD) platform Grasshopper, this framework provides real-time visualization, enabling dynamic control of robot configurations. Results show that the proposed method achieves precise alignment of the robot's backbone with minimal computation. This approach not only simplifies the coefficient identification process but also demonstrates the advantages of EA in multi-objective optimization, contributing to efficient modeling and control of continuum robots., Comment: 10 pages, 20 figures

We utilize serendipitous observations from the Asteroid Terrestrial-impact Last Alert System (ATLAS) and the Zwicky Transient Facility (ZTF) in addition to targeted follow-up observations from the Las Cumbres Observatory (LCO) and Liverpool Telescope to analyze the first observed instance of cometary activity by the newly-discovered Jupiter-family comet C/2023 RN3 (ATLAS), whose orbital dynamics place it close to residing on a Centaur-like orbit. Across our 7-month baseline, we observe an epoch of cometary activity commencing in August 2023 with an increase in brightness of >5.4 mag. The lightcurve of 2023 RN3 indicates the presence of continuous cometary activity across our observations, suggesting the onset of a new period of sustained activity. We find no evidence of any outbursts on top of the observed brightening, nor do we find any significant color evolution across our observations. 2023 RN3 is visibly extended in LCO and Liverpool Telescope observations, indicating the presence of a spatially-extended coma. Numerical integration of 2023 RN3's orbit reveals the comet to have recently undergone a slight increase in semimajor axis due to a planetary encounter with Jupiter, however whether this orbital change could trigger 2023 RN3's cometary activity is unclear. Our estimate for the maximum dust production metric of Afrho ~400 cm is consistent with previous measurements for the Jupiter-family comet and Centaur populations., Comment: 20 pages, 9 figures

Entanglement is an essential resource for various quantum-information tasks. When a target system shares entanglement with another memory system and is stored reliably, one can use entanglement at a later time -- this is quantum memory. In practice, entanglement can be exceedingly fragile during a system's evolution. In particular, no entanglement can survive when a so-called entanglement-breaking channel acts on the target system. Are entanglement-breaking channels really useless for maintaining entanglement? As a single channel, this is certainly the case; it cannot be useful for quantum memory. However, in this work, we show that putting together two entanglement-breaking channels in a broadcasting scenario can activate their ability to maintain entanglement -- the channel's quantum memory resource can be super-activated., Comment: 4+3 pages, 2 figures

We analyze contextual representations in neural autoregressive language models, emphasizing long-range contexts that span several thousand tokens. Our methodology employs a perturbation setup and the metric \emph{Anisotropy-Calibrated Cosine Similarity}, to capture the degree of contextualization of long-range patterns from the perspective of representation geometry. We begin the analysis with a case study on standard decoder-only Transformers, demonstrating that similar perplexity can exhibit markedly different downstream task performance, which can be explained by the difference in contextualization of long-range content. Next, we extend the analysis to other models, covering recent novel architectural designs and various training configurations. The representation-level results illustrate a reduced capacity for high-complexity (i.e., less compressible) sequences across architectures, and that fully recurrent models rely heavily on local context, whereas hybrid models more effectively encode the entire sequence structure. Finally, preliminary analysis of model size and training configurations on the encoding of long-range context suggest potential directions for improving existing language models., Comment: 17 pages, 9 figures

The multiplicities of positive and negative pions, kaons and unidentified hadrons produced in deep-inelastic scattering are measured in bins of the Bjorken scaling variable $x$, the relative virtual-photon energy $y$ and the fraction of the virtual-photon energy transferred to the final-state hadron $z$. Data were obtained by the COMPASS Collaboration using a 160 GeV muon beam of both electric charges and a liquid hydrogen target. These measurements cover the kinematic domain with photon virtuality $Q^2 > 1$ (GeV/$c)^2$, $0.004 < x < 0.4$, $0.1 < y < 0.7$ and $0.2 < z < 0.85$, in accordance with the kinematic domain used in earlier published COMPASS multiplicity measurements with an isoscalar target. The calculation of radiative corrections was improved by using the Monte Carlo generator DJANGOH, which results in up to 12\% larger corrections in the low-$x$ region., Comment: 19 pages, 29 figures

Quantum-centric supercomputing presents a compelling framework for large-scale hybrid quantum-classical tasks. Although quantum machine learning (QML) offers theoretical benefits in various applications, challenges such as large-size data encoding in the input stage and the reliance on quantum resources in the inference stage limit its practicality for tasks like fine-tuning large language models (LLMs). Quantum parameter generation, a novel approach of QML, addresses these limitations by using quantum neural networks (QNNs) to generate classical model weights (parameters) exclusively during training, thereby decoupling inference from quantum hardware. In this work, we introduce Quantum Parameter Adaptation (QPA) in the framework of quantum parameter generation, which integrates QNNs with a classical multi-layer perceptron mapping model to generate parameters for fine-tuning methods. Using Gemma-2 and GPT-2 as case studies, QPA demonstrates significant parameter reduction for parameter-efficient fine-tuning methods, such as Low-Rank Adaptation (LoRA), while maintaining comparable or improved performance in text generation tasks. Specifically, QPA reduces the number of parameters to $52.06\%$ of the original LoRA for GPT-2 with a slight performance gain of $0.75\%$, and to $16.84\%$ for Gemma-2, with a marginal performance improvement of $0.07\%$. These results highlight QPA's ability to achieve efficient parameter reduction without sacrificing performance in the quantum parameter generation framework. This work showcases the potential of quantum-enhanced parameter reduction, offering a scalable quantum-classical solution for fine-tuning LLMs while preserving the feasibility of inference on classical hardware., Comment: 21 pages, 6 figures

The magnetar SGR 1935+2154 is the only known Galactic source of fast radio bursts (FRBs). FRBs from SGR 1935+2154 were first detected by CHIME/FRB and STARE2 in 2020 April, after the conclusion of the LIGO, Virgo, and KAGRA Collaborations' O3 observing run. Here we analyze four periods of gravitational wave (GW) data from the GEO600 detector coincident with four periods of FRB activity detected by CHIME/FRB, as well as X-ray glitches and X-ray bursts detected by NICER and NuSTAR close to the time of one of the FRBs. We do not detect any significant GW emission from any of the events. Instead, using a short-duration GW search (for bursts $\leq$ 1 s) we derive 50\% (90\%) upper limits of $10^{48}$ ($10^{49}$) erg for GWs at 300 Hz and $10^{49}$ ($10^{50}$) erg at 2 kHz, and constrain the GW-to-radio energy ratio to $\leq 10^{14} - 10^{16}$. We also derive upper limits from a long-duration search for bursts with durations between 1 and 10 s. These represent the strictest upper limits on concurrent GW emission from FRBs., Comment: 15 pages of text including references, 4 figures, 5 tables

Diffusion and flow-matching models achieve remarkable generative performance but at the cost of many sampling steps, this slows inference and limits applicability to time-critical tasks. The ReFlow procedure can accelerate sampling by straightening generation trajectories. However, ReFlow is an iterative procedure, typically requiring training on simulated data, and results in reduced sample quality. To mitigate sample deterioration, we examine the design space of ReFlow and highlight potential pitfalls in prior heuristic practices. We then propose seven improvements for training dynamics, learning and inference, which are verified with thorough ablation studies on CIFAR10 $32 \times 32$, AFHQv2 $64 \times 64$, and FFHQ $64 \times 64$. Combining all our techniques, we achieve state-of-the-art FID scores (without / with guidance, resp.) for fast generation via neural ODEs: $2.23$ / $1.98$ on CIFAR10, $2.30$ / $1.91$ on AFHQv2, $2.84$ / $2.67$ on FFHQ, and $3.49$ / $1.74$ on ImageNet-64, all with merely $9$ neural function evaluations.

Network pruning has emerged as a potential solution to make LLMs cheaper to deploy. However, existing LLM pruning approaches universally rely on the C4 dataset as the calibration data for calculating pruning scores, leaving its optimality unexplored. In this study, we evaluate the choice of calibration data on LLM pruning, across a wide range of datasets that are most commonly used in LLM training and evaluation, including four pertaining datasets as well as three categories of downstream tasks encompassing nine datasets. Each downstream dataset is prompted with In-Context Learning (ICL) and Chain-of-Thought (CoT), respectively. Besides the already intriguing observation that the choice of calibration data significantly impacts the performance of pruned LLMs, our results also uncover several subtle and often unexpected findings, summarized as follows: (1) C4 is not the optimal choice for LLM pruning, even among commonly used pre-training datasets; (2) arithmetic datasets, when used as calibration data, performs on par or even better than pre-training datasets; (3) pruning with downstream datasets does not necessarily help the corresponding downstream task, compared to pre-training data; (4) ICL is widely beneficial to all data categories, whereas CoT is only useful on certain tasks. Our findings shed light on the importance of carefully selecting calibration data for LLM pruning and pave the way for more efficient deployment of these powerful models in real-world applications. We release our code at: https://github.com/abx393/llm-pruning-calibration-data., Comment: EMNLP 2024

The piecewise-stationary bandit problem is an important variant of the multi-armed bandit problem that further considers abrupt changes in the reward distributions. The main theme of the problem is the trade-off between exploration for detecting environment changes and exploitation of traditional bandit algorithms. While this problem has been extensively investigated, existing works either assume knowledge about the number of change points $M$ or require extremely high computational complexity. In this work, we revisit the piecewise-stationary bandit problem from a minimalist perspective. We propose a novel and generic exploration mechanism, called diminishing exploration, which eliminates the need for knowledge about $M$ and can be used in conjunction with an existing change detection-based algorithm to achieve near-optimal regret scaling. Simulation results show that despite oblivious of $M$, equipping existing algorithms with the proposed diminishing exploration generally achieves better empirical regret than the traditional uniform exploration.

Understanding the computational complexity of learning efficient classical programs in various learning models has been a fundamental and important question in classical computational learning theory. In this work, we study the computational complexity of quantum state learning, which can be seen as a quantum generalization of distributional learning introduced by Kearns et.al [STOC94]. Previous works by Chung and Lin [TQC21], and B\u{a}descu and O$'$Donnell [STOC21] study the sample complexity of the quantum state learning and show that polynomial copies are sufficient if unknown quantum states are promised efficiently generatable. However, their algorithms are inefficient, and the computational complexity of this learning problem remains unresolved. In this work, we study the computational complexity of quantum state learning when the states are promised to be efficiently generatable. We show that if unknown quantum states are promised to be pure states and efficiently generateable, then there exists a quantum polynomial time algorithm $A$ and a language $L \in PP$ such that $A^L$ can learn its classical description. We also observe the connection between the hardness of learning quantum states and quantum cryptography. We show that the existence of one-way state generators with pure state outputs is equivalent to the average-case hardness of learning pure states. Additionally, we show that the existence of EFI implies the average-case hardness of learning mixed states., Comment: 24 pages

We propose a novel neural network architecture, the normalized Transformer (nGPT) with representation learning on the hypersphere. In nGPT, all vectors forming the embeddings, MLP, attention matrices and hidden states are unit norm normalized. The input stream of tokens travels on the surface of a hypersphere, with each layer contributing a displacement towards the target output predictions. These displacements are defined by the MLP and attention blocks, whose vector components also reside on the same hypersphere. Experiments show that nGPT learns much faster, reducing the number of training steps required to achieve the same accuracy by a factor of 4 to 20, depending on the sequence length.

The study of magnetoresistance (MR) phenomena has been pivotal in advancing magnetic sensors and spintronic devices. Helimagnets present an intriguing avenue for spintronics research. Theoretical predictions suggest that MR magnitude in the helimagnetic (HM) regime surpasses that in the ferromagnetic (FM) regime by over an order of magnitude. However, in metallic helimagnets like manganese phosphide, MR in the HM phase remains modest (10%), limiting its application in MR devices. Here, a groundbreaking approach is presented to achieve a giant low field MR effect in nanostructured manganese phosphide films by leveraging confinement and strain effects along with spin helicity. Unlike the modest MR observed in bulk manganese phosphide single crystals and large grain polycrystalline films, which exhibit a small negative MR in the FM region (2%) increasing to 8% in the HM region across 10-300 K, a grain size-dependent giant positive MR (90%) is discovered near FM to HM transition temperature (110 K), followed by a rapid decline to a negative MR below 55 K in manganese phosphide nanocrystalline films. These findings illuminate a novel strain-mediated spin helicity phenomenon in nanostructured helimagnets, presenting a promising pathway for the development of high-performance MR sensors and spintronic devices through the strategic utilization of confinement and strain effects.

Traditionally, unmanned aerial vehicles (UAVs) rely on CMOS-based cameras to collect images about the world below. One of the most successful applications of UAVs is to generate orthomosaics or orthomaps, in which a series of images are integrated together to develop a larger map. However, the use of CMOS-based cameras with global or rolling shutters mean that orthomaps are vulnerable to challenging light conditions, motion blur, and high-speed motion of independently moving objects under the camera. Event cameras are less sensitive to these issues, as their pixels are able to trigger asynchronously on brightness changes. This work introduces the first orthomosaic approach using event cameras. In contrast to existing methods relying only on CMOS cameras, our approach enables map generation even in challenging light conditions, including direct sunlight and after sunset., Comment: 7 pages, 7 figures

Several benchmarks have concluded that our best vision-language models (e.g., CLIP) are lacking in compositionality. Given an image, these benchmarks probe a model's ability to identify its associated caption amongst a set of compositional distractors. In response, a surge of recent proposals show improvements by finetuning CLIP with distractors as hard negatives. Our investigations reveal that these improvements have, in fact, been significantly overstated -- because existing benchmarks do not probe whether finetuned vision-language models remain invariant to hard positives. By curating an evaluation dataset with 112,382 hard negatives and hard positives, we uncover that including hard positives decreases CLIP's performance by 12.9%, while humans perform effortlessly at 99%. CLIP finetuned with hard negatives results in an even larger decrease, up to 38.7%. With this finding, we then produce a 1,775,259 image-text training set with both hard negative and hard positive captions. By training with both, we see improvements on existing benchmarks while simultaneously improving performance on hard positives, indicating a more robust improvement in compositionality. Our work suggests the need for future research to rigorously test and improve CLIP's understanding of semantic relationships between related "positive" concepts., Comment: ECCV 2024

We study the strong gravitational lensing by the Kerr-like wormholes with an additional parameter to specify the location of the throat. We classify the roots of the radial potential derived from the null geodesic equations. We focus on the throat together with other roots to become either double root or triple root, potentially giving the divergence of the deflection angle of the light rays in the strong deflection limit (SDL). In particular, while the logarithmic divergence is known as the double roots are approached, the more stronger power-law (non-logarithmic) divergence is found for the triple roots. In addition, the effective potential in terms of the proper distance from the throat is constructed with which to realize how the light rays can either travel within one spacetime, where the observers are located or pass through the throat into another spacetime, where different observers reside. The observational effects, such as relativistic images resulting from the deflection of light by wormholes, are discussed., Comment: 39 pages, 11 figures, 4 tables

Coordinating the motion of multiple robots in cluttered environments remains a computationally challenging task. We study the problem of minimizing the execution time of a set of geometric paths by a team of robots with state-dependent actuation constraints. We propose a Time-Optimal Path Parameterization (TOPP) algorithm for multiple car-like agents, where the modulation of the timing of every robot along its assigned path is employed to ensure collision avoidance and dynamic feasibility. This is achieved through the use of a priority queue to determine the order of trajectory execution for each robot while taking into account all possible collisions with higher priority robots in a spatiotemporal graph. We show a 10-20% reduction in makespan against existing state-of-the-art methods and validate our approach through simulations and hardware experiments.

Continuum robots, which often rely on interdisciplinary and multimedia collaborations, have been increasingly recognized for their potential to revolutionize the field of human-computer interaction (HCI) in varied applications due to their adaptive, responsive, and flexible characteristics. Despite their promises, the lack of an integrated framework poses a significant limitation for both users and developers, resulting in inefficiency and complexity during preliminary developments. Thus, this paper introduces a unified framework for continuum robotic systems that addresses these challenges by integrating system architecture, dynamics computation, and control strategy within a computer-aided design (CAD) platform. The proposed method allows for efficient modeling and quick preview of the robot performance, and thus facilitating iterative design and implementation, with a view to enhancing the quality of robot developments., Comment: 9 pages, 10 figures, 1 table

Community and organizational policies are typically designed in a top-down, centralized fashion, with limited input from impacted stakeholders. This can result in policies that are misaligned with community needs or perceived as illegitimate. How can we support more collaborative, participatory approaches to policy design? In this paper, we present PolicyCraft, a system that structures collaborative policy design through case-grounded deliberation. Building on past research that highlights the value of concrete cases in establishing common ground, PolicyCraft supports users in collaboratively proposing, critiquing, and revising policies through discussion and voting on cases. A field study across two university courses showed that students using PolicyCraft reached greater consensus and developed better-supported course policies, compared with those using a baseline system that did not scaffold their use of concrete cases. Reflecting on our findings, we discuss opportunities for future HCI systems to help groups more effectively bridge between abstract policies and concrete cases.

This paper presents our experience with TREBLE, a two-year initiative to build the modular base in Android, a Java-based mobile platform running on the Linux kernel. Our TREBLE architecture splits the hardware independent core framework written in Java from the hardware dependent vendor implementations (e.g., user space device drivers, vendor native libraries, and kernel written in C/C++). Cross-layer communications between them are done via versioned, stable inter-process communication interfaces whose backward compatibility is tested by using two API compliance suites. Based on this architecture, we repackage the key Android software components that suffered from crucial post-launch security bugs as separate images. That not only enables separate ownerships but also independent updates of each image by interested ecosystem entities. We discuss our experience of delivering TREBLE architectural changes to silicon vendors and device makers using a yearly release model. Our experiments and industry rollouts support our hypothesis that giving more freedom to all ecosystem entities and creating an equilibrium are a transformation necessary to further scale the world largest open ecosystem with over two billion active devices., Comment: \c{opyright} K. S. Yim et al. | ACM 2019. This is the author's version of the work. It is posted here for your personal use. Not for redistribution. The definitive Version of Record was published in ACM TECS, https://doi.org/10.1145/3358237

Informational contributions to thermodynamics can be studied in isolation by considering systems with fully-degenerate Hamiltonians. In this regime, being in non-equilibrium -- termed informational non-equilibrium -- provides thermodynamic resources, such as extractable work, solely from the information content. The usefulness of informational non-equilibrium creates an incentive to obtain more of it, motivating the question of how to concentrate it: can we increase the local informational non-equilibrium of a product state $\rho\otimes\rho$ under a global closed system (unitary) evolution? We fully solve this problem analytically, showing that it is impossible for two-qubits, and it is always possible to find states achieving this in higher dimensions. The notion of bound resources in this framework is then discussed, along with initial global correlations' ability to activate concentration. Finally, we apply our results to study the concentration of purity and intrinsic randomness., Comment: 5+2 pages, 2 figures

In real-world applications with Large Language Models (LLMs), external retrieval mechanisms - such as Search-Augmented Generation (SAG), tool utilization, and Retrieval-Augmented Generation (RAG) - are often employed to enhance the quality of augmented generations in dialogues. These approaches often come with multi-turn dialogue, where each interaction is enriched by relevant information retrieved from external sources. Existing benchmarks either assess LLMs' chat abilities in multi-turn dialogues or their use of retrieval for augmented responses in single-turn settings. However, there is a gap in evaluating LLMs' ability to leverage retrieval for more precise responses across multiple turns. To address this limitation, we introduce RAD-Bench (Retrieval Augmented Dialogue), a benchmark designed to evaluate LLMs' capabilities in multi-turn dialogues following retrievals, essential for their deployment in context-rich applications. RAD-Bench evaluates two key abilities of LLMs: Retrieval Synthesis and Retrieval Reasoning. These are measured using discriminative questions and retrieved contexts, and corresponding reference answers, assessing how effectively LLMs integrate and reason with context to maintain and enhance conversation quality over multiple turns. Our evaluation results on commonly used LLMs reveal that model performance deteriorates as additional layers of conditions or constraints are applied across conversation turns, even when accurate retrieved contexts are provided.

In recent years, Couder and collaborators have initiated a series of studies on walking droplets. Experimentally, they found that at frequencies and amplitudes close to the onset of Faraday waves, droplets on the surface of silicone oil can survive and walk at a roughly constant speed due to resonance. Droplets excite local ripples from the Faraday instability when they bounce from the liquid surface. This tightly coupled particle-wave entity, although a complex yet entirely classical system, exhibits many phenomena that are strikingly similar to those of quantum systems, such as slit interference and diffraction, tunneling probability, and Anderson localization. In this Letter, we focus on the tunneling time of droplets. Specifically, we explore (1) how it changes with the width of an acrylic barrier, which gives rise to the potential barrier when the depth of the silicone oil is reduced to prevent the generation of ripples that can feed energy back to the droplet, and (2) the distribution of tunneling times at the same barrier width. Both results turn out to be similar to the numerical outcome of the Bohmian mechanics, which strengthens the analogy to a quantum system. Furthermore, we successfully derive analytic expressions for these properties by revising the multiple scattering theory and constructing a ``skipping stone" model. Provided that the resemblance in tunneling behavior of walking droplets to Bohmian particles is not coincidental, we discuss the lessons for the Copenhagen interpretation of quantum mechanics that so far fails to explain both characteristics adequately.

We studied an interesting experiment that showed a half-buried chopstick lifting a full bottle of granules off the table. In Janssen theory, the friction force provided by the container wall helps alleviate the weight of the granules. How can a thin rod with a much less contact area support the full weight plus that of the container? Insights are gained by allowing the friction on the wall to change direction before solving the Janssen equation. We obtained the analytic expression for the critical depth of granules that enables a successful lift off. In addition, we established that the stick and slip phenomenon exists during a failed lift off by analyzing the frequency of fluctuations in the pull force. Finally, a photoelasticity experiment was employed to directly visualize the stress field sensitive to the pull force, and verify the directional change of friction force from the wall., Comment: 6 pages, 4 figures

The ability to slow down light at the single-photon level has applications in quantum information processing and other quantum technologies. We demonstrate two methods, both using just a single artificial atom, enabling dynamic control over microwave light velocities in waveguide quantum electrodynamics (waveguide QED). Our methods are based on two distinct mechanisms harnessing the balance between radiative and non-radiative decay rates of a superconducting artificial atom in front of a mirror. In the first method, we tune the radiative decay of the atom using interference effects due to the mirror; in the second method, we pump the atom to control its non-radiative decay through the Autler--Townes effect. When the half the radiative decay rate exceeds the non-radiative decay rate, we observe positive group delay; conversely, dominance of the non-radiative decay rate results in negative group delay. Our results advance signal-processing capabilities in waveguide QED.

Large Language Model (LLM) pretraining traditionally relies on autoregressive language modeling on randomly sampled data blocks from web-scale datasets. We take inspiration from human learning techniques like spaced repetition to hypothesize that random data sampling for LLMs leads to high training cost and low quality models which tend to forget data. In order to effectively commit web-scale information to long-term memory, we propose the LFR (Learn, Focus, and Review) pedagogy, a new dynamic training paradigm which focuses and repeatedly reviews complex data blocks at systematic intervals based on the model's learning pace and progress. LFR records the model perplexities for different data blocks and frequently revisits blocks with higher perplexity which are more likely to be forgotten. We pretrain the GPT-2 models (124M - 1.5B) from scratch on the OpenWebText dataset using LFR. We test on downstream tasks from the language modeling, question answering, translation, and problem solving domains to achieve consistently lower perplexity and higher accuracy than the baseline OpenAI models, while obtaining a 20x pretraining speed-up.

In the realm of drug discovery, DNA-encoded library (DEL) screening technology has emerged as an efficient method for identifying high-affinity compounds. However, DEL screening faces a significant challenge: noise arising from nonspecific interactions within complex biological systems. Neural networks trained on DEL libraries have been employed to extract compound features, aiming to denoise the data and uncover potential binders to the desired therapeutic target. Nevertheless, the inherent structure of DEL, constrained by the limited diversity of building blocks, impacts the performance of compound encoders. Moreover, existing methods only capture compound features at a single level, further limiting the effectiveness of the denoising strategy. To mitigate these issues, we propose a Multimodal Pretraining DEL-Fusion model (MPDF) that enhances encoder capabilities through pretraining and integrates compound features across various scales. We develop pretraining tasks applying contrastive objectives between different compound representations and their text descriptions, enhancing the compound encoders' ability to acquire generic features. Furthermore, we propose a novel DEL-fusion framework that amalgamates compound information at the atomic, submolecular, and molecular levels, as captured by various compound encoders. The synergy of these innovations equips MPDF with enriched, multi-scale features, enabling comprehensive downstream denoising. Evaluated on three DEL datasets, MPDF demonstrates superior performance in data processing and analysis for validation tasks. Notably, MPDF offers novel insights into identifying high-affinity molecules, paving the way for improved DEL utility in drug discovery.

Abridged: Very Low Luminosity Objects (VeLLOs) are deeply embedded, and extremely faint objects and are thought to be in the quiescent phase of the episodic accretion process. They fill an important gap in our understanding of star formation. The VeLLO in the isolated DC3272+18 cloud has undergone an outburst, and is thus an ideal target for investigating the chemical inventory in the gas phase of an object of its type. Observations with the Atacama Pathfinder EXperiment (APEX) in four spectral windows in the frequency range of 213.6--272.4~GHz have been carried out to identify molecules that can be directly linked to the past outburst, utilize the line fluxes, column densities, and the abundance ratios of the detected species to characterize the different physical components of the VeLLO, and probe for the presence of complex organic molecules. Nitric oxide (NO) is detected for the first time in a source of this type, and its formation could be induced by the sublimation of grain-surface species during the outburst. A pathway to form NO directly in the gas phase is from the photodissociation products created after the sublimation of H$_2$O and NH$_3$ from the ices. While the present time water snowline has likely retreated to pre-outburst small radius, the volatile NO species is still extensively present in the gas phase, as evident by its high column density relative to methanol in the observations. This suggests that NO could be potentially used to trace the water snowline in outbursting sources. In order to rule out non-thermal desorption processes that could also have led to the formation of NO, this proposition has to be verified with future observations at higher spatial resolution, and by searching for NO in additional targets., Comment: 19 pages, 5 figures, 5 tables. Accepted for publication in A&A

We investigate the problem of predicting the output behavior of unknown quantum channels. Given query access to an $n$-qubit channel $E$ and an observable $O$, we aim to learn the mapping \begin{equation*} \rho \mapsto \mathrm{Tr}(O E[\rho]) \end{equation*} to within a small error for most $\rho$ sampled from a distribution $D$. Previously, Huang, Chen, and Preskill proved a surprising result that even if $E$ is arbitrary, this task can be solved in time roughly $n^{O(\log(1/\epsilon))}$, where $\epsilon$ is the target prediction error. However, their guarantee applied only to input distributions $D$ invariant under all single-qubit Clifford gates, and their algorithm fails for important cases such as general product distributions over product states $\rho$. In this work, we propose a new approach that achieves accurate prediction over essentially any product distribution $D$, provided it is not "classical" in which case there is a trivial exponential lower bound. Our method employs a "biased Pauli analysis," analogous to classical biased Fourier analysis. Implementing this approach requires overcoming several challenges unique to the quantum setting, including the lack of a basis with appropriate orthogonality properties. The techniques we develop to address these issues may have broader applications in quantum information., Comment: 20 pages, comments welcome

Large Language Models (LLMs) have demonstrated remarkable proficiency in various natural language generation (NLG) tasks. Previous studies suggest that LLMs' generation process involves uncertainty. However, existing approaches to uncertainty estimation mainly focus on sequence-level uncertainty, overlooking individual pieces of information within sequences. These methods fall short in separately assessing the uncertainty of each component in a sequence. In response, we propose a novel framework for Concept-Level Uncertainty Estimation (CLUE) for LLMs. We leverage LLMs to convert output sequences into concept-level representations, breaking down sequences into individual concepts and measuring the uncertainty of each concept separately. We conduct experiments to demonstrate that CLUE can provide more interpretable uncertainty estimation results compared with sentence-level uncertainty, and could be a useful tool for various tasks such as hallucination detection and story generation.

Platform-based laborers face unprecedented challenges and working conditions that result from algorithmic opacity, insufficient data transparency, and unclear policies and regulations. The CSCW and HCI communities increasingly turn to worker data collectives as a means to advance related policy and regulation, hold platforms accountable for data transparency and disclosure, and empower the collective worker voice. However, fundamental questions remain for designing, governing and sustaining such data infrastructures. In this workshop, we leverage frameworks such as data feminism to design sustainable and power-aware data collectives that tackle challenges present in various types of online labor platforms (e.g., ridesharing, freelancing, crowdwork, carework). While data collectives aim to support worker collectives and complement relevant policy initiatives, the goal of this workshop is to encourage their designers to consider topics of governance, privacy, trust, and transparency. In this one-day session, we convene research and advocacy community members to reflect on critical platform work issues (e.g., worker surveillance, discrimination, wage theft, insufficient platform accountability) as well as to collaborate on codesigning data collectives that ethically and equitably address these concerns by supporting working collectivism and informing policy development.

We study classes of constant-depth circuits with gates that compute restricted polynomial threshold functions, recently introduced by [Kum23] as a family that strictly generalizes $\mathsf{AC}^0$. Denoting these circuit families $\mathsf{bPTFC}^0[k]$ for $\textit{bounded polynomial threshold circuits}$ parameterized by an integer-valued degree-bound $k$, we prove three hardness results separating these classes from constant-depth quantum circuits ($\mathsf{QNC}^0$). $\hspace{2em}$ - We prove that the parity halving problem [WKS+19], which $\mathsf{QNC}^0$ over qubits can solve with certainty, remains average-case hard for polynomial size $\mathsf{bPTFC}^0[k]$ circuits for all $k=\mathcal{O}(n^{1/(5d)})$. $\hspace{2em}$ - We construct a new family of relation problems based on computing $\mathsf{mod}\ p$ for each prime $p>2$, and prove a separation of $\mathsf{QNC}^0$ circuits over higher dimensional quantum systems (`qupits') against $\mathsf{bPTFC}^0[k]$ circuits for the same degree-bound parameter as above. $\hspace{2em}$ - We prove that both foregoing results are noise-robust under the local stochastic noise model, by introducing fault-tolerant implementations of non-Clifford $\mathsf{QNC}^0/|\overline{T^{1/p}}>$ circuits, that use logical magic states as advice. $\mathsf{bPTFC}^0[k]$ circuits can compute certain classes of Polynomial Threshold Functions (PTFs), which in turn serve as a natural model for neural networks and exhibit enhanced expressivity and computational capabilities. Furthermore, for large enough values of $k$, $\mathsf{bPTFC}^0[k]$ contains $\mathsf{TC}^0$ as a subclass. The main challenges we overcome include establishing classical average-case lower bounds, designing non-local games with quantum-classical gaps in winning probabilities and developing noise-resilient non-Clifford quantum circuits necessary to extend beyond qubits to higher dimensions.

To improve logical anomaly detection, some previous works have integrated segmentation techniques with conventional anomaly detection methods. Although these methods are effective, they frequently lead to unsatisfactory segmentation results and require manual annotations. To address these drawbacks, we develop an unsupervised component segmentation technique that leverages foundation models to autonomously generate training labels for a lightweight segmentation network without human labeling. Integrating this new segmentation technique with our proposed Patch Histogram module and the Local-Global Student-Teacher (LGST) module, we achieve a detection AUROC of 95.3% in the MVTec LOCO AD dataset, which surpasses previous SOTA methods. Furthermore, our proposed method provides lower latency and higher throughput than most existing approaches.

Accurate and efficient fluid flow models are essential for applications relating to many physical phenomena including geophysical, aerodynamic, and biological systems. While these flows may exhibit rich and multiscale dynamics, in many cases underlying low-rank structures exist which describe the bulk of the motion. These structures tend to be spatially large and temporally slow, and may contain most of the energy in a given flow. The extraction and parsimonious representation of these low-rank dynamics from high-dimensional data is a key challenge. Inspired by the success of physics-informed machine learning methods, we propose a spectrally-informed approach to extract low-rank models of fluid flows by leveraging known spectral properties in the learning process. We incorporate this knowledge by imposing regularizations on the learned dynamics, which bias the training process towards learning low-frequency structures with corresponding higher power. We demonstrate the effectiveness of this method to improve prediction and produce learned models which better match the underlying spectral properties of prototypical fluid flows., Comment: 13 pages, 10 figures

In actor-critic-based reinforcement learning algorithms such as Twin Delayed Deep Deterministic policy gradient (TD3), insufficient exploration of the spatial space can result in suboptimal policies when controlling 7-DOF robotic arms. To address this issue, we propose a novel Exploration-Enhanced Contrastive Learning (EECL) module that improves exploration by providing additional rewards for encountering novel states. Our module stores previously explored states in a buffer and identifies new states by comparing them with historical data using Euclidean distance within a K-dimensional tree (KDTree) framework. When the agent explores new states, exploration rewards are assigned. These rewards are then integrated into the TD3 algorithm, ensuring that the Q-learning process incorporates these signals, promoting more effective strategy optimization. We evaluate our method on the robosuite panda lift task, demonstrating that it significantly outperforms the baseline TD3 in terms of both efficiency and convergence speed in the tested environment., Comment: 4 pages, 2 figures, IEEE-ICKII-2024

The vast and complicated large-qubit state space forbids us to comprehensively capture the dynamics of modern quantum computers via classical simulations or quantum tomography. However, recent progress in quantum learning theory invokes a crucial question: given a quantum circuit containing d tunable RZ gates and G-d Clifford gates, can a learner perform purely classical inference to efficiently predict its linear properties using new classical inputs, after learning from data obtained by incoherently measuring states generated by the same circuit but with different classical inputs? In this work, we prove that the sample complexity scaling linearly in d is necessary and sufficient to achieve a small prediction error, while the corresponding computational complexity may scale exponentially in d. Building upon these derived complexity bounds, we further harness the concept of classical shadow and truncated trigonometric expansion to devise a kernel-based learning model capable of trading off prediction error and computational complexity, transitioning from exponential to polynomial scaling in many practical settings. Our results advance two crucial realms in quantum computation: the exploration of quantum algorithms with practical utilities and learning-based quantum system certification. We conduct numerical simulations to validate our proposals across diverse scenarios, encompassing quantum information processing protocols, Hamiltonian simulation, and variational quantum algorithms up to 60 qubits.

Language models (LMs) trained on web-scale datasets are largely successful due to their ability to memorize large amounts of training data, even if only present in a few examples. These capabilities are often desirable in evaluation on tasks such as question answering but raise questions about whether these models can exhibit genuine reasoning or succeed only at mimicking patterns from the training data. This distinction is particularly salient in forecasting tasks, where the answer is not present in the training data, and the model must reason to make logical deductions. We present Reasoning and Tools for Forecasting (RTF), a framework of reasoning-and-acting (ReAct) agents that can dynamically retrieve updated information and run numerical simulation with equipped tools. We evaluate our model with questions from competitive forecasting platforms and demonstrate that our method is competitive with and can outperform human predictions. This suggests that LMs, with the right tools, can indeed think and adapt like humans, offering valuable insights for real-world decision-making.

The relative twist angle between layers of near-lattice-matched van der Waals materials is critical for the emergent correlated phenomena associated with moire flat bands. However, the concept of angle rotation control is not exclusive to moir\'e superlattices in which electrons directly experience a twist-angle-dependent periodic potential. Instead, it can also be employed to induce programmable symmetry-breaking perturbations with the goal of stabilizing desired correlated states. Here, we experimentally demonstrate `moireless' twist-tuning of superconductivity together with other correlated orders in Bernal bilayer graphene proximitized by tungsten diselenide. The alignment between the two materials systematically controls the strength of the induced Ising spin-orbit coupling (SOC), profoundly altering the phase diagram. As Ising SOC is increased, superconductivity onsets at a higher displacement field and features a higher critical temperature, reaching up to 0.5K. Within the main superconducting dome and in the strong Ising SOC limit, we find an unusual phase transition characterized by a nematic redistribution of holes among trigonally warped Fermi pockets and enhanced resilience to in-plane magnetic fields. The behavior of the superconducting phase is well captured by our theoretical model, which emphasizes the role of interband interactions between Fermi pockets arising due to interaction-enhanced symmetry breaking. Moreover, we identify two additional superconducting regions, one of which descends from an inter-valley coherent normal state and exhibits a Pauli-limit violation ratio exceeding 40, among the highest for all known superconductors. Our results provide new insights into ultra-clean graphene-based superconductors and underscore the potential of utilizing moireless-twist engineering across a range of van der Waals heterostructures., Comment: main text with four figures, extended data and supplementary information

Quantum machine learning, which involves running machine learning algorithms on quantum devices, may be one of the most significant flagship applications for these devices. Unlike its classical counterparts, the role of data in quantum machine learning has not been fully understood. In this work, we quantify the performances of quantum machine learning in the landscape of quantum data. Provided that the encoding of quantum data is sufficiently random, the performance, we find that the training efficiency and generalization capabilities in quantum machine learning will be exponentially suppressed with the increase in the number of qubits, which we call "the curse of random quantum data". Our findings apply to both the quantum kernel method and the large-width limit of quantum neural networks. Conversely, we highlight that through meticulous design of quantum datasets, it is possible to avoid these curses, thereby achieving efficient convergence and robust generalization. Our conclusions are corroborated by extensive numerical simulations., Comment: 40 pages, 8 figures

Solving large-scale robust portfolio optimization problems is challenging due to the high computational demands associated with an increasing number of assets, the amount of data considered, and market uncertainty. To address this issue, we propose an extended supporting hyperplane approximation approach for efficiently solving a class of distributionally robust portfolio problems for a general class of additively separable utility functions and polyhedral ambiguity distribution set, applied to a large-scale set of assets. Our technique is validated using a large-scale portfolio of the S&P 500 index constituents, demonstrating robust out-of-sample trading performance. More importantly, our empirical studies show that this approach significantly reduces computational time compared to traditional concave Expected Log-Growth (ELG) optimization, with running times decreasing from several thousand seconds to just a few. This method provides a scalable and practical solution to large-scale robust portfolio optimization, addressing both theoretical and practical challenges., Comment: Submitted to possible publication

Soft robots have many advantages over rigid robots thanks to their compliant and passive nature. However, it is generally challenging to model the dynamics of soft robots due to their high spatial dimensionality, making it difficult to use model-based methods to accurately control soft robots. It often requires direct numerical simulation of partial differential equations to simulate soft robots. This not only requires an accurate numerical model, but also makes soft robot modeling slow and expensive. Deep learning algorithms have shown promises in data-driven modeling of soft robots. However, these algorithms usually require a large amount of data, which are difficult to obtain in either simulation or real-world experiments of soft robots. In this work, we propose KNODE-Cosserat, a framework that combines first-principle physics models and neural ordinary differential equations. We leverage the best from both worlds -- the generalization ability of physics-based models and the fast speed of deep learning methods. We validate our framework in both simulation and real-world experiments. In both cases, we show that the robot model significantly improves over the baseline models under different metrics., Comment: 8 pages, 11 figures, 4 tables

Magic state distillation is a crucial yet resource-intensive process in fault-tolerant quantum computation. The protocol's overhead, defined as the number of input magic states required per output magic state with an error rate below $\epsilon$, typically grows as $\mathcal{O}(\log^\gamma(1/\epsilon))$. Achieving smaller overheads, i.e., smaller exponents $\gamma$, is highly desirable; however, all existing protocols require polylogarithmically growing overheads with some $\gamma > 0$, and identifying the smallest achievable exponent $\gamma$ for distilling magic states of qubits has remained challenging. To address this issue, we develop magic state distillation protocols for qubits with efficient, polynomial-time decoding that achieve an $\mathcal{O}(1)$ overhead, meaning the optimal exponent $\gamma = 0$; this improves over the previous best of $\gamma \approx 0.678$ due to Hastings and Haah. In our construction, we employ algebraic geometry codes to explicitly present asymptotically good quantum codes for $2^{10}$-dimensional qudits that support transversally implementable logical gates in the third level of the Clifford hierarchy. The use of asymptotically good codes with non-vanishing rate and relative distance leads to the constant overhead. These codes can be realised by representing each $2^{10}$-dimensional qudit as a set of $10$ qubits, using stabiliser operations on qubits. The $10$-qubit magic states distilled with these codes can be converted to and from conventional magic states for the controlled-controlled-$Z$ ($CCZ$) and $T$ gates on qubits with only a constant overhead loss, making it possible to achieve constant-overhead distillation of such standard magic states for qubits. These results resolve the fundamental open problem in quantum information theory concerning the construction of magic state distillation protocols with the optimal exponent., Comment: 55 pages, Added comments on independent work

The uncertainty principle is one of quantum theory's most foundational features. It underpins a quantum phenomenon called measurement incompatibility -- two physical observables of a single quantum system may not always be measured simultaneously. Apart from being fundamentally important, measurement incompatibility is also a powerful resource in the broad quantum science and technologies, with wide applications to cryptography, communication, random number generation, and device-independent tasks. Since every physical system is unavoidably subject to noise, an important, yet still open, question is how to characterise the ability of noisy quantum dynamics to preserve measurement incompatibility. This work fills this gap by providing the first resource theory of this ability, termed incompatibility preservability. We quantify incompatibility preservability by a robustness measure. Then, we introduce an operational task, entanglement-assisted filter game, to completely characterise both the robustness measure and the conversion of incompatibility preservability. Our results provide a general framework to describe how noisy dynamics affect the uncertainty principle's signature., Comment: 4+5 pages; 4 figures

We study a new variant of graph coloring by adding a connectivity constraint. A path in a vertex-colored graph is called conflict-free if there is a color that appears exactly once on its vertices. A connected graph $G$ is said to be strongly conflict-free vertex-connection $k$-colorable if $G$ admits a vertex $k$-coloring such that any two distinct vertices of $G$ are connected by a conflict-free $shortest$ path. Among others, we show that deciding whether a given graph is strongly conflict-free vertex-connection $3$-colorable is NP-complete even when restricted to $3$-colorable graphs with diameter $3$, radius $2$ and domination number $3$, and, assuming the Exponential Time Hypothesis (ETH), cannot be solved in $2^{o(n)}$ time on such restricted input graphs with $n$ vertices. This hardness result is quite strong when compared to the ordinary $3$-COLORING problem: it is known that $3$-COLORING is solvable in polynomial time in graphs with bounded domination number, and assuming ETH, cannot be solved in $2^{o(\sqrt{n})}$ time in $n$-vertex graphs with diameter $3$ and radius $2$. On the positive side, we point out that a strong conflict-free vertex-connection coloring with minimum color number of a given split graph or a co-bipartite graph can be computed in polynomial time., Comment: The full version of a COCOON 2024 paper

This paper investigates the application of millimeter-wave (mmWave) radar for the estimation of human vital signs. Aiming to obtain more accurate frequency estimation for periodic signals of respiration and heartbeat, we propose the harmonic MUSIC (HMUSIC) algorithm to consider harmonic components for frequency estimation of vital sign signals. In the experiments, we tested different subjects' vital signs. Experimental results demonstrate that the 89-th percentile errors in respiration rate and the 88-th percentile errors in heartbeat rate are less than 3 respirations per minute and 5 beats per minute.

Geometrically local quantum codes, comprised of qubits and checks embedded in $\mathbb{R}^D$ with local check operators, have been a subject of significant interest. A key challenge is identifying the optimal code construction that maximizes both dimension and distance. Recent advancements have produced several constructions, but these either depend on specific good quantum low-density parity-check (qLDPC) codes or are limited to three dimensions. In this work, we introduce a construction that can transform any good qLDPC code into an optimal geometrically local quantum code. Our approach hinges on a novel procedure that extracts a two-dimensional structure from an arbitrary three-term chain complex. We expect that this procedure will find broader applications in areas such as weight reduction and the geometric realization of chain complexes., Comment: 25 pages, 15 figures

Humans describe complex scenes with compositionality, using simple text descriptions enriched with links and relationships. While vision-language research has aimed to develop models with compositional understanding capabilities, this is not reflected yet in existing datasets which, for the most part, still use plain text to describe images. In this work, we propose a new annotation strategy, graph-based captioning (GBC) that describes an image using a labelled graph structure, with nodes of various types. The nodes in GBC are created using, in a first stage, object detection and dense captioning tools nested recursively to uncover and describe entity nodes, further linked together in a second stage by highlighting, using new types of nodes, compositions and relations among entities. Since all GBC nodes hold plain text descriptions, GBC retains the flexibility found in natural language, but can also encode hierarchical information in its edges. We demonstrate that GBC can be produced automatically, using off-the-shelf multimodal LLMs and open-vocabulary detection models, by building a new dataset, GBC10M, gathering GBC annotations for about 10M images of the CC12M dataset. We use GBC10M to showcase the wealth of node captions uncovered by GBC, as measured with CLIP training. We show that using GBC nodes' annotations -- notably those stored in composition and relation nodes -- results in significant performance boost on downstream models when compared to other dataset formats. To further explore the opportunities provided by GBC, we also propose a new attention mechanism that can leverage the entire GBC graph, with encouraging experimental results that show the extra benefits of incorporating the graph structure. Our datasets are released at \url{https://huggingface.co/graph-based-captions}., Comment: 47 pages, 33 figures