Search

Multimodal models leverage large-scale pre-training to achieve strong but still imperfect performance on tasks such as image captioning, visual question answering, and cross-modal retrieval. In this paper, we present a simple and efficient method for correcting errors in trained contrastive image-text retrieval models with no additional training, called Nearest Neighbor Normalization (NNN). We show an improvement on retrieval metrics in both text retrieval and image retrieval for all of the contrastive models that we tested (CLIP, BLIP, ALBEF, SigLIP, BEiT) and for both of the datasets that we used (MS-COCO and Flickr30k). NNN requires a reference database, but does not require any training on this database, and can even increase the retrieval accuracy of a model after finetuning.

The single variant spin cycloid and associated antiferromagnetic order in multiferroic BiFeO3 can provide a direct and predictable magnetoelectric coupling to ferroelectric order for deterministic switching, and also a key to fundamental understanding of spin transport and magnon-based applications in the system. (111) oriented BiFeO3 supplies an easy magnetic plane for the spin cycloid, but despite previous efforts, achieving deterministic switching of the single spin cycloid over multiple cycles remains challenging due to the presence of multiple spin cycloid domains in the (111) plane. Here we show that anisotropic in-plane strain engineering can stabilize a single antiferromagnetic domain and provide robust, deterministic switching. We grow BiFeO3 on orthorhombic NdGaO3 (011)o [(111)pc] substrates, breaking the spin cycloid degeneracy and by imposing a uniaxial strain in the (111) plane. This stabilization is confirmed through direct imaging with scanning NV microscopy and non-resonant X-ray diffraction. Remarkably, we achieved deterministic and non-volatile 180{\deg} switching of ferroelectric and associated antiferromagnetic domains over 1,000 cycles, significantly outperforming existing approaches. Our findings underscore that anisotropic strain engineering opens up exciting possibilities for (111)pc monodomain BiFeO3 in potential magnetoelectric and emerging magnonic applications., Comment: 17 pages, 4 figures

Predicting low-energy molecular conformations given a molecular graph is an important but challenging task in computational drug discovery. Existing state-of-the-art approaches either resort to large scale transformer-based models that diffuse over conformer fields, or use computationally expensive methods to generate initial structures and diffuse over torsion angles. In this work, we introduce Equivariant Transformer Flow (ET-Flow). We showcase that a well-designed flow matching approach with equivariance and harmonic prior alleviates the need for complex internal geometry calculations and large architectures, contrary to the prevailing methods in the field. Our approach results in a straightforward and scalable method that directly operates on all-atom coordinates with minimal assumptions. With the advantages of equivariance and flow matching, ET-Flow significantly increases the precision and physical validity of the generated conformers, while being a lighter model and faster at inference. Code is available https://github.com/shenoynikhil/ETFlow., Comment: NeurIPS 2024

QR codes have become ubiquitous in daily life, enabling rapid information exchange. With the increasing adoption of smart wearable devices, there is a need for efficient, and friction-less QR code reading capabilities from Egocentric point-of-views. However, adapting existing phone-based QR code readers to egocentric images poses significant challenges. Code reading from egocentric images bring unique challenges such as wide field-of-view, code distortion and lack of visual feedback as compared to phones where users can adjust the position and framing. Furthermore, wearable devices impose constraints on resources like compute, power and memory. To address these challenges, we present EgoQR, a novel system for reading QR codes from egocentric images, and is well suited for deployment on wearable devices. Our approach consists of two primary components: detection and decoding, designed to operate on high-resolution images on the device with minimal power consumption and added latency. The detection component efficiently locates potential QR codes within the image, while our enhanced decoding component extracts and interprets the encoded information. We incorporate innovative techniques to handle the specific challenges of egocentric imagery, such as varying perspectives, wider field of view, and motion blur. We evaluate our approach on a dataset of egocentric images, demonstrating 34% improvement in reading the code compared to an existing state of the art QR code readers., Comment: Submitted to ICLR 2025

We present an analysis of 152,355 radio sources identified in the second data release of the LOFAR Two Metre Sky Survey (LoTSS-DR2) with Sloan Digital Sky Survey (SDSS) spectroscopic redshifts in the range 0.00 < z < 0.57. Using Monte Carlo simulations we determine the reliability of each source exhibiting an excess in radio luminosity relative to that predicted from their Ha emission, and, for a subset of 124,023 sources we combine this measurement with a full BPT analysis. Using these two independent diagnostics we determine the reliability of each source hosting a supermassive black hole of high or low Eddington-scaled accretion rate, and combine the measurements to determine the reliability of sources belonging to each of four physical classes of objects: star forming galaxies (SFGs), radio-quiet active galactic nuclei (RQAGN), and high- or low-excitation radio galaxies (HERGs or emission-line LERGs). The result is a catalogue which enables user-defined samples of radio sources with a reliability threshold suited to their science goal e.g. prioritising purity or completeness. Here we select high-confidence samples of radio sources (>90% reliability) to report: 38,588 radio-excess AGN in the LoTSS DR2 sample (362 HERGs, and 12,648 emission-line LERGs), together with 38,729 SFGs, and 18,726 RQAGN. We validate these results through comparison to literature using independent emission-line measurements, and to widely-adopted WISE photometric selection techniques. While our use of SDSS spectroscopy limits our current analysis to ~4 percent of the LoTSS-DR2 catalogue, our method is directly applicable to data from the forthcoming WEAVE-LOFAR survey which will obtain over a million spectra of 144 MHz selected sources., Comment: Accepted for publication in MNRAS

The role of radio mode AGN feedback on galaxy evolution is still under debate. In this study, we utilize a combination of radio continuum observations and optical integral field spectroscopic (IFS) data to explore the impact of radio AGN on the evolution of their host galaxies at both global and sub-galactic scales. We construct a comprehensive radio-IFS sample comprising 5578 galaxies with redshift z < 0.15 by cross-matching the LOFAR Two-Metre Sky Survey (LoTSS) with the MaNGA survey. We revisit the tight linear radio continuum - star formation relation and quantify its intrinsic scatter, then use the relation to classify 616 radio-excess AGN with excessive radio luminosities over that expected from their star formation rate. Massive radio AGN host galaxies are predominantly quiescent systems, but the quenching level shows no correlation with the jet luminosity. The mass assembly histories derived from the stellar population synthesis model fitting agree with the cosmological simulations incorporating radio-mode AGN feedback models. We observe that radio AGN hosts grow faster than a control sample of galaxies matched in stellar mass, and the quenching age (~ 5 Gyr) is at larger lookback times than the typical radio jet age (< 1 Gyr). By stacking the spectra in different radial bins and comparing results for radio AGN hosts and their controls, we find emission line excess features in the nuclear region of radio AGN hosts. This indicates that radio AGN are ionizing the surrounding interstellar medium in the vicinity of the nucleus. We also find that ongoing star formation in the outer regions of the galaxy is weaker if a radio jet is detected. Our findings support the scenario that the observed present-day radio AGN activity is not responsible for the past quenching of their hosts, but may help the host galaxies maintain quiescence through ionizing and heating the surrounding gas., Comment: 16 pages, 15 figures, submitted to A&A

Ridesharing platforms such as Uber, Lyft, and DiDi have grown in popularity due to their on-demand availability, ease of use, and commute cost reductions, among other benefits. However, not all ridesharing promises have panned out. Recent studies demonstrate that the expected drop in traffic congestion and reduction in greenhouse gas (GHG) emissions have not materialized. This is primarily due to the substantial distances traveled by the ridesharing vehicles without passengers between rides, known as deadhead miles. Recent work has focused on reducing the impact of deadhead miles while considering additional metrics such as rider waiting time, GHG emissions from deadhead miles, or driver earnings. Unfortunately, prior studies consider these environmental and equity-based metrics individually despite them being interrelated. In this paper, we propose a Learning-based Equity-Aware Decarabonization approach, LEAD, for ridesharing platforms. LEAD targets minimizing emissions while ensuring that the driver's utility, defined as the difference between the trip distance and the deadhead miles, is fairly distributed. LEAD uses reinforcement learning to match riders to drivers based on the expected future utility of drivers and the expected carbon emissions of the platform without increasing the rider waiting times. Extensive experiments based on a real-world ride-sharing dataset show that LEAD improves fairness by 2$\times$ when compared to emission-aware ride-assignment and reduces emissions by 70% while ensuring fairness within 66% of the fair baseline. It also reduces the rider wait time, by at least 40%, compared to various baselines. Additionally, LEAD corrects the imbalance in previous emission-aware ride assignment algorithms that overassigned rides to low-emission vehicles.

We study carbon-aware spatiotemporal workload management, which seeks to address the growing environmental impact of data centers. We formalize this as an online problem called spatiotemporal online allocation with deadline constraints ($\mathsf{SOAD}$), in which an online player completes a workload (e.g., a batch compute job) by moving and scheduling the workload across a network subject to a deadline $T$. At each time step, a service cost function is revealed, representing, e.g., the carbon intensity of servicing a workload at each location, and the player must irrevocably decide the current allocation. Furthermore, whenever the player moves the allocation, it incurs a movement cost defined by a metric space $(X,d)$ that captures, e.g., the overhead of migrating a compute job. $\mathsf{SOAD}$ formalizes the open problem of combining general metrics and deadline constraints in the online algorithms literature, unifying problems such as metrical task systems and online search. We propose a competitive algorithm for $\mathsf{SOAD}$ along with a matching lower bound that proves it is optimal. Our main algorithm, ${\rm C{\scriptsize ARBON}C{\scriptsize LIPPER}}$, is a learning-augmented algorithm that takes advantage of predictions (e.g., carbon intensity forecasts) and achieves an optimal consistency-robustness trade-off. We evaluate our proposed algorithms for carbon-aware spatiotemporal workload management on a simulated global data center network, showing that ${\rm C{\scriptsize ARBON}C{\scriptsize LIPPER}}$ significantly improves performance compared to baseline methods and delivers meaningful carbon reductions., Comment: 50 pages, 21 figures

The contributions of this technical note are twofold. Firstly, we formulate an optimization problem to obtain a linear representation of a nonlinear vector field based on a system's trajectory. We also prove that its cost function is strictly convex, given the trajectory is persistently exciting. Under certain observability conditions, we provide results that guarantee the Hurwitz stability of the global minimizer. Secondly, we present a novel algorithm based on point-wise geometric flows to estimate the boundary of the region of attraction. We show that the algorithm converges to the exact boundary of the region of attraction under certain assumptions on the system dynamics. Finally, we validate the results using simulations on various nonlinear autonomous systems.

We present deep 1.4-4.8 um JWST-NIRCam imaging of the Serpens Main star-forming region and identify 20 candidate protostellar outflows, most with bipolar structure and identified driving sources. The outflow position angles (PAs) are strongly correlated, and aligned within +/- 24 degrees of the major axis of the Serpens filament. These orientations are further aligned with the angular momentum vectors of the two disk shadows in this region. We estimate that the probability of this number of young stars being co-aligned if sampled from a uniform PA distribution is 10^-4. This in turn suggests that the aligned protostars, which seem to be at similar evolutionary stages based on their outflow dynamics, formed at similar times with a similar spin inherited from a local cloud filament. Further, there is tentative evidence for a systematic change in average position angle between the north-western and south-eastern cluster, as well as increased scatter in the PAs of the south-eastern protostars. SOFIA-HAWC+ archival dust polarization observations of Serpens Main at 154 and 214 um are perpendicular to the dominant jet orientation in NW region in particular. We measure and locate shock knots and edges for all of the outflows and provide an identifying catalog. We suggest that Serpens main is a cluster that formed from an isolated filament, and due to its youth retains its primordial outflow alignment., Comment: 18 pages, 7 figures

Machine learning and quantum machine learning (QML) have gained significant importance, as they offer powerful tools for tackling complex computational problems across various domains. This work gives an extensive overview of QML uses in quantitative finance, an important discipline in the financial industry. We examine the connection between quantum computing and machine learning in financial applications, spanning a range of use cases including fraud detection, underwriting, Value at Risk, stock market prediction, portfolio optimization, and option pricing by overviewing the corpus of literature concerning various financial subdomains., Comment: comments are welcome

Satellite image time series (SITS) segmentation is crucial for many applications like environmental monitoring, land cover mapping and agricultural crop type classification. However, training models for SITS segmentation remains a challenging task due to the lack of abundant training data, which requires fine grained annotation. We propose S4 a new self-supervised pre-training approach that significantly reduces the requirement for labeled training data by utilizing two new insights: (a) Satellites capture images in different parts of the spectrum such as radio frequencies, and visible frequencies. (b) Satellite imagery is geo-registered allowing for fine-grained spatial alignment. We use these insights to formulate pre-training tasks in S4. We also curate m2s2-SITS, a large-scale dataset of unlabeled, spatially-aligned, multi-modal and geographic specific SITS that serves as representative pre-training data for S4. Finally, we evaluate S4 on multiple SITS segmentation datasets and demonstrate its efficacy against competing baselines while using limited labeled data.

Spectral energy distribution (SED) fitting has been extensively used to determine the nature of the faint radio source population. Recent efforts have combined fits from multiple SED-fitting codes to account for the host galaxy and any active nucleus that may be present. We show that it is possible to produce similar-quality classifications using a single energy-balance SED fitting code, Prospector, to model up to 26 bands of UV$-$far-infrared aperture-matched photometry for $\sim$31,000 sources in the ELAIS-N1 field from the LOFAR Two-Metre Sky Survey (LoTSS) Deep fields first data release. One of a new generation of SED-fitting codes, Prospector accounts for potential contributions from radiative active galactic nuclei (AGN) when estimating galaxy properties, including star formation rates (SFRs) derived using non-parametric star formation histories. Combining this information with radio luminosities, we classify 92 per cent of the radio sources as a star-forming galaxy, high/low-excitation radio galaxy, or radio-quiet AGN and study the population demographics as a function of 150 MHz flux density, luminosity, SFR, stellar mass, redshift and apparent $r$-band magnitude. Finally, we use Prospector SED fits to investigate the SFR$-$150 MHz luminosity relation for a sample of $\sim$$133,000~3.6~\mu$m-selected $z<1$ sources, finding that the stellar mass dependence is significantly weaker than previously reported, and may disappear altogether at $\log_{10} (\mathrm{SFR}/M_\odot~\mathrm{yr}^{-1}) > 0.5$. This approach makes it significantly easier to classify radio sources from LoTSS and elsewhere, and may have important implications for future studies of star-forming galaxies at radio wavelengths., Comment: 20 pages, 13 figures, accepted for publication in MNRAS

Audio signals can reveal intimate details about a person's life, including their conversations, health status, emotions, location, and personal preferences. Unauthorized access or misuse of this information can have profound personal and social implications. In an era increasingly populated by devices capable of audio recording, safeguarding user privacy is a critical obligation. This work studies the ethical and privacy concerns in current audio classification systems. We discuss the challenges and research directions in designing privacy-preserving audio sensing systems. We propose privacy-preserving audio features that can be used to classify wide range of audio classes, while being privacy preserving.

Mid-infrared emission features probe the properties of ionized gas, and hot or warm molecular gas. The Orion Bar is a frequently studied photodissociation region (PDR) containing large amounts of gas under these conditions, and was observed with the MIRI IFU aboard JWST as part of the "PDRs4All" program. The resulting IR spectroscopic images of high angular resolution (0.2") reveal a rich observational inventory of mid-IR emission lines, and spatially resolve the substructure of the PDR, with a mosaic cutting perpendicularly across the ionization front and three dissociation fronts. We extracted five spectra that represent the ionized, atomic, and molecular gas layers, and measured the most prominent gas emission lines. An initial analysis summarizes the physical conditions of the gas and the potential of these data. We identified around 100 lines, report an additional 18 lines that remain unidentified, and measured the line intensities and central wavelengths. The H I recombination lines originating from the ionized gas layer bordering the PDR, have intensity ratios that are well matched by emissivity coefficients from H recombination theory, but deviate up to 10% due contamination by He I lines. We report the observed emission lines of various ionization stages of Ne, P, S, Cl, Ar, Fe, and Ni, and show how certain line ratios vary between the five regions. We observe the pure-rotational H$_2$ lines in the vibrational ground state from 0-0 S(1) to 0-0 S(8), and in the first vibrationally excited state from 1-1 S(5) to 1-1 S(9). We derive H$_2$ excitation diagrams, and approximate the excitation with one thermal (~700 K) component representative of an average gas temperature, and one non-thermal component (~2700 K) probing the effect of UV pumping. We compare these results to an existing model for the Orion Bar PDR and highlight the differences with the observations., Comment: 26 pages, 12 figures, 3 tables. Submitted to A&A, under review (1st revision)

The phase-field method has become a useful tool for the simulation of classical metallurgical phase transformations as well as other phenomena related to materials science. The thermodynamic consistency that forms the basis of these formulations lends to its strong predictive capabilities and utility. However, a strong impediment to the usage of the method for typical applied problems of industrial and academic relevance is the significant overhead with regard to the code development and know-how required for quantitative model formulations. In this paper, we report the development of an open-source phase-field software stack that contains generic formulations for the simulation of multi-phase and multi-component phase transformations. The solvers incorporate thermodynamic coupling that allows the realization of simulations with real alloys in scenarios directly relevant to the materials industry. Further, the solvers utilize parallelization strategies using either multiple CPUs or GPUs to provide cross-platform portability and usability on available supercomputing machines. Finally, the solver stack also contains a graphical user interface to gradually introduce the usage of the software. The user interface also provides a collection of post-processing tools that allow the estimation of useful metrics related to microstructural evolution.

Motivated by an imperative to reduce the carbon emissions of cloud data centers, this paper studies the online carbon-aware resource scaling problem with unknown job lengths (OCSU) and applies it to carbon-aware resource scaling for executing computing workloads. The task is to dynamically scale resources (e.g., the number of servers) assigned to a job of unknown length such that it is completed before a deadline, with the objective of reducing the carbon emissions of executing the workload. The total carbon emissions of executing a job originate from the emissions of running the job and excess carbon emitted while switching between different scales (e.g., due to checkpoint and resume). Prior work on carbon-aware resource scaling has assumed accurate job length information, while other approaches have ignored switching losses and require carbon intensity forecasts. These assumptions prohibit the practical deployment of prior work for online carbon-aware execution of scalable computing workload. We propose LACS, a theoretically robust learning-augmented algorithm that solves OCSU. To achieve improved practical average-case performance, LACS integrates machine-learned predictions of job length. To achieve solid theoretical performance, LACS extends the recent theoretical advances on online conversion with switching costs to handle a scenario where the job length is unknown. Our experimental evaluations demonstrate that, on average, the carbon footprint of LACS lies within 1.2% of the online baseline that assumes perfect job length information and within 16% of the offline baseline that, in addition to the job length, also requires accurate carbon intensity forecasts. Furthermore, LACS achieves a 32% reduction in carbon footprint compared to the deadline-aware carbon-agnostic execution of the job.

There has been a significant societal push towards sustainable practices, including in computing. Modern interactive workloads such as geo-distributed web-services exhibit various spatiotemporal and performance flexibility, enabling the possibility to adapt the location, time, and intensity of processing to align with the availability of renewable and low-carbon energy. An example is a web application hosted across multiple cloud regions, each with varying carbon intensity based on their local electricity mix. Distributed load-balancing enables the exploitation of low-carbon energy through load migration across regions, reducing web applications carbon footprint. In this paper, we present CASPER, a carbon-aware scheduling and provisioning system that primarily minimizes the carbon footprint of distributed web services while also respecting their Service Level Objectives (SLO). We formulate CASPER as an multi-objective optimization problem that considers both the variable carbon intensity and latency constraints of the network. Our evaluation reveals the significant potential of CASPER in achieving substantial reductions in carbon emissions. Compared to baseline methods, CASPER demonstrates improvements of up to 70% with no latency performance degradation.

Learned reweighting (LRW) approaches to supervised learning use an optimization criterion to assign weights for training instances, in order to maximize performance on a representative validation dataset. We pose and formalize the problem of optimized selection of the validation set used in LRW training, to improve classifier generalization. In particular, we show that using hard-to-classify instances in the validation set has both a theoretical connection to, and strong empirical evidence of generalization. We provide an efficient algorithm for training this meta-optimized model, as well as a simple train-twice heuristic for careful comparative study. We demonstrate that LRW with easy validation data performs consistently worse than LRW with hard validation data, establishing the validity of our meta-optimization problem. Our proposed algorithm outperforms a wide range of baselines on a range of datasets and domain shift challenges (Imagenet-1K, CIFAR-100, Clothing-1M, CAMELYON, WILDS, etc.), with ~1% gains using VIT-B on Imagenet. We also show that using naturally hard examples for validation (Imagenet-R / Imagenet-A) in LRW training for Imagenet improves performance on both clean and naturally hard test instances by 1-2%. Secondary analyses show that using hard validation data in an LRW framework improves margins on test data, hinting at the mechanism underlying our empirical gains. We believe this work opens up new research directions for the meta-optimization of meta-learning in a supervised learning context., Comment: Accepted at CVPR 2024

In this report, we introduce the Gemini 1.5 family of models, representing the next generation of highly compute-efficient multimodal models capable of recalling and reasoning over fine-grained information from millions of tokens of context, including multiple long documents and hours of video and audio. The family includes two new models: (1) an updated Gemini 1.5 Pro, which exceeds the February version on the great majority of capabilities and benchmarks; (2) Gemini 1.5 Flash, a more lightweight variant designed for efficiency with minimal regression in quality. Gemini 1.5 models achieve near-perfect recall on long-context retrieval tasks across modalities, improve the state-of-the-art in long-document QA, long-video QA and long-context ASR, and match or surpass Gemini 1.0 Ultra's state-of-the-art performance across a broad set of benchmarks. Studying the limits of Gemini 1.5's long-context ability, we find continued improvement in next-token prediction and near-perfect retrieval (>99%) up to at least 10M tokens, a generational leap over existing models such as Claude 3.0 (200k) and GPT-4 Turbo (128k). Finally, we highlight real-world use cases, such as Gemini 1.5 collaborating with professionals on completing their tasks achieving 26 to 75% time savings across 10 different job categories, as well as surprising new capabilities of large language models at the frontier; when given a grammar manual for Kalamang, a language with fewer than 200 speakers worldwide, the model learns to translate English to Kalamang at a similar level to a person who learned from the same content.

This paper presents a novel application of explainable AI (XAI) for root-causing performance degradation in machine learning models that learn continuously from user engagement data. In such systems a single feature corruption can cause cascading feature, label and concept drifts. We have successfully applied this technique to improve the reliability of models used in personalized advertising. Performance degradation in such systems manifest as prediction anomalies in the models. These models are typically trained continuously using features that are produced by hundreds of real time data processing pipelines or derived from other upstream models. A failure in any of these pipelines or an instability in any of the upstream models can cause feature corruption, causing the model's predicted output to deviate from the actual output and the training data to become corrupted. The causal relationship between the features and the predicted output is complex, and root-causing is challenging due to the scale and dynamism of the system. We demonstrate how temporal shifts in the global feature importance distribution can effectively isolate the cause of a prediction anomaly, with better recall than model-to-feature correlation methods. The technique appears to be effective even when approximating the local feature importance using a simple perturbation-based method, and aggregating over a few thousand examples. We have found this technique to be a model-agnostic, cheap and effective way to monitor complex data pipelines in production and have deployed a system for continuously analyzing the global feature importance distribution of continuously trained models., Comment: Submitted to The 2nd World Conference on eXplainable Artificial Intelligence, 17-19 July, 2024, Malta, Valletta

Most low-mass stars form in stellar clusters that also contain massive stars, which are sources of far-ultraviolet (FUV) radiation. Theoretical models predict that this FUV radiation produces photo-dissociation regions (PDRs) on the surfaces of protoplanetary disks around low-mass stars, impacting planet formation within the disks. We report JWST and Atacama Large Millimetere Array observations of a FUV-irradiated protoplanetary disk in the Orion Nebula. Emission lines are detected from the PDR; modelling their kinematics and excitation allows us to constrain the physical conditions within the gas. We quantify the mass-loss rate induced by the FUV irradiation, finding it is sufficient to remove gas from the disk in less than a million years. This is rapid enough to affect giant planet formation in the disk.

We introduce and study a family of online metric problems with long-term constraints. In these problems, an online player makes decisions $\mathbf{x}_t$ in a metric space $(X,d)$ to simultaneously minimize their hitting cost $f_t(\mathbf{x}_t)$ and switching cost as determined by the metric. Over the time horizon $T$, the player must satisfy a long-term demand constraint $\sum_{t} c(\mathbf{x}_t) \geq 1$, where $c(\mathbf{x}_t)$ denotes the fraction of demand satisfied at time $t$. Such problems can find a wide array of applications to online resource allocation in sustainable energy/computing systems. We devise optimal competitive and learning-augmented algorithms for the case of bounded hitting cost gradients and weighted $\ell_1$ metrics, and further show that our proposed algorithms perform well in numerical experiments., Comment: Accepted to ICML 2024. 31 pages, 12 figures

The Bloch sphere provides an elegant way of visualizing a qubit. Analogous representation of the simplest composite state of two-qubits has attracted significant attention. Here we present a detailed mathematical analysis of the real-matrix parametrization and associated geometric picturization of arbitrary two-qubit states - up to their local SL2C equivalence, in terms of canonical ellipsoids inscribed within the Bloch sphere., Comment: 12 pages, 8 figures, one table; typos in Example 2 corrected; Submitted to "Bound States and Quantum Correlations - ARP Rau special collection"

We explore phases of free fermions on arenas that do not tessellate a manifold. Specializing to arboreal arenas described by tree graphs which possess a notion of translation symmetry, we study possible fermionic phases in the BDI symmetry class on the $p$-coordinated Bethe lattice. We find that there are $p$ distinct obstructed atomic insulating phases that are characterized by distinct edge states, pattern of entanglement, and a winding characteristic that we define here. These distinct insulting phases are always separated by a metallic region in the parameter space rather than isolated quantum critical points. The metallic region itself comprises several distinct metallic phases that are distinguished by the winding characteristic and correlation functions. The correlation functions of distinct subsystems display non-analytic behavior at distinct points in the metallic region, signaling a cascade of subsystem transitions. An intriguing feature of these arboreal metals is the presence of truncated subsystems with zero energy boundary modes despite being gapless. This work suggests new opportunities for synthetic quantum systems to realize these novel phases., Comment: 12 pages, 6 figures, Supplemental Material (20 pages, 12 figures)

We introduce Lumos, the first end-to-end multimodal question-answering system with text understanding capabilities. At the core of Lumos is a Scene Text Recognition (STR) component that extracts text from first person point-of-view images, the output of which is used to augment input to a Multimodal Large Language Model (MM-LLM). While building Lumos, we encountered numerous challenges related to STR quality, overall latency, and model inference. In this paper, we delve into those challenges, and discuss the system architecture, design choices, and modeling techniques employed to overcome these obstacles. We also provide a comprehensive evaluation for each component, showcasing high quality and efficiency., Comment: Accepted to KDD 2024 (ADS Track)

In recent years, there has been an increased emphasis on reducing the carbon emissions from electricity consumption. Many organizations have set ambitious targets to reduce the carbon footprint of their operations as a part of their sustainability goals. The carbon footprint of any consumer of electricity is computed as the product of the total energy consumption and the carbon intensity of electricity. Third-party carbon information services provide information on carbon intensity across regions that consumers can leverage to modulate their energy consumption patterns to reduce their overall carbon footprint. In addition, to accelerate their decarbonization process, large electricity consumers increasingly acquire power purchase agreements (PPAs) from renewable power plants to obtain renewable energy credits that offset their "brown" energy consumption. There are primarily two methods for attributing carbon-free energy, or renewable energy credits, to electricity consumers: location-based and market-based. These two methods yield significantly different carbon intensity values for various consumers. As there is a lack of consensus which method to use for carbon-free attribution, a concurrent application of both approaches is observed in practice. In this paper, we show that such concurrent applications can cause discrepancies in the carbon savings reported by carbon optimization techniques. Our analysis across three state-of-the-art carbon optimization techniques shows possible overestimation of up to 55.1% in the carbon reductions reported by the consumers and even increased emissions for consumers in some cases. We also find that carbon optimization techniques make different decisions under the market-based method and location-based method, and the market-based method can yield up to 28.2% less carbon savings than those claimed by the location-based method for consumers without PPAs.