Search

Cataract surgery is the most common surgical procedure globally, with a disproportionately higher burden in developing countries. While automated surgical video analysis has been explored in general surgery, its application to ophthalmic procedures remains limited. Existing works primarily focus on Phaco cataract surgery, an expensive technique not accessible in regions where cataract treatment is most needed. In contrast, Manual Small-Incision Cataract Surgery (MSICS) is the preferred low-cost, faster alternative in high-volume settings and for challenging cases. However, no dataset exists for MSICS. To address this gap, we introduce Sankara-MSICS, the first comprehensive dataset containing 53 surgical videos annotated for 18 surgical phases and 3,527 frames with 13 surgical tools at the pixel level. We benchmark this dataset on state-of-the-art models and present ToolSeg, a novel framework that enhances tool segmentation by introducing a phase-conditional decoder and a simple yet effective semi-supervised setup leveraging pseudo-labels from foundation models. Our approach significantly improves segmentation performance, achieving a $23.77\%$ to $38.10\%$ increase in mean Dice scores, with a notable boost for tools that are less prevalent and small. Furthermore, we demonstrate that ToolSeg generalizes to other surgical settings, showcasing its effectiveness on the CaDIS dataset.

The driver warning system that alerts the human driver about potential risks during driving is a key feature of an advanced driver assistance system. Existing driver warning technologies, mainly the forward collision warning and unsafe lane change warning, can reduce the risk of collision caused by human errors. However, the current design methods have several major limitations. Firstly, the warnings are mainly generated in a one-shot manner without modeling the ego driver's reactions and surrounding objects, which reduces the flexibility and generality of the system over different scenarios. Additionally, the triggering conditions of warning are mostly rule-based threshold-checking given the current state, which lacks the prediction of the potential risk in a sufficiently long future horizon. In this work, we study the problem of optimally generating driver warnings by considering the interactions among the generated warning, the driver behavior, and the states of ego and surrounding vehicles on a long horizon. The warning generation problem is formulated as a partially observed Markov decision process (POMDP). An optimal warning generation framework is proposed as a solution to the proposed POMDP. The simulation experiments demonstrate the superiority of the proposed solution to the existing warning generation methods., Comment: ICRA 2024

We study the problem of estimating the body movements of a camera wearer from egocentric videos. Current methods for ego-body pose estimation rely on temporally dense sensor data, such as IMU measurements from spatially sparse body parts like the head and hands. However, we propose that even temporally sparse observations, such as hand poses captured intermittently from egocentric videos during natural or periodic hand movements, can effectively constrain overall body motion. Naively applying diffusion models to generate full-body pose from head pose and sparse hand pose leads to suboptimal results. To overcome this, we develop a two-stage approach that decomposes the problem into temporal completion and spatial completion. First, our method employs masked autoencoders to impute hand trajectories by leveraging the spatiotemporal correlations between the head pose sequence and intermittent hand poses, providing uncertainty estimates. Subsequently, we employ conditional diffusion models to generate plausible full-body motions based on these temporally dense trajectories of the head and hands, guided by the uncertainty estimates from the imputation. The effectiveness of our method was rigorously tested and validated through comprehensive experiments conducted on various HMD setup with AMASS and Ego-Exo4D datasets., Comment: Accepted at NeurIPS 2024

Human navigation is facilitated through the association of actions with landmarks, tapping into our ability to recognize salient features in our environment. Consequently, navigational instructions for humans can be extremely concise, such as short verbal descriptions, indicating a small memory requirement and no reliance on complex and overly accurate navigation tools. Conversely, current autonomous navigation schemes rely on accurate positioning devices and algorithms as well as extensive streams of sensory data collected from the environment. Inspired by this human capability and motivated by the associated technological gap, in this work we propose a hierarchical end-to-end meta-learning scheme that enables a mobile robot to navigate in a previously unknown environment upon presentation of only a few sample images of a set of landmarks along with their corresponding high-level navigation actions. This dramatically simplifies the wayfinding process and enables easy adoption to new environments. For few-shot waypoint detection, we implement a metric-based few-shot learning technique through distribution embedding. Waypoint detection triggers the multi-task low-level maneuver controller module to execute the corresponding high-level navigation action. We demonstrate the effectiveness of the scheme using a small-scale autonomous vehicle on novel indoor navigation tasks in several previously unseen environments., Comment: Appeared at the 40th Anniversary of the IEEE International Conference on Robotics and Automation (ICRA@40), 23-26 September, 2024, Rotterdam, The Netherlands. 9 pages, 5 figures

It is well-known that in the large central charge $c$ limit, the Virasoro blocks in CFT$_2$ suffer from information loss when studying the BTZ black holes in the dual AdS$_3$ theory. Recent studies have proved that non-perturbative $e^{-c}$ corrections resolves information loss. Here, we study universal features of the spectrum of heavy states from general properties of finite $c$ Virasoro conformal blocks. Furthermore, we investigate the behavior of correlators beyond "Page" time $ t \sim S_{BH}$, entropy of the black hole. In the Lorentzian regime, we wish to better understand the transition from an exponential decay behaviour of Virasoro blocks in the heavy-light limit to a power-law decay at late times., Comment: This revised version updates my current affiliation, replaces the previous institute's logo, and replaces the inaccurate use of the term thesis with the term project submitted as part of the MSc degree. 76 pages, 20 figures with additional references added

Community health workers (CHWs) provide last-mile healthcare services but face challenges due to limited medical knowledge and training. This paper describes the design, deployment, and evaluation of ASHABot, an LLM-powered, experts-in-the-loop, WhatsApp-based chatbot to address the information needs of CHWs in India. Through interviews with CHWs and their supervisors and log analysis, we examine factors affecting their engagement with ASHABot, and ASHABot's role in addressing CHWs' informational needs. We found that ASHABot provided a private channel for CHWs to ask rudimentary and sensitive questions they hesitated to ask supervisors. CHWs trusted the information they received on ASHABot and treated it as an authoritative resource. CHWs' supervisors expanded their knowledge by contributing answers to questions ASHABot failed to answer, but were concerned about demands on their workload and increased accountability. We emphasize positioning LLMs as supplemental fallible resources within the community healthcare ecosystem, instead of as replacements for supervisor support.

Large Language Models (LLMs) are widely used in healthcare, but limitations like hallucinations, incomplete information, and bias hinder their reliability. To address these, researchers released the Build Your Own expert Bot (BYOeB) platform, enabling developers to create LLM-powered chatbots with integrated expert verification. CataractBot, its first implementation, provides expert-verified responses to cataract surgery questions. A pilot evaluation showed its potential; however the study had a small sample size and was primarily qualitative. In this work, we conducted a large-scale 24-week deployment of CataractBot involving 318 patients and attendants who sent 1,992 messages, with 91.71% of responses verified by seven experts. Analysis of interaction logs revealed that medical questions significantly outnumbered logistical ones, hallucinations were negligible, and experts rated 84.52% of medical answers as accurate. As the knowledge base expanded with expert corrections, system performance improved by 19.02%, reducing expert workload. These insights guide the design of future LLM-powered chatbots., Comment: The first two authors contributed equally to this research

In this work, we explore the effect of neutrino nonstandard interactions (NSI) involving the charm quark at SND@LHC. Using an effective description of new physics in terms of four-fermion operators involving a charm quark, we constrain the Wilson coefficients of the effective interaction from two and three-body charmed meson decays. In our fit, we include charmed meson decays not only to pseudoscalar final states but also to vector final states and include decays to the $\eta$ and $\eta^\prime$ final states. We also consider constraints from charmed baryon decays. We then study the effect of new physics in neutrino scattering processes, involving charm production at SND@LHC, for various benchmark new physics couplings obtained from the low energy fits. Finally, we also study the effects of lepton universality violation (LUV) assuming that the new physics coupling is not lepton universal., Comment: 23 pages, 3 figures; included neutrino production modes that affect results, typos corrected, references added

We provide an algorithm that, given an $n$-vertex $m$-edge Eulerian graph with polynomially bounded weights, computes an $\breve{O}(n\log^{2} n \cdot \varepsilon^{-2})$-edge $\varepsilon$-approximate Eulerian sparsifier with high probability in $\breve{O}(m\log^3 n)$ time (where $\breve{O}(\cdot)$ hides $\text{polyloglog}(n)$ factors). Due to a reduction from [Peng-Song, STOC '22], this yields an $\breve{O}(m\log^3 n + n\log^6 n)$-time algorithm for solving $n$-vertex $m$-edge Eulerian Laplacian systems with polynomially-bounded weights with high probability, improving upon the previous state-of-the-art runtime of $\Omega(m\log^8 n + n\log^{23} n)$. We also give a polynomial-time algorithm that computes $O(\min(n\log n \cdot \varepsilon^{-2} + n\log^{5/3} n \cdot \varepsilon^{-4/3}, n\log^{3/2} n \cdot \varepsilon^{-2}))$-edge sparsifiers, improving the best such sparsity bound of $O(n\log^2 n \cdot \varepsilon^{-2} + n\log^{8/3} n \cdot \varepsilon^{-4/3})$ [Sachdeva-Thudi-Zhao, ICALP '24]. Finally, we show that our techniques extend to yield the first $O(m\cdot\text{polylog}(n))$ time algorithm for computing $O(n\varepsilon^{-1}\cdot\text{polylog}(n))$-edge graphical spectral sketches, as well as a natural Eulerian generalization we introduce. In contrast to prior Eulerian graph sparsification algorithms which used either short cycle or expander decompositions, our algorithms use a simple efficient effective resistance decomposition scheme we introduce. Our algorithms apply a natural sampling scheme and electrical routing (to achieve degree balance) to such decompositions. Our analysis leverages new asymmetric variance bounds specialized to Eulerian Laplacians and tools from discrepancy theory.

Enabling engagement of manga by visually impaired individuals presents a significant challenge due to its inherently visual nature. With the goal of fostering accessibility, this paper aims to generate a dialogue transcript of a complete manga chapter, entirely automatically, with a particular emphasis on ensuring narrative consistency. This entails identifying (i) what is being said, i.e., detecting the texts on each page and classifying them into essential vs non-essential, and (ii) who is saying it, i.e., attributing each dialogue to its speaker, while ensuring the same characters are named consistently throughout the chapter. To this end, we introduce: (i) Magiv2, a model that is capable of generating high-quality chapter-wide manga transcripts with named characters and significantly higher precision in speaker diarisation over prior works; (ii) an extension of the PopManga evaluation dataset, which now includes annotations for speech-bubble tail boxes, associations of text to corresponding tails, classifications of text as essential or non-essential, and the identity for each character box; and (iii) a new character bank dataset, which comprises over 11K characters from 76 manga series, featuring 11.5K exemplar character images in total, as well as a list of chapters in which they appear. The code, trained model, and both datasets can be found at: https://github.com/ragavsachdeva/magi

Long-form question answering (LFQA) aims to provide thorough and in-depth answers to complex questions, enhancing comprehension. However, such detailed responses are prone to hallucinations and factual inconsistencies, challenging their faithful evaluation. This work introduces HaluQuestQA, the first hallucination dataset with localized error annotations for human-written and model-generated LFQA answers. HaluQuestQA comprises 698 QA pairs with 1.8k span-level error annotations for five different error types by expert annotators, along with preference judgments. Using our collected data, we thoroughly analyze the shortcomings of long-form answers and find that they lack comprehensiveness and provide unhelpful references. We train an automatic feedback model on this dataset that predicts error spans with incomplete information and provides associated explanations. Finally, we propose a prompt-based approach, Error-informed refinement, that uses signals from the learned feedback model to refine generated answers, which we show reduces errors and improves answer quality across multiple models. Furthermore, humans find answers generated by our approach comprehensive and highly prefer them (84%) over the baseline answers., Comment: Code and data are available: https://github.com/UKPLab/arxiv2024-lfqa-hallucination

We introduce a novel technique for enhancing the robustness of light-pulse atom interferometers against the pulse infidelities that typically limit their sensitivities. The technique uses quantum optimal control to favorably harness the multipath interference of the stray trajectories produced by imperfect atom-optics operations. We apply this method to a resonant atom interferometer and achieve thousand-fold phase amplification, representing a fifty-fold improvement over the performance observed without optimized control. Moreover, we find that spurious interference can arise from the interplay of spontaneous emission and many-pulse sequences and demonstrate optimization strategies to mitigate this effect. Given the ubiquity of spontaneous emission in quantum systems, these results may be valuable for improving the performance of a diverse array of quantum sensors. We anticipate our findings will significantly benefit the performance of matter-wave interferometers for a variety of applications, including dark matter, dark energy, and gravitational wave detection.

We give the first almost-linear total time algorithm for deciding if a flow of cost at most $F$ still exists in a directed graph, with edge costs and capacities, undergoing decremental updates, i.e., edge deletions, capacity decreases, and cost increases. This implies almost-linear time algorithms for approximating the minimum-cost flow value and $s$-$t$ distance on such decremental graphs. Our framework additionally allows us to maintain decremental strongly connected components in almost-linear time deterministically. These algorithms also improve over the current best known runtimes for statically computing minimum-cost flow, in both the randomized and deterministic settings. We obtain our algorithms by taking the dual perspective, which yields cut-based algorithms. More precisely, our algorithm computes the flow via a sequence of $m^{1+o(1)}$ dynamic min-ratio cut problems, the dual analog of the dynamic min-ratio cycle problem that underlies recent fast algorithms for minimum-cost flow. Our main technical contribution is a new data structure that returns an approximately optimal min-ratio cut in amortized $m^{o(1)}$ time by maintaining a tree-cut sparsifier. This is achieved by devising a new algorithm to maintain the dynamic expander hierarchy of [Goranci-R\"{a}cke-Saranurak-Tan, SODA 2021] that also works in capacitated graphs. All our algorithms are deterministc, though they can be sped up further using randomized techniques while still working against an adaptive adversary., Comment: 61 pages, Accepted to FOCS 2024

In this paper, we propose a new hybrid temporal computing (HTC) framework that leverages both pulse rate and temporal data encoding to design ultra-low energy hardware accelerators. Our approach is inspired by the recently proposed temporal computing, or race logic, which encodes data values as single delays, leading to significantly lower energy consumption due to minimized signal switching. However, race logic is limited in its applications due to inherent restrictions. The new HTC framework overcomes these limitations by encoding signals in both temporal and pulse rate formats for multiplication and in temporal format for propagation. This approach maintains reduced switch energy while being general enough to implement a wide range of arithmetic operations. We demonstrate how HTC multiplication is performed for both unipolar and bipolar data encoding and present the basic designs for multipliers, adders, and MAC units. Additionally, we implement two hardware accelerators: a Finite Impulse Response (FIR) filter and a Discrete Cosine Transform (DCT)/iDCT engine for image compression and DSP applications. Experimental results show that the HTC MAC has a significantly smaller power and area footprint compared to the Unary MAC design and is orders of magnitude faster. Compared to the CBSC MAC, the HTC MAC reduces power consumption by $45.2\%$ and area footprint by $50.13\%$. For the FIR design, the HTC design significantly outperforms the Unary design on all metrics. Compared to the CBSC design, the HTC-based FIR filter reduces power consumption by $36.61\%$ and area cost by $45.85\%$. The HTC-based DCT filter retains the quality of the original image with a decent PSNR, while consuming $23.34\%$ less power and occupying $18.20\%$ less area than the CBSC MAC-based DCT filter., Comment: 7 pages, 8 figures and 3 tables

Graph condensation (GC) is an emerging technique designed to learn a significantly smaller graph that retains the essential information of the original graph. This condensed graph has shown promise in accelerating graph neural networks while preserving performance comparable to those achieved with the original, larger graphs. Additionally, this technique facilitates downstream applications like neural architecture search and deepens our understanding of redundancies in large graphs. Despite the rapid development of GC methods, particularly for node classification, a unified evaluation framework is still lacking to systematically compare different GC methods or clarify key design choices for improving their effectiveness. To bridge these gaps, we introduce \textbf{GC4NC}, a comprehensive framework for evaluating diverse GC methods on node classification across multiple dimensions including performance, efficiency, privacy preservation, denoising ability, NAS effectiveness, and transferability. Our systematic evaluation offers novel insights into how condensed graphs behave and the critical design choices that drive their success. These findings pave the way for future advancements in GC methods, enhancing both performance and expanding their real-world applications. Our code is available at \url{https://github.com/Emory-Melody/GraphSlim/tree/main/benchmark}., Comment: 22 pages

In this paper, we investigate the question of whether the electrical flow routing is a good oblivious routing scheme on an $m$-edge graph $G = (V, E)$ that is a $\Phi$-expander, i.e. where $\lvert \partial S \rvert \geq \Phi \cdot \mathrm{vol}(S)$ for every $S \subseteq V, \mathrm{vol}(S) \leq \mathrm{vol}(V)/2$. Beyond its simplicity and structural importance, this question is well-motivated by the current state-of-the-art of fast algorithms for $\ell_{\infty}$ oblivious routings that reduce to the expander-case which is in turn solved by electrical flow routing. Our main result proves that the electrical routing is an $O(\Phi^{-1} \log m)$-competitive oblivious routing in the $\ell_1$- and $\ell_\infty$-norms. We further observe that the oblivious routing is $O(\log^2 m)$-competitive in the $\ell_2$-norm and, in fact, $O(\log m)$-competitive if $\ell_2$-localization is $O(\log m)$ which is widely believed. Using these three upper bounds, we can smoothly interpolate to obtain upper bounds for every $p \in [2, \infty]$ and $q$ given by $1/p + 1/q = 1$. Assuming $\ell_2$-localization in $O(\log m)$, we obtain that in $\ell_p$ and $\ell_q$, the electrical oblivious routing is $O(\Phi^{-(1-2/p)}\log m)$ competitive. Using the currently known result for $\ell_2$-localization, this ratio deteriorates by at most a sublogarithmic factor for every $p, q \neq 2$. We complement our upper bounds with lower bounds that show that the electrical routing for any such $p$ and $q$ is $\Omega(\Phi^{-(1-2/p)}\log m)$-competitive. This renders our results in $\ell_1$ and $\ell_{\infty}$ unconditionally tight up to constants, and the result in any $\ell_p$- and $\ell_q$-norm to be tight in case of $\ell_2$-localization in $O(\log m)$., Comment: To appear in ICALP 2024

We introduce a comprehensive quantum solver for binary combinatorial optimization problems on gate-model quantum computers that outperforms any published alternative and consistently delivers correct solutions for problems with up to 127 qubits. We provide an overview of the internal workflow, describing the integration of a customized ansatz and variational parameter update strategy, efficient error suppression in hardware execution, and QPU-overhead-free post-processing to correct for bit-flip errors. We benchmark this solver on IBM quantum computers for several classically nontrivial unconstrained binary optimization problems -- the entire optimization is conducted on hardware with no use of classical simulation or prior knowledge of the solution. First, we demonstrate the ability to correctly solve Max-Cut instances for random regular graphs with a variety of densities using up to 120 qubits, where the graph topologies are not matched to device connectivity. Next, we apply the solver to higher-order binary optimization and successfully search for the ground state energy of a 127-qubit spin-glass model with linear, quadratic, and cubic interaction terms. Use of this new quantum solver increases the likelihood of finding the minimum energy by up to $\sim1,500\times$ relative to published results using a DWave annealer, and it can find the correct solution when the annealer fails. Furthermore, for both problem types, the Q-CTRL solver outperforms a heuristic local solver used to indicate the relative difficulty of the problems pursued. Overall, these results represent the largest quantum optimizations successfully solved on hardware to date, and demonstrate the first time a gate-model quantum computer has been able to outperform an annealer for a class of binary optimization problems., Comment: 19 pages, 8 figures

Let $(\tau, V_{\tau})$ be a finite dimensional representation of a maximal compact subgroup $K$ of a connected non-compact semisimple Lie group $G$, and let $\Gamma$ be a uniform torsion-free lattice in $G$. We obtain an infinitesimal version of the celebrated Matsushima-Murakami formula, which relates the dimension of the space of automorphic forms associated to $\tau$ and multiplicities of irreducible $\tau^\vee$-spherical spectra in $L^2(\Gamma \backslash G)$. This result gives a promising tool to study the joint spectra of all central operators on the homogenous bundle associated to the locally symmetric space and hence its infinitesimal $\tau$-isospectrality. Along with this we prove that the almost equality of $\tau$-spherical spectra of two lattices assures the equality of their $\tau$-spherical spectra., Comment: 18 pages, minor revisions

Computer simulations of blood flow in patients with anomalous aortic origin of a coronary artery (AAOCA) have the promise to provide insight into this complex disease. They provide an in-silico experimental platform to explore possible mechanisms of myocardial ischemia, a potentially deadly complication for patients with this defect. This paper focuses on the question of model calibration for fluid-structure interaction models of pediatric AAOCA patients. Imaging and cardiac catheterization data provide partial information for model construction and calibration. However, parameters for downstream boundary conditions needed for these models are difficult to estimate. Further, important model predictions, like fractional flow reserve (FFR), are sensitive to these parameters. We describe an approach to calibrate downstream boundary condition parameters to clinical measurements of resting FFR. The calibrated models are then used to predict FFR at stress, an invasively measured quantity that can be used in the clinical evaluation of these patients. We find reasonable agreement between the model predicted and clinically measured FFR at stress, indicating the credibility of this modeling framework for predicting hemodynamics of pediatric AAOCA patients. This approach could lead to important clinical applications since it may serve as a tool for risk stratifying children with AAOCA.

Local spatial models such as Geographically Weighted Regression (GWR) and Multiscale Geographically Weighted Regression (MGWR) serve as instrumental tools to capture intrinsic contextual effects through the estimates of the local intercepts and behavioral contextual effects through estimates of the local slope parameters. GWR and MGWR provide simple implementation yet powerful frameworks that could be extended to various disciplines that handle spatial data. This bibliography aims to serve as a comprehensive compilation of peer-reviewed papers that have utilized GWR or MGWR as a primary analytical method to conduct spatial analyses and acts as a useful guide to anyone searching the literature for previous examples of local statistical modeling in a wide variety of application fields., Comment: 423 pages

Our goal is to enable embodied agents to learn inductively generalizable spatial concepts, e.g., learning staircase as an inductive composition of towers of increasing height. Given a human demonstration, we seek a learning architecture that infers a succinct ${program}$ representation that explains the observed instance. Additionally, the approach should generalize inductively to novel structures of different sizes or complex structures expressed as a hierarchical composition of previously learned concepts. Existing approaches that use code generation capabilities of pre-trained large (visual) language models, as well as purely neural models, show poor generalization to a-priori unseen complex concepts. Our key insight is to factor inductive concept learning as (i) ${\it Sketch:}$ detecting and inferring a coarse signature of a new concept (ii) ${\it Plan:}$ performing MCTS search over grounded action sequences (iii) ${\it Generalize:}$ abstracting out grounded plans as inductive programs. Our pipeline facilitates generalization and modular reuse, enabling continual concept learning. Our approach combines the benefits of the code generation ability of large language models (LLM) along with grounded neural representations, resulting in neuro-symbolic programs that show stronger inductive generalization on the task of constructing complex structures in relation to LLM-only and neural-only approaches. Furthermore, we demonstrate reasoning and planning capabilities with learned concepts for embodied instruction following.