Search

Developers using LLMs in their applications and agents have provided plenty of anecdotal evidence that in-context-learning (ICL) is fragile. In addition to the quantity and quality of examples, we show that the order in which the in-context examples are listed in the prompt affects the output of the LLM and, consequently, their performance. In this paper, we present OptiSeq, which introduces a score based on log probabilities of LLM outputs to prune the universe of possible example orderings in few-shot ICL and recommend the best order(s) by distinguishing between correct and incorrect outputs resulting from different order permutations. Through a detailed empirical evaluation on multiple LLMs, datasets and prompts, we demonstrate that OptiSeq improves accuracy by 6 - 10.5 percentage points across multiple tasks.

We analyze lensless imaging systems with estimation-theoretic techniques based on Fisher information. Our analysis evaluates multiple optical encoder designs on objects with varying sparsity, in the context of both Gaussian and Poisson noise models. Our simulations verify that lensless imaging system performance is object-dependent and highlight tradeoffs between encoder multiplexing and object sparsity, showing quantitatively that sparse objects tolerate higher levels of multiplexing than dense objects. Insights from our analysis promise to inform and improve optical encoder designs for lensless imaging., Comment: 7 pages, 3 figures

In mission-critical domains such as law enforcement and medical diagnosis, the ability to explain and interpret the outputs of deep learning models is crucial for ensuring user trust and supporting informed decision-making. Despite advancements in explainability, existing methods often fall short in providing explanations that mirror the depth and clarity of those given by human experts. Such expert-level explanations are essential for the dependable application of deep learning models in law enforcement and medical contexts. Additionally, we recognize that most explanations in real-world scenarios are communicated primarily through natural language. Addressing these needs, we propose a novel approach that utilizes characteristic descriptors to explain model decisions by identifying their presence in images, thereby generating expert-like explanations. Our method incorporates a concept bottleneck layer within the model architecture, which calculates the similarity between image and descriptor encodings to deliver inherent and faithful explanations. Through experiments in face recognition and chest X-ray diagnosis, we demonstrate that our approach offers a significant contrast over existing techniques, which are often limited to the use of saliency maps. We believe our approach represents a significant step toward making deep learning systems more accountable, transparent, and trustworthy in the critical domains of face recognition and medical diagnosis.

The increasing prevalence of AI-generated content alongside human-written text underscores the need for reliable discrimination methods. To address this challenge, we propose a novel framework with textual embeddings from Pre-trained Language Models (PLMs) to distinguish AI-generated and human-authored text. Our approach utilizes Embedding Fusion to integrate semantic information from multiple Language Models, harnessing their complementary strengths to enhance performance. Through extensive evaluation across publicly available diverse datasets, our proposed approach demonstrates strong performance, achieving classification accuracy greater than 96% and a Matthews Correlation Coefficient (MCC) greater than 0.93. This evaluation is conducted on a balanced dataset of texts generated from five well-known Large Language Models (LLMs), highlighting the effectiveness and robustness of our novel methodology.

In safeguarding mission-critical systems, such as Unmanned Aerial Vehicles (UAVs), preserving the privacy of path trajectories during navigation is paramount. While the combination of Reinforcement Learning (RL) and Fully Homomorphic Encryption (FHE) holds promise, the computational overhead of FHE presents a significant challenge. This paper proposes an innovative approach that leverages Knowledge Distillation to enhance the practicality of secure UAV navigation. By integrating RL and FHE, our framework addresses vulnerabilities to adversarial attacks while enabling real-time processing of encrypted UAV camera feeds, ensuring data security. To mitigate FHE's latency, Knowledge Distillation is employed to compress the network, resulting in an impressive 18x speedup without compromising performance, as evidenced by an R-squared score of 0.9499 compared to the original model's score of 0.9631. Our methodology underscores the feasibility of processing encrypted data for UAV navigation tasks, emphasizing security alongside performance efficiency and timely processing. These findings pave the way for deploying autonomous UAVs in sensitive environments, bolstering their resilience against potential security threats., Comment: arXiv admin note: text overlap with arXiv:2404.17225

The main areas of research in VLSI system design include area, high speed, and power-efficient data route logic systems. The amount of time needed to send a carry through the adder limits the pace at which addition can occur in digital adders. One of the quickest adders, the Carry Select Adder (CSLA), is utilized by various data processing processors to carry out quick arithmetic operations. It is evident from the CSLA's structure that there is room to cut back on both the area and the delay. This work employs a straightforward and effective gate-level adjustment (in a regular structure) that significantly lowers the CSLA's area and delay. In light of this adjustment Square-Root Carry Select Adder (SQRT CSLA) designs with bit lengths of 8, 16, 32, and 64. When compared to the standard SQRT CSLA, the suggested design significantly reduces both area and latency. Xilinx ISE tool is used for Simulation and synthesis. The performance of the recommended designs in terms of delay is estimated in this study using the standard designs. The study of the findings indicates that the suggested CSLA structure outperforms the standard SQRT CSLA.

The family of mappings of the plane possessing a biquadratic invariant, which is known collectively as QRT maps, is composed of two involutions, one preserving a vertical shift and the other preserving a horizontal shift in the plane. In this paper, we extend the map by replacing each shift by the group operation on each of two families of elliptic curves, whose product forms a bielliptic surface., Comment: 17 pages; typos corrected

The k-Tree algorithm [Wagner 02] is a non-trivial algorithm for the average-case k-SUM problem that has found widespread use in cryptanalysis. Its input consists of k lists, each containing n integers from a range of size m. Wagner's original heuristic analysis suggested that this algorithm succeeds with constant probability if n = m^{1/(\log{k}+1)}, and that in this case it runs in time O(kn). Subsequent rigorous analysis of the algorithm [Lyubashevsky 05, Shallue 08, Joux-Kippen-Loss 24] has shown that it succeeds with high probability if the input list sizes are significantly larger than this. We present a broader rigorous analysis of the k-Tree algorithm, showing upper and lower bounds on its success probability and complexity for any size of the input lists. Our results confirm Wagner's heuristic conclusions, and also give meaningful bounds for a wide range of list sizes that are not covered by existing analyses. We present analytical bounds that are asymptotically tight, as well as an efficient algorithm that computes (provably correct) bounds for a wide range of concrete parameter settings. We also do the same for the k-Tree algorithm over Z_m. Finally, we present experimental evaluation of the tightness of our results.

Introduction: Altered neurometabolism is an important pathological mechanism in many neurological diseases and brain cancer, which can be mapped non-invasively by Magnetic Resonance Spectroscopic Imaging (MRSI). Advanced MRSI using non-cartesian compressed-sense acquisition enables fast high-resolution metabolic imaging but has lengthy reconstruction times that limits throughput and needs expert user interaction. Here, we present a robust and efficient Deep Learning reconstruction to obtain high-quality metabolic maps. Methods: Fast high-resolution whole-brain metabolic imaging was performed at 3.4 mm$^3$ isotropic resolution with acquisition times between 4:11-9:21 min:s using ECCENTRIC pulse sequence on a 7T MRI scanner. Data were acquired in a high-resolution phantom and 27 human participants, including 22 healthy volunteers and 5 glioma patients. A deep neural network using recurring interlaced convolutional layers with joint dual-space feature representation was developed for deep learning ECCENTRIC reconstruction (Deep-ER). 21 subjects were used for training and 6 subjects for testing. Deep-ER performance was compared to conventional iterative Total Generalized Variation reconstruction using image and spectral quality metrics. Results: Deep-ER demonstrated 600-fold faster reconstruction than conventional methods, providing improved spatial-spectral quality and metabolite quantification with 12%-45% (P<0.05) higher signal-to-noise and 8%-50% (P<0.05) smaller Cramer-Rao lower bounds. Metabolic images clearly visualize glioma tumor heterogeneity and boundary. Conclusion: Deep-ER provides efficient and robust reconstruction for sparse-sampled MRSI. The accelerated acquisition-reconstruction MRSI is compatible with high-throughput imaging workflow. It is expected that such improved performance will facilitate basic and clinical MRSI applications.

In this publication, leading feminist thinkers from Africa, the Middle East and North Africa, South Asia, Latin America and the Caribbean, share perspectives on why we must move beyond gross domestic product (GDP) as the primary measure of societal progress. Drawing on regional examples, they demonstrate how the GDP-growth obsession is skewing policy-making towards measures that drive austerity, debt, corporate power, and exploitative labour practices, whilst invisibilising unpaid care and domestic work. In a call to action, the authors highlight Indigenous and feminist alternatives that need to be pursued to catalyse a shift towards economies that centre care of people and the planet. Illustrations by Vidushi Yadav and Medhavini Yadav.

We consider the Riemann-Hilbert correspondence associated with the $q$-difference sixth Painlev\'e equation in the crystal limit, i.e. $q\rightarrow 0$, and show two main results. First, the limit of this generically highly transcendental mapping is shown to exist. Second, we show that the limiting map is bi-rational and describe it explicitly., Comment: 22 pages, 1 figure

This study aims to learn a translation from visible to infrared imagery, bridging the domain gap between the two modalities so as to improve accuracy on downstream tasks including object detection. Previous approaches attempt to perform bi-domain feature fusion through iterative optimization or end-to-end deep convolutional networks. However, we pose the problem as similar to that of image translation, adopting a two-stage training strategy with a Generative Adversarial Network and an object detection model. The translation model learns a conversion that preserves the structural detail of visible images while preserving the texture and other characteristics of infrared images. Images so generated are used to train standard object detection frameworks including Yolov5, Mask and Faster RCNN. We also investigate the usefulness of integrating a super-resolution step into our pipeline to further improve model accuracy, and achieve an improvement of as high as 5.3% mAP.

Large language models have demonstrated impressive performance on commonsense tasks; however, these tasks are often posed as multiple-choice questions, allowing models to exploit systematic biases. Commonsense is also inherently probabilistic with multiple correct answers. The purpose of "boiling water" could be making tea and cooking, but it also could be killing germs. Existing tasks do not capture the probabilistic nature of common sense. To this end, we present commonsense frame completion (CFC), a new generative task that evaluates common sense via multiple open-ended generations. We also propose a method of probabilistic evaluation that strongly correlates with human judgments. Humans drastically outperform strong language model baselines on our dataset, indicating this approach is both a challenging and useful evaluation of machine common sense., Comment: ACL 2024 Camera Ready

In this paper, we consider a six parameter family of affine Segre surfaces embedded in $\mathbb C^6$. For generic values of the parameters, this family is associated to the $q$-difference sixth Painlev\'e equation. We show that different limiting forms of this family give Segre surfaces that are isomorphic as affine varieties to the the monodromy manifolds of each Painlev\'e differential equation., Comment: 66 pages, 4 figures

With the rapid surge in the prevalence of Large Language Models (LLMs), individuals are increasingly turning to conversational AI for initial insights across various domains, including health-related inquiries such as disease diagnosis. Many users seek potential causes on platforms like ChatGPT or Bard before consulting a medical professional for their ailment. These platforms offer valuable benefits by streamlining the diagnosis process, alleviating the significant workload of healthcare practitioners, and saving users both time and money by avoiding unnecessary doctor visits. However, Despite the convenience of such platforms, sharing personal medical data online poses risks, including the presence of malicious platforms or potential eavesdropping by attackers. To address privacy concerns, we propose a novel framework combining FHE and Deep Learning for a secure and private diagnosis system. Operating on a question-and-answer-based model akin to an interaction with a medical practitioner, this end-to-end secure system employs Fully Homomorphic Encryption (FHE) to handle encrypted input data. Given FHE's computational constraints, we adapt deep neural networks and activation functions to the encryted domain. Further, we also propose a faster algorithm to compute summation of ciphertext elements. Through rigorous experiments, we demonstrate the efficacy of our approach. The proposed framework achieves strict security and privacy with minimal loss in performance.

Autonomous Unmanned Aerial Vehicles (UAVs) have become essential tools in defense, law enforcement, disaster response, and product delivery. These autonomous navigation systems require a wireless communication network, and of late are deep learning based. In critical scenarios such as border protection or disaster response, ensuring the secure navigation of autonomous UAVs is paramount. But, these autonomous UAVs are susceptible to adversarial attacks through the communication network or the deep learning models - eavesdropping / man-in-the-middle / membership inference / reconstruction. To address this susceptibility, we propose an innovative approach that combines Reinforcement Learning (RL) and Fully Homomorphic Encryption (FHE) for secure autonomous UAV navigation. This end-to-end secure framework is designed for real-time video feeds captured by UAV cameras and utilizes FHE to perform inference on encrypted input images. While FHE allows computations on encrypted data, certain computational operators are yet to be implemented. Convolutional neural networks, fully connected neural networks, activation functions and OpenAI Gym Library are meticulously adapted to the FHE domain to enable encrypted data processing. We demonstrate the efficacy of our proposed approach through extensive experimentation. Our proposed approach ensures security and privacy in autonomous UAV navigation with negligible loss in performance.

Modern face recognition systems utilize deep neural networks to extract salient features from a face. These features denote embeddings in latent space and are often stored as templates in a face recognition system. These embeddings are susceptible to data leakage and, in some cases, can even be used to reconstruct the original face image. To prevent compromising identities, template protection schemes are commonly employed. However, these schemes may still not prevent the leakage of soft biometric information such as age, gender and race. To alleviate this issue, we propose a novel technique that combines Fully Homomorphic Encryption (FHE) with an existing template protection scheme known as PolyProtect. We show that the embeddings can be compressed and encrypted using FHE and transformed into a secure PolyProtect template using polynomial transformation, for additional protection. We demonstrate the efficacy of the proposed approach through extensive experiments on multiple datasets. Our proposed approach ensures irreversibility and unlinkability, effectively preventing the leakage of soft biometric attributes from face embeddings without compromising recognition accuracy.

Recent advances in artificial intelligence (AI) capabilities have increased the eagerness of companies to integrate AI into software systems. While AI can be used to have a positive impact on several dimensions of sustainability, this is often overshadowed by its potential negative influence. While many studies have explored sustainability factors in isolation, there is insufficient holistic coverage of potential sustainability benefits or costs that practitioners need to consider during decision-making for AI adoption. We therefore aim to synthesize trade-offs related to sustainability in the context of integrating AI into software systems. We want to make the sustainability benefits and costs of integrating AI more transparent and accessible for practitioners. The study was conducted in collaboration with a Dutch financial organization. We first performed a rapid review that led to the inclusion of 151 research papers. Afterward, we conducted six semi-structured interviews to enrich the data with industry perspectives. The combined results showcase the potential sustainability benefits and costs of integrating AI. The labels synthesized from the review regarding potential sustainability benefits were clustered into 16 themes, with "energy management" being the most frequently mentioned one. 11 themes were identified in the interviews, with the top mentioned theme being "employee wellbeing". Regarding sustainability costs, the review discovered seven themes, with "deployment issues" being the most popular one, followed by "ethics & society". "Environmental issues" was the top theme from the interviews. Our results provide valuable insights to organizations and practitioners for understanding the potential sustainability implications of adopting AI., Comment: Accepted for publication at the 8th International Workshop on Green and Sustainable Software (GREENS'24), collocated with ICSA'24

Let $X$ be a closed, connected, oriented surface of genus $g$, with a hyperbolic metric chosen at random according to the Weil--Petersson measure on the moduli space of Riemannian metrics. Let $\lambda_1=\lambda_1(X)$ bethe first non-zero eigenvalue of the Laplacian on $X$ or, in other words, the spectral gap.In this paper we give a full road-map to prove that for arbitrarily small~$\alpha>0$,\begin{align*} \Pwp{\lambda_1 \leq \frac{1}{4} - \alpha^2 } \Lim_{g\To +\infty} 0.\end{align*}The full proofs are deferred to separate papers.

Usually when solving differential or difference equations via series solutions one encounters divergent series in which the coefficients grow like a factorial. Surprisingly, in the $q$-world the $n$th coefficient is often of the size $q^{-\frac12 n(n-1)}$, in which $q\in(0,1)$ is fixed. Hence, the divergence is much stronger, and one has to introduce alternative Borel and Laplace transforms to make sense of these formal series. We will discuss exponentially-improved asymptotics for the basic hypergeometric function ${}_2\phi_0$ and for solutions of the $q$-difference first Painlev\'e equation $q{\rm P}_{\rm I}$. These are optimal truncated expansions, and re-expansions in terms of new $q$-hyperterminant functions. The re-expansions do incorporate the Stokes phenomena., Comment: 15 pages, 1 figure