Search

Personalized search represents a problem where retrieval models condition on historical user interaction data in order to improve retrieval results. However, personalization is commonly perceived as opaque and not amenable to control by users. Further, personalization necessarily limits the space of items that users are exposed to. Therefore, prior work notes a tension between personalization and users' ability for discovering novel items. While discovery of novel items in personalization setups may be resolved through search result diversification, these approaches do little to allow user control over personalization. Therefore, in this paper, we introduce an approach for controllable personalized search. Our model, CtrlCE presents a novel cross-encoder model augmented with an editable memory constructed from users historical items. Our proposed memory augmentation allows cross-encoder models to condition on large amounts of historical user data and supports interaction from users permitting control over personalization. Further, controllable personalization for search must account for queries which don't require personalization, and in turn user control. For this, we introduce a calibrated mixing model which determines when personalization is necessary. This allows system designers using CtrlCE to only obtain user input for control when necessary. In multiple datasets of personalized search, we show CtrlCE to result in effective personalization as well as fulfill various key goals for controllable personalized search., Comment: Work in progress

Personalization of Large Language Models (LLMs) has recently become increasingly important with a wide range of applications. Despite the importance and recent progress, most existing works on personalized LLMs have focused either entirely on (a) personalized text generation or (b) leveraging LLMs for personalization-related downstream applications, such as recommendation systems. In this work, we bridge the gap between these two separate main directions for the first time by introducing a taxonomy for personalized LLM usage and summarizing the key differences and challenges. We provide a formalization of the foundations of personalized LLMs that consolidates and expands notions of personalization of LLMs, defining and discussing novel facets of personalization, usage, and desiderata of personalized LLMs. We then unify the literature across these diverse fields and usage scenarios by proposing systematic taxonomies for the granularity of personalization, personalization techniques, datasets, evaluation methods, and applications of personalized LLMs. Finally, we highlight challenges and important open problems that remain to be addressed. By unifying and surveying recent research using the proposed taxonomies, we aim to provide a clear guide to the existing literature and different facets of personalization in LLMs, empowering both researchers and practitioners.

GPT-4o is an autoregressive omni model that accepts as input any combination of text, audio, image, and video, and generates any combination of text, audio, and image outputs. It's trained end-to-end across text, vision, and audio, meaning all inputs and outputs are processed by the same neural network. GPT-4o can respond to audio inputs in as little as 232 milliseconds, with an average of 320 milliseconds, which is similar to human response time in conversation. It matches GPT-4 Turbo performance on text in English and code, with significant improvement on text in non-English languages, while also being much faster and 50\% cheaper in the API. GPT-4o is especially better at vision and audio understanding compared to existing models. In line with our commitment to building AI safely and consistent with our voluntary commitments to the White House, we are sharing the GPT-4o System Card, which includes our Preparedness Framework evaluations. In this System Card, we provide a detailed look at GPT-4o's capabilities, limitations, and safety evaluations across multiple categories, focusing on speech-to-speech while also evaluating text and image capabilities, and measures we've implemented to ensure the model is safe and aligned. We also include third-party assessments on dangerous capabilities, as well as discussion of potential societal impacts of GPT-4o's text and vision capabilities.

This paper investigates the design of a unified search engine to serve multiple retrieval-augmented generation (RAG) agents, each with a distinct task, backbone large language model (LLM), and retrieval-augmentation strategy. We introduce an iterative approach where the search engine generates retrieval results for these RAG agents and gathers feedback on the quality of the retrieved documents during an offline phase. This feedback is then used to iteratively optimize the search engine using a novel expectation-maximization algorithm, with the goal of maximizing each agent's utility function. Additionally, we adapt this approach to an online setting, allowing the search engine to refine its behavior based on real-time individual agents feedback to better serve the results for each of them. Experiments on diverse datasets from the Knowledge-Intensive Language Tasks (KILT) benchmark demonstrates that our approach significantly on average outperforms competitive baselines across 18 RAG models. We also demonstrate that our method effectively ``personalizes'' the retrieval process for each RAG agent based on the collected feedback. Finally, we provide a comprehensive ablation study to explore various aspects of our method.

This paper provides an overview of a problem in information-theoretic privacy mechanism design, addressing two scenarios in which private data is either observable or hidden. In each scenario, different privacy measures are used, including bounded mutual information and two types of per-letter privacy constraints. Considering the first scenario, an agent observes useful data that is correlated with private data, and wants to disclose the useful information to a user. Due to the privacy concerns, direct disclosure is prohibited. Hence, a privacy mechanism is designed to generate disclosed data which maximizes the revealed information about the useful data while satisfying a privacy constraint. In the second scenario, the agent has additionally access to the private data. We discuss how the Functional Representation Lemma, the Strong Functional Representation Lemma, and their extended versions are useful for designing low-complexity privacy mechanisms that achieve optimal privacy-utility trade-offs under certain constraints. Furthermore, another privacy design problem is presented where part of the private attribute is more private than the remaining part. Finally, we provide applications including semantic communications, caching and delivery, and compression designs, where the approach can be applied., Comment: arXiv admin note: text overlap with arXiv:2212.12475

Integrated sensing and communications (ISAC) is a promising component of 6G networks, fusing communication and radar technologies to facilitate new services. Additionally, the use of extremely large-scale antenna arrays (ELLA) at the ISAC common receiver not only facilitates terahertz-rate communication links but also significantly enhances the accuracy of target detection in radar applications. In practical scenarios, communication scatterers and radar targets often reside in close proximity to the ISAC receiver. This, combined with the use of ELLA, fundamentally alters the electromagnetic characteristics of wireless and radar channels, shifting from far-field planar-wave propagation to near-field spherical wave propagation. Under the far-field planar-wave model, the phase of the array response vector varies linearly with the antenna index. In contrast, in the near-field spherical wave model, this phase relationship becomes nonlinear. This shift presents a fundamental challenge: the widely-used Fourier analysis can no longer be directly applied for target detection and communication channel estimation at the ISAC common receiver. In this work, we propose a feasible solution to address this fundamental issue. Specifically, we demonstrate that there exists a high-dimensional space in which the phase nonlinearity can be expressed as linear. Leveraging this insight, we develop a lifted super-resolution framework that simultaneously performs communication channel estimation and extracts target parameters with high precision.

Evaluating the creativity of large language models (LLMs) in story writing is difficult because LLM-generated stories could seemingly look creative but be very similar to some existing stories in their huge and proprietary training corpus. To overcome this challenge, we introduce a novel benchmark dataset with varying levels of prompt specificity: CS4 ($\mathbf{C}$omparing the $\mathbf{S}$kill of $\mathbf{C}$reating $\mathbf{S}$tories by $\mathbf{C}$ontrolling the $\mathbf{S}$ynthesized $\mathbf{C}$onstraint $\mathbf{S}$pecificity). By increasing the number of requirements/constraints in the prompt, we can increase the prompt specificity and hinder LLMs from retelling high-quality narratives in their training data. Consequently, CS4 empowers us to indirectly measure the LLMs' creativity without human annotations. Our experiments on LLaMA, Gemma, and Mistral not only highlight the creativity challenges LLMs face when dealing with highly specific prompts but also reveal that different LLMs perform very differently under different numbers of constraints and achieve different balances between the model's instruction-following ability and narrative coherence. Additionally, our experiments on OLMo suggest that Learning from Human Feedback (LHF) can help LLMs select better stories from their training data but has limited influence in boosting LLMs' ability to produce creative stories that are unseen in the training corpora. The benchmark is released at https://github.com/anirudhlakkaraju/cs4_benchmark.

A statistical cache-aided compression problem with a privacy constraint is studied, where a server has access to a database of $N$ files, $(Y_1,...,Y_N)$, each of size $F$ bits and is linked through a shared channel to $K$ users, where each has access to a local cache memory of size $MF$ bits. During the placement phase, the server fills the users' caches without prior knowledge of their demands, while the delivery phase takes place after the users send their demands to the server. We assume that each file in database $Y_i$ is arbitrarily correlated with a private attribute $X$, and an adversary is assumed to have access to the shared channel. The users and the server have access to a shared key $W$. The goal is to design the cache contents and the delivered message $\cal C$ such that the average length of $\mathcal{C}$ is minimized, while satisfying: i. The response $\cal C$ does not reveal any information about $X$, i.e., $I(X;\mathcal{C})=0$; ii. User $i$ can decode its demand, $Y_{d_i}$, by using the shared key $W$, $\cal C$, and its local cache $Z_i$. In a previous work, we have proposed a variable-length coding scheme that combines privacy-aware compression with coded caching techniques. In this paper, we propose a new achievability scheme using minimum entropy coupling concept and a greedy entropy-based algorithm. We show that the proposed scheme improves the previous results. Moreover, considering two special cases we improve the obtained bounds using the common information concept.

Privacy-preserving methods for personalizing large language models (LLMs) are relatively under-explored. There are two schools of thought on this topic: (1) generating personalized outputs by personalizing the input prompt through retrieval augmentation from the user's personal information (RAG-based methods), and (2) parameter-efficient fine-tuning of LLMs per user that considers efficiency and space limitations (PEFT-based methods). This paper presents the first systematic comparison between two approaches on a wide range of personalization tasks using seven diverse datasets. Our results indicate that RAG-based and PEFT-based personalization methods on average yield 14.92% and 1.07% improvements over the non-personalized LLM, respectively. We find that combining RAG with PEFT elevates these improvements to 15.98%. Additionally, we identify a positive correlation between the amount of user data and PEFT's effectiveness, indicating that RAG is a better choice for cold-start users (i.e., user's with limited personal data).

The equation of motion is the important equation for obtain the extra force and Raychaudhuri equation. By considering an explicitly coupling between an arbitrary function of the scalar Gauss-Bonnet, G and the Lagrangian density of matter, it is shown that an extra force normal to their four-velocities arises. In this paper, we obtain the extra force and the generalized Raychaudhuri equation in F(G) modified theory of gravity in an imperfect fluid for the massive particle by divergence of energy momentum tensor so we earn extra force an Raychaudhuri equation in a compared with f(R) modified gravity for perfect fluid this conclusion giving the evolution of the kinematical quantities and describing the relative accelerations of nearby particles .

In this article, we study the fundamental limits in the design of fair and/or private representations achieving perfect demographic parity and/or perfect privacy through the lens of information theory. More precisely, given some useful data $X$ that we wish to employ to solve a task $T$, we consider the design of a representation $Y$ that has no information of some sensitive attribute or secret $S$, that is, such that $I(Y;S) = 0$. We consider two scenarios. First, we consider a design desiderata where we want to maximize the information $I(Y;T)$ that the representation contains about the task, while constraining the level of compression (or encoding rate), that is, ensuring that $I(Y;X) \leq r$. Second, inspired by the Conditional Fairness Bottleneck problem, we consider a design desiderata where we want to maximize the information $I(Y;T|S)$ that the representation contains about the task which is not shared by the sensitive attribute or secret, while constraining the amount of irrelevant information, that is, ensuring that $I(Y;X|T,S) \leq r$. In both cases, we employ extended versions of the Functional Representation Lemma and the Strong Functional Representation Lemma and study the tightness of the obtained bounds. Every result here can also be interpreted as a coding with perfect privacy problem by considering the sensitive attribute as a secret.

Finite-state abstractions (a.k.a. symbolic models) present a promising avenue for the formal verification and synthesis of controllers in continuous-space control systems. These abstractions provide simplified models that capture the fundamental behaviors of the original systems. However, the creation of such abstractions typically relies on the availability of precise knowledge concerning system dynamics, which might not be available in many real-world applications. In this work, we introduce an innovative, data-driven, and compositional approach to generate finite abstractions for interconnected systems that consist of discrete-time control subsystems with unknown dynamics. These subsystems interact through an unknown static interconnection map. Our methodology for abstracting the interconnected system involves constructing abstractions for individual subsystems and incorporating an abstraction of the interconnection map., Comment: This manuscript of 19 pages and 7 figures is a preprint under review with a journal

Diagnosability is a system theoretical property characterizing whether fault occurrences in a system can always be detected within a finite time. In this paper, we investigate the verification of diagnosability for cyber-physical systems with continuous state sets. We develop an abstraction-free and automata-based framework to verify (the lack of) diagnosability, leveraging a notion of hybrid barrier certificates. To this end, we first construct a (delta,K)-deterministic finite automaton that captures the occurrence of faults targeted for diagnosis. Then, the verification of diagnosability property is converted into a safety verification problem over a product system between the automaton and the augmented version of the dynamical system. We demonstrate that this verification problem can be addressed by computing hybrid barrier certificates for the product system. To this end, we introduce two systematic methods, leveraging sum-of-squares programming and counter-example guided inductive synthesis to search for such certificates. Additionally, if the system is found to be diagnosable, we propose methodologies to construct a diagnoser to identify fault occurrences online. Finally, we showcase the effectiveness of our methods through a case study.

This paper studies the last iterate of subgradient method with Polyak step size when applied to the minimization of a nonsmooth convex function with bounded subgradients. We show that the subgradient method with Polyak step size achieves a convergence rate $\mathcal{O}\left(\tfrac{1}{\sqrt[4]{N}}\right)$ in terms of the final iterate. An example is provided to show that this rate is exact and cannot be improved. We introduce an adaptive Polyak step size for which the subgradient method enjoys a convergence rate $\mathcal{O}\left(\tfrac{1}{\sqrt{N}}\right)$ for the last iterate. Its convergence rate matches exactly the lower bound on the performance of any black-box method on the considered problem class. Additionally, we propose an adaptive Polyak method with a momentum term, where the step sizes are independent of the number of iterates. We establish that the algorithm also attains the optimal convergence rate. We investigate the alternating projection method. We derive a convergence rate $\left( \frac{2N }{ 2N+1 } \right)^N\tfrac{R}{\sqrt{2N+1}}$ for the last iterate, where $R$ is a bound on the distance between the initial iterate and a solution. An example is also provided to illustrate the exactness of the rate.

In this paper, we consider the collaborative attitude estimation problem for a multi-agent system. The agents are equipped with sensors that provide directional measurements and relative attitude measurements. We present a bottom-up approach where each agent runs an extended Kalman filter (EKF) locally using directional measurements and augments this with relative attitude measurements provided by neighbouring agents. The covariance estimates of the relative attitude measurements are geometrically corrected to compensate for relative attitude between the agent that makes the measurement and the agent that uses the measurement before being fused with the local estimate using the convex combination ellipsoid (CCE) method to avoid data incest. Simulations are undertaken to numerically evaluate the performance of the proposed algorithm., Comment: To be presented at IFAC MTNS 2024

In the field of language modeling, models augmented with retrieval components have emerged as a promising solution to address several challenges faced in the natural language processing (NLP) field, including knowledge grounding, interpretability, and scalability. Despite the primary focus on NLP, we posit that the paradigm of retrieval-enhancement can be extended to a broader spectrum of machine learning (ML) such as computer vision, time series prediction, and computational biology. Therefore, this work introduces a formal framework of this paradigm, Retrieval-Enhanced Machine Learning (REML), by synthesizing the literature in various domains in ML with consistent notations which is missing from the current literature. Also, we found that while a number of studies employ retrieval components to augment their models, there is a lack of integration with foundational Information Retrieval (IR) research. We bridge this gap between the seminal IR research and contemporary REML studies by investigating each component that comprises the REML framework. Ultimately, the goal of this work is to equip researchers across various disciplines with a comprehensive, formally structured framework of retrieval-enhanced models, thereby fostering interdisciplinary future research.

Knowledge-intensive visual question answering requires models to effectively use external knowledge to help answer visual questions. A typical pipeline includes a knowledge retriever and an answer generator. However, a retriever that utilizes local information, such as an image patch, may not provide reliable question-candidate relevance scores. Besides, the two-tower architecture also limits the relevance score modeling of a retriever to select top candidates for answer generator reasoning. In this paper, we introduce an additional module, a multi-modal reranker, to improve the ranking quality of knowledge candidates for answer generation. Our reranking module takes multi-modal information from both candidates and questions and performs cross-item interaction for better relevance score modeling. Experiments on OK-VQA and A-OKVQA show that multi-modal reranker from distant supervision provides consistent improvements. We also find a training-testing discrepancy with reranking in answer generation, where performance improves if training knowledge candidates are similar to or noisier than those used in testing.

At its core, information access and seeking is an interactive process. In existing search engines, interactions are limited to a few pre-defined actions, such as "requery", "click on a document", "scrolling up/down", "going to the next result page", "leaving the search engine", etc. A major benefit of moving towards generative IR systems is enabling users with a richer expression of information need and feedback and free-form interactions in natural language and beyond. In other words, the actions users take are no longer limited by the clickable links and buttons available on the search engine result page and users can express themselves freely through natural language. This can go even beyond natural language, through images, videos, gestures, and sensors using multi-modal generative IR systems. This chapter briefly discusses the role of interaction in generative IR systems. We will first discuss different ways users can express their information needs by interacting with generative IR systems. We then explain how users can provide explicit or implicit feedback to generative IR systems and how they can consume such feedback. Next, we will cover how users interactively can refine retrieval results. We will expand upon mixed-initiative interactions and discuss clarification and preference elicitation in more detail. We then discuss proactive generative IR systems, including context-aware recommendation, following up past conversations, contributing to multi-party conversations, and feedback requests. Providing explanation is another interaction type that we briefly discuss in this chapter. We will also briefly describe multi-modal interactions in generative information retrieval. Finally, we describe emerging frameworks and solutions for user interfaces with generative AI systems., Comment: Draft of a chapter intended to appear in a forthcoming book on generative information retrieval, co-edited by Chirag Shah and Ryen White

We compute probabilistic controlled invariant sets for nonlinear systems using Gaussian process state space models, which are data-driven models that account for unmodeled and unknown nonlinear dynamics. We investigate the relationship between robust and probabilistic invariance, leveraging this relationship to design state-feedback controllers that maximize the probability of the system staying within the probabilistic controlled invariant set. We propose a semi-definite-programming-based optimization scheme for designing the state-feedback controllers subject to input constraints. The effectiveness of our results are demonstrated and validated on a quadrotor, both in simulation and on a physical platform.

Instruction-tuned large language models have demonstrated remarkable capabilities in following human instructions across various domains. However, their proficiency remains notably deficient in many low-resource languages. To address this challenge, we begin by introducing FarsInstruct a comprehensive instruction dataset designed to enhance the instruction following ability of large language models specifically for the Persian language a significant yet underrepresented language globally. FarsInstruct encompasses a wide range of task types and datasets, each containing a mix of straightforward to complex manual written instructions, as well as translations from the Public Pool of Prompts, ensuring a rich linguistic and cultural representation. Furthermore, we introduce Co-CoLA, a framework designed to enhance the multi-task adaptability of LoRA-tuned models. Through extensive experimental analyses, our study showcases the effectiveness of the FarsInstruct dataset coupled with training by the Co-CoLA framework, in improving the performance of large language models within the Persian context. As of the current writing, FarsInstruct comprises 197 templates across 21 distinct datasets, and we intend to update it consistently, thus augmenting its applicability.

Broadband NIRS (bNIRS) is an extension of fNIRS that provides the same assessment of oxygenation biomarkers along with a valuable marker for oxygen metabolism at a cellular level, the oxidation state of cytochrome-c-oxidase (oxCCO). bNIRS implements many (100s) NIR wavelengths in the full NIR spectrum to address this and provide insight to tissue energetics. To supply these many wavelengths of light, broadband sources are required, and spectrometers are employed to distinguish power per wavelength. Current multi-channel bNIRS instruments are bulky and only semi-portable due to technological limitations. We propose a design for a bNIRS device that has been miniaturized to allow for portable use. This design leverages the innovations in photonic devices that have created a new line of microspectrometers and broadband NIR high-power LEDs; the Hamamatsu SMD-type spectrometer C14384MA and the Ushio SMBBIR45-1100 LED. This first-of-itskind device, referred to as microCYRIL (after its two predecessors CYRIL and miniCYRIL), has been developed for oxygenation and metabolism measurements with dual channel operation. To verify functionality, concentration changes in oxygenated (HbO2) and deoxygenated (HHb) haemoglobin and oxCCO were successfully tracked during a cuff-induced venous and arterial occlusion., Comment: To be published in Advances in Experimental Medicine and Biology

Topic models are widely used to analyze document collections. While they are valuable for discovering latent topics in a corpus when analysts are unfamiliar with the corpus, analysts also commonly start with an understanding of the content present in a corpus. This may be through categories obtained from an initial pass over the corpus or a desire to analyze the corpus through a predefined set of categories derived from a high level theoretical framework (e.g. political ideology). In these scenarios analysts desire a topic modeling approach which incorporates their understanding of the corpus while supporting various forms of interaction with the model. In this work, we present EdTM, as an approach for label name supervised topic modeling. EdTM models topic modeling as an assignment problem while leveraging LM/LLM based document-topic affinities and using optimal transport for making globally coherent topic-assignments. In experiments, we show the efficacy of our framework compared to few-shot LLM classifiers, and topic models based on clustering and LDA. Further, we show EdTM's ability to incorporate various forms of analyst feedback and while remaining robust to noisy analyst inputs., Comment: Pre-print; Work in progress

Aditya-L1 is the first Indian solar mission placed at the first Lagrangian (L1) point to study the Sun. A fluxgate magnetometer (MAG) is one of the seven payloads and one of the three in-situ payloads onboard to measure the interplanetary magnetic field (IMF) coming from the Sun towards the Earth. At present, the Aditya-L1 spacecraft is in a halo-orbit around the L1 point and the MAG payload is ON is continuously measuring the IMF. This paper presents the first measurements of the IMF by MAG.

This study is the first attempt to explore the impact of clarification question modality on user preference in search engines. We introduce the multi-modal search clarification dataset, MIMICS-MM, containing clarification questions with associated expert-collected and model-generated images. We analyse user preferences over different clarification modes of text, image, and combination of both through crowdsourcing by taking into account image and text quality, clarity, and relevance. Our findings demonstrate that users generally prefer multi-modal clarification over uni-modal approaches. We explore the use of automated image generation techniques and compare the quality, relevance, and user preference of model-generated images with human-collected ones. The study reveals that text-to-image generation models, such as Stable Diffusion, can effectively generate multi-modal clarification questions. By investigating multi-modal clarification, this research establishes a foundation for future advancements in search systems.

Long-text generation is seemingly ubiquitous in real-world applications of large language models such as generating an email or writing a review. Despite the fundamental importance and prevalence of long-text generation in many practical applications, existing work on personalized generation has focused on the generation of very short text. To overcome these limitations, we study the problem of personalized long-text generation, that is, generating long-text that is personalized for a specific user while being practically useful for the vast majority of real-world applications that naturally require the generation of longer text. In this work, we demonstrate the importance of user-specific personalization for long-text generation tasks and develop the Long-text Language Model Personalization (LongLaMP) Benchmark. LongLaMP provides a comprehensive and diverse evaluation framework for personalized long-text generation. Extensive experiments on LongLaMP for zero-shot and fine-tuned language tasks demonstrate the effectiveness of the proposed benchmark and its utility for developing and evaluating techniques for personalized long-text generation across a wide variety of long-text generation tasks. The results highlight the importance of personalization across a wide variety of long-text generation tasks. Finally, we release the benchmark for others to use for this important problem.

Let U be a $\phi $-Johnson amenable Banach algebra in which $\phi$ is a non-zero multiplicative linear functional on U. Suppose that X is a Banach U-bimodule such that $a.x=\phi(a)x$ for all a in U and x in X or $x.a=\phi(a)x$ for all a in U and x in X. We show that every continuous Jordan derivation from U to X is a derivation, and every continuous Lie derivation from U to X decomposed into the sum of a continuous derivation and a continuous center-valued trace.

Let $\mathscr{A}$ be a unital $C^*$-algebra with unit $e$ and let $\nu\in(0, 1)$. We introduce the concept of the $\nu$-weighted contraharmonic of two positive definite elements $a$ and $b$ of $\mathscr{A}$ by \begin{align*} {C}_{\nu}(a, b):= (1-\nu)\nu^{-1}b + \nu (1-\nu)^{-1}a - \left((1-\nu)a^{-1}+\nu b^{-1}\right)^{-1}. \end{align*} We show that \begin{align*} {C}_{\nu}(a, b)= \displaystyle{\max_{x+y=e}}\left\{(1-\nu)^{-1}\left(\nu a - x^*ax\right) + \nu^{-1}\left((1-\nu)b - y^*by\right)\right\}, \end{align*} and then apply it to present some properties of this weighted mean.

Sexism in online content is a pervasive issue that necessitates effective classification techniques to mitigate its harmful impact. Online platforms often have sexist comments and posts that create a hostile environment, especially for women and minority groups. This content not only spreads harmful stereotypes but also causes emotional harm. Reliable methods are essential to find and remove sexist content, making online spaces safer and more welcoming. Therefore, the sEXism Identification in Social neTworks (EXIST) challenge addresses this issue at CLEF 2024. This study aims to improve sexism identification in bilingual contexts (English and Spanish) by leveraging natural language processing models. The tasks are to determine whether a text is sexist and what the source intention behind it is. We fine-tuned the XLM-RoBERTa model and separately used GPT-3.5 with few-shot learning prompts to classify sexist content. The XLM-RoBERTa model exhibited robust performance in handling complex linguistic structures, while GPT-3.5's few-shot learning capability allowed for rapid adaptation to new data with minimal labeled examples. Our approach using XLM-RoBERTa achieved 4th place in the soft-soft evaluation of Task 1 (sexism identification). For Task 2 (source intention), we achieved 2nd place in the soft-soft evaluation., Comment: 8 pages, 6 tables

Image generation and editing have seen a great deal of advancements with the rise of large-scale diffusion models that allow user control of different modalities such as text, mask, depth maps, etc. However, controlled editing of videos still lags behind. Prior work in this area has focused on using 2D diffusion models to globally change the style of an existing video. On the other hand, in many practical applications, editing localized parts of the video is critical. In this work, we propose a method to edit videos using a pre-trained inpainting image diffusion model. We systematically redesign the forward path of the model by replacing the self-attention modules with an extended version of attention modules that creates frame-level dependencies. In this way, we ensure that the edited information will be consistent across all the video frames no matter what the shape and position of the masked area is. We qualitatively compare our results with state-of-the-art in terms of accuracy on several video editing tasks like object retargeting, object replacement, and object removal tasks. Simulations demonstrate the superior performance of the proposed strategy.

We define a function on the $C^{\ast}$-algebra of all bounded linear Hilbert space operators, which generalizes the operator radii, and we present some basic properties of this function. Our results extend several results in the literature.

Several numerical radius inequalities in the framework of $C^*$-algebras are proved in this paper. These results, which are based on an extension of Buzano inequality for elements in a pre-Hilbert $C^*$-module, generalize earlier numerical radius inequalities.

Control barrier certificates have proven effective in formally guaranteeing the safety of the control systems. However, designing a control barrier certificate is a time-consuming and computationally expensive endeavor that requires expert input in the form of domain knowledge and mathematical maturity. Additionally, when a system undergoes slight changes, the new controller and its correctness certificate need to be recomputed, incurring similar computational challenges as those faced during the design of the original controller. Prior approaches have utilized transfer learning to transfer safety guarantees in the form of a barrier certificate while maintaining the control invariant. Unfortunately, in practical settings, the source and the target environments often deviate substantially in their control inputs, rendering the aforementioned approach impractical. To address this challenge, we propose integrating \emph{inverse dynamics} -- a neural network that suggests required action given a desired successor state -- of the target system with the barrier certificate of the source system to provide formal proof of safety. In addition, we propose a validity condition that, when met, guarantees correctness of the controller. We demonstrate the effectiveness of our approach through three case studies., Comment: Extended Version, submitted to ADHS 2024

We study a multi-task private semantic communication problem, in which an encoder has access to an information source arbitrarily correlated with some latent private data. A user has $L$ tasks with priorities. The encoder designs a message to be revealed which is called the semantic of the information source. Due to the privacy constraints the semantic can not be disclosed directly and the encoder adds noise to produce disclosed data. The goal is to design the disclosed data that maximizes the weighted sum of the utilities achieved by the user while satisfying a privacy constraint on the private data. In this work, we first consider a single-task scenario and design the added noise utilizing various methods including the extended versions of the Functional Representation Lemma, Strong Functional Representation Lemma, and separation technique. We then study the multi-task scenario and derive a simple design of the source semantics. We show that in the multi-task scenario the main problem can be divided into multiple parallel single-task problems.

We study the concepts of Birkhoff--James orthogonality and parallelism in Hilbert space operators, induced by the operator radius norm $w_{\rho}(\cdot)$. In particular, we completely characterize Birkhoff--James orthogonality and parallelism with respect to $w_{\rho}(\cdot)$. As an application of the results presented, we obtain a well-known characterization due to R.~Bhatia and P.~\v{S}emrl for the classical Birkhoff--James orthogonality of Hilbert space operators. Some other related results are also discussed.

The problem of constructing optimal AIFV codes is a special case of that of constructing minimum cost Markov Chains. This paper provides the first complete proof of correctness for the previously known iterative algorithm for constructing such Markov chains. A recent work describes how to efficiently solve the Markov Chain problem by first constructing a Markov Chain Polytope and then running the Ellipsoid algorithm for linear programming on it. This paper's second result is that, in the AIFV case, a special property of the polytope instead permits solving the corresponding linear program using simple binary search, Comment: Expanded version of paper appearing in ISIT 2024. arXiv admin note: text overlap with arXiv:2401.11622

It is possible to improve upon Tunstall coding using a collection of multiple parse trees. The best such results so far are Iwata and Yamamoto's maximum cost AIVF codes. The most efficient algorithm for designing such codes is an iterative one that could run in exponential time. In this paper, we show that this problem fits into the framework of a newly developed technique that uses linear programming with the Ellipsoid method to solve the minimum cost Markov chain problem. This permits constructing maximum cost AIVF codes in (weakly) polynomial time., Comment: Expanded version of paper appearing on ISIT 2024

The field of conversational information seeking, which is rapidly gaining interest in both academia and industry, is changing how we interact with search engines through natural language interactions. Existing datasets and methods are mostly evaluating reactive conversational information seeking systems that solely provide response to every query from the user. We identify a gap in building and evaluating proactive conversational information seeking systems that can monitor a multi-party human conversation and proactively engage in the conversation at an opportune moment by retrieving useful resources and suggestions. In this paper, we introduce a large-scale dataset for proactive document retrieval that consists of over 2.8 million conversations. We conduct crowdsourcing experiments to obtain high-quality and relatively complete relevance judgments through depth-k pooling. We also collect annotations related to the parts of the conversation that are related to each document, enabling us to evaluate proactive retrieval systems. We introduce normalized proactive discounted cumulative gain (npDCG) for evaluating these systems, and further provide benchmark results for a wide range of models, including a novel model we developed for this task. We believe that the developed dataset, called ProCIS, paves the path towards developing proactive conversational information seeking systems.

We introduce and study a new notion of amenability called symmetric pseudo-amenability. We obtain some properties of symmetrically pseudo-amenable Banach algebras and with examples, we compare this type of amenability with some other types of amenability. We also provide some special classes of symmetrically pseudo-amenable Banach algebras. Finally, Jordan and Lie derivations from a class of Banach algebras into appropriate Banach bimodules are investigated using the notion of symmetric pseudo-amenability.

This paper introduces Stochastic RAG--a novel approach for end-to-end optimization of retrieval-augmented generation (RAG) models that relaxes the simplifying assumptions of marginalization and document independence, made in most prior work. Stochastic RAG casts the retrieval process in RAG as a stochastic sampling without replacement process. Through this formulation, we employ straight-through Gumbel-top-k that provides a differentiable approximation for sampling without replacement and enables effective end-to-end optimization for RAG. We conduct extensive experiments on seven diverse datasets on a wide range of tasks, from open-domain question answering to fact verification to slot-filling for relation extraction and to dialogue systems. By applying this optimization method to a recent and effective RAG model, we advance state-of-the-art results on six out of seven datasets., Comment: To appear in the proceedings of SIGIR 2024

This paper introduces uRAG--a framework with a unified retrieval engine that serves multiple downstream retrieval-augmented generation (RAG) systems. Each RAG system consumes the retrieval results for a unique purpose, such as open-domain question answering, fact verification, entity linking, and relation extraction. We introduce a generic training guideline that standardizes the communication between the search engine and the downstream RAG systems that engage in optimizing the retrieval model. This lays the groundwork for us to build a large-scale experimentation ecosystem consisting of 18 RAG systems that engage in training and 18 unknown RAG systems that use the uRAG as the new users of the search engine. Using this experimentation ecosystem, we answer a number of fundamental research questions that improve our understanding of promises and challenges in developing search engines for machines.

This paper explores SynTOD, a new synthetic data generation approach for developing end-to-end Task-Oriented Dialogue (TOD) Systems capable of handling complex tasks such as intent classification, slot filling, conversational question-answering, and retrieval-augmented response generation, without relying on crowdsourcing or real-world data. SynTOD utilizes a state transition graph to define the desired behavior of a TOD system and generates diverse, structured conversations through random walks and response simulation using large language models (LLMs). In our experiments, using graph-guided response simulations leads to significant improvements in intent classification, slot filling and response relevance compared to naive single-prompt simulated conversations. We also investigate the end-to-end TOD effectiveness of different base and instruction-tuned LLMs, with and without the constructed synthetic conversations. Finally, we explore how various LLMs can evaluate responses in a TOD system and how well they are correlated with human judgments. Our findings pave the path towards quick development and evaluation of domain-specific TOD systems. We release our datasets, models, and code for research purposes.

This paper introduces PAG-a novel optimization and decoding approach that guides autoregressive generation of document identifiers in generative retrieval models through simultaneous decoding. To this aim, PAG constructs a set-based and sequential identifier for each document. Motivated by the bag-of-words assumption in information retrieval, the set-based identifier is built on lexical tokens. The sequential identifier, on the other hand, is obtained via quantizing relevance-based representations of documents. Extensive experiments on MSMARCO and TREC Deep Learning Track data reveal that PAG outperforms the state-of-the-art generative retrieval model by a large margin (e.g., 15.6% MRR improvements on MS MARCO), while achieving 22x speed up in terms of query latency., Comment: Accepted to SIGIR 2024

Evaluating retrieval-augmented generation (RAG) presents challenges, particularly for retrieval models within these systems. Traditional end-to-end evaluation methods are computationally expensive. Furthermore, evaluation of the retrieval model's performance based on query-document relevance labels shows a small correlation with the RAG system's downstream performance. We propose a novel evaluation approach, eRAG, where each document in the retrieval list is individually utilized by the large language model within the RAG system. The output generated for each document is then evaluated based on the downstream task ground truth labels. In this manner, the downstream performance for each document serves as its relevance label. We employ various downstream task metrics to obtain document-level annotations and aggregate them using set-based or ranking metrics. Extensive experiments on a wide range of datasets demonstrate that eRAG achieves a higher correlation with downstream RAG performance compared to baseline methods, with improvements in Kendall's $\tau$ correlation ranging from 0.168 to 0.494. Additionally, eRAG offers significant computational advantages, improving runtime and consuming up to 50 times less GPU memory than end-to-end evaluation.

This paper studies retrieval-augmented approaches for personalizing large language models (LLMs), which potentially have a substantial impact on various applications and domains. We propose the first attempt to optimize the retrieval models that deliver a limited number of personal documents to large language models for the purpose of personalized generation. We develop two optimization algorithms that solicit feedback from the downstream personalized generation tasks for retrieval optimization -- one based on reinforcement learning whose reward function is defined using any arbitrary metric for personalized generation and another based on knowledge distillation from the downstream LLM to the retrieval model. This paper also introduces a pre- and post-generation retriever selection model that decides what retriever to choose for each LLM input. Extensive experiments on diverse tasks from the language model personalization (LaMP) benchmark reveal statistically significant improvements in six out of seven datasets.