Search

Specifying all desirable properties of a language model is challenging, but certain requirements seem essential. Given samples from an unknown language, the trained model should produce valid strings not seen in training and be expressive enough to capture the language's full richness. Otherwise, outputting invalid strings constitutes "hallucination," and failing to capture the full range leads to "mode collapse." We ask if a language model can meet both requirements. We investigate this within a statistical language generation setting building on Gold and Angluin. Here, the model receives random samples from a distribution over an unknown language K, which belongs to a possibly infinite collection of languages. The goal is to generate unseen strings from K. We say the model generates from K with consistency and breadth if, as training size increases, its output converges to all unseen strings in K. Kleinberg and Mullainathan [KM24] asked if consistency and breadth in language generation are possible. We answer this negatively: for a large class of language models, including next-token prediction models, this is impossible for most collections of candidate languages. This contrasts with [KM24]'s result, showing consistent generation without breadth is possible for any countable collection of languages. Our finding highlights that generation with breadth fundamentally differs from generation without breadth. As a byproduct, we establish near-tight bounds on the number of samples needed for generation with or without breadth. Finally, our results offer hope: consistent generation with breadth is achievable for any countable collection of languages when negative examples (strings outside K) are available alongside positive ones. This suggests that post-training feedback, which encodes negative examples, can be crucial in reducing hallucinations while limiting mode collapse., Comment: Abstract shortened to fit arXiv limit

LLMs have demonstrated impressive proficiency in generating coherent and high-quality text, making them valuable across a range of text-generation tasks. However, rigorous evaluation of this generated content is crucial, as ensuring its quality remains a significant challenge due to persistent issues such as factual inaccuracies and hallucinations. This paper introduces two fine-tuned general-purpose LLM autoevaluators, REC-12B and REC-70B, specifically designed to evaluate generated text across several dimensions: faithfulness, instruction following, coherence, and completeness. These models not only provide ratings for these metrics but also offer detailed explanations and verifiable citations, thereby enhancing trust in the content. Moreover, the models support various citation modes, accommodating different requirements for latency and granularity. Extensive evaluations on diverse benchmarks demonstrate that our general-purpose LLM auto-evaluator, REC-70B, outperforms state-of-the-art LLMs, excelling in content evaluation by delivering better quality explanations and citations with minimal bias. It achieves Rank \#1 as a generative model on the RewardBench leaderboard\footnote{\url{https://huggingface.co/spaces/allenai/reward-bench}} under the model name \texttt{TextEval-Llama3.1-70B}. Our REC dataset and models are released at \url{https://github.com/adelaidehsu/REC}.

The rise in popularity of social media platforms, has resulted in millions of new, content pieces being created every day. This surge in content creation underscores the need to pay attention to our design choices as they can greatly impact how long content remains relevant. In today's landscape where regularly recommending new content is crucial, particularly in the absence of detailed information, a variety of factors such as UI features, algorithms and system settings contribute to shaping the journey of content across the platform. While previous research has focused on how new content affects users' experiences, this study takes a different approach by analyzing these decisions considering the content itself. Through a series of carefully crafted experiments we explore how seemingly small decisions can influence the longevity of content, measured by metrics like Content Progression (CVP) and Content Survival (CSR). We also emphasize the importance of recognizing the stages that content goes through underscoring the need to tailor strategies for each stage as a one size fits all approach may not be effective. Additionally we argue for a departure from traditional experimental setups in the study of content lifecycles, to avoid potential misunderstandings while proposing advanced techniques, to achieve greater precision and accuracy in the evaluation process.

Inverse propensity-score weighted (IPW) estimators are prevalent in causal inference for estimating average treatment effects in observational studies. Under unconfoundedness, given accurate propensity scores and $n$ samples, the size of confidence intervals of IPW estimators scales down with $n$, and, several of their variants improve the rate of scaling. However, neither IPW estimators nor their variants are robust to inaccuracies: even if a single covariate has an $\varepsilon>0$ additive error in the propensity score, the size of confidence intervals of these estimators can increase arbitrarily. Moreover, even without errors, the rate with which the confidence intervals of these estimators go to zero with $n$ can be arbitrarily slow in the presence of extreme propensity scores (those close to 0 or 1). We introduce a family of Coarse IPW (CIPW) estimators that captures existing IPW estimators and their variants. Each CIPW estimator is an IPW estimator on a coarsened covariate space, where certain covariates are merged. Under mild assumptions, e.g., Lipschitzness in expected outcomes and sparsity of extreme propensity scores, we give an efficient algorithm to find a robust estimator: given $\varepsilon$-inaccurate propensity scores and $n$ samples, its confidence interval size scales with $\varepsilon+1/\sqrt{n}$. In contrast, under the same assumptions, existing estimators' confidence interval sizes are $\Omega(1)$ irrespective of $\varepsilon$ and $n$. Crucially, our estimator is data-dependent and we show that no data-independent CIPW estimator can be robust to inaccuracies., Comment: Accepted for presentation at the 37th Conference on Learning Theory (COLT) 2024

We study the estimation of distributional parameters when samples are shown only if they fall in some unknown set $S \subseteq \mathbb{R}^d$. Kontonis, Tzamos, and Zampetakis (FOCS'19) gave a $d^{\mathrm{poly}(1/\varepsilon)}$ time algorithm for finding $\varepsilon$-accurate parameters for the special case of Gaussian distributions with diagonal covariance matrix. Recently, Diakonikolas, Kane, Pittas, and Zarifis (COLT'24) showed that this exponential dependence on $1/\varepsilon$ is necessary even when $S$ belongs to some well-behaved classes. These works leave the following open problems which we address in this work: Can we estimate the parameters of any Gaussian or even extend beyond Gaussians? Can we design $\mathrm{poly}(d/\varepsilon)$ time algorithms when $S$ is a simple set such as a halfspace? We make progress on both of these questions by providing the following results: 1. Toward the first question, we give a $d^{\mathrm{poly}(\ell/\varepsilon)}$ time algorithm for any exponential family that satisfies some structural assumptions and any unknown set $S$ that is $\varepsilon$-approximable by degree-$\ell$ polynomials. This result has two important applications: 1a) The first algorithm for estimating arbitrary Gaussian distributions from samples truncated to an unknown $S$; and 1b) The first algorithm for linear regression with unknown truncation and Gaussian features. 2. To address the second question, we provide an algorithm with runtime $\mathrm{poly}(d/\varepsilon)$ that works for a set of exponential families (containing all Gaussians) when $S$ is a halfspace or an axis-aligned rectangle. Along the way, we develop tools that may be of independent interest, including, a reduction from PAC learning with positive and unlabeled samples to PAC learning with positive and negative samples that is robust to certain covariate shifts., Comment: Accepted for presentation at the 65th IEEE Symposium on Foundations of Computer Science (FOCS), 2024; abstract shortened for arXiv

Gravitational lensing is a powerful concept in the Astrophysics to study black holes. The gravitational field of a massive object like a galaxy or black hole bends and magnifies the light from a distant object behind it. The Schwarzchild black hole that are the simplest type of black hole, having no charge or angular momentum have been useful to observe the gravitational lensing. In the presented work, Schwarzchild black hole has been simulated keeping the spiral, elliptical, lenticular and irregular galaxies at the background to obtain the gravitational lensing., Comment: 8 pages, 7 figures

In the case of compute-intensive machine learning, efficient operating system scheduling is crucial for performance and energy efficiency. This paper conducts a comparative study over FIFO(First-In-First-Out) and RR(Round-Robin) scheduling policies with the application of real-time machine learning training processes and data pipelines on Ubuntu-based systems. Knowing a few patterns of CPU usage and energy consumption, we identify which policy (the exclusive or the shared) provides higher performance and/or lower energy consumption for typical modern workloads. Results of this study would help in providing better operating system schedulers for modern systems like Ubuntu, working to improve performance and reducing energy consumption in compute intensive workloads.

To protect consumer privacy, the California Consumer Privacy Act (CCPA) mandates that businesses provide consumers with a straightforward way to opt out of the sale and sharing of their personal information. However, the control that businesses enjoy over the opt-out process allows them to impose hurdles on consumers aiming to opt out, including by employing dark patterns. Motivated by the enactment of the California Privacy Rights Act (CPRA), which strengthens the CCPA and explicitly forbids certain dark patterns in the opt-out process, we investigate how dark patterns are used in opt-out processes and assess their compliance with CCPA regulations. Our research reveals that websites employ a variety of dark patterns. Some of these patterns are explicitly prohibited under the CCPA; others evidently take advantage of legal loopholes. Despite the initial efforts to restrict dark patterns by policymakers, there is more work to be done.

Service robots are increasingly deployed in public spaces, performing functional tasks such as making deliveries. To better integrate them into our social environment and enhance their adoption, we consider integrating social identities within delivery robots along with their functional identity. We conducted a virtual reality-based pilot study to explore people's perceptions and acceptance of delivery robots that perform prosocial behavior. Preliminary findings from thematic analysis of semi-structured interviews illustrate people's ambivalence about dual identity. We discussed the emerging themes in light of social identity theory, framing effect, and human-robot intergroup dynamics. Building on these insights, we propose that the next generation of delivery robots should use peer-based framing, an updated value proposition, and an interactive design that places greater emphasis on expressing intentionality and emotional responses.

Micromobility devices, such as e-scooters and delivery robots, hold promise for eco-friendly and cost-effective alternatives for future urban transportation. However, their lack of societal acceptance remains a challenge. Therefore, we must consider ways to promote prosocial behavior in micromobility interactions. We investigate how post-ride feedback can encourage the prosocial behavior of e-scooter riders while interacting with sidewalk users, including pedestrians and delivery robots. Using a web-based platform, we measure the prosocial behavior of e-scooter riders. Results found that post-ride feedback can successfully promote prosocial behavior, and objective measures indicated better gap behavior, lower speeds at interaction, and longer stopping time around other sidewalk actors. The findings of this study demonstrate the efficacy of post-ride feedback and provide a step toward designing methodologies to improve the prosocial behavior of mobility users., Comment: In 16th International ACM Conference on Automotive User Interfaces and Interactive Vehicular Applications (AutomotiveUI'24), September 22-25, 2024, Stanford, CA, USA. 11 pages

This paper considers a distributionally robust chance constraint model with a general ambiguity set. We show that a sample based approximation of this model converges under suitable sufficient conditions. We also show that upper and lower bounds on the optimal value of the model can be estimated statistically. Specific ambiguity sets are discussed as examples.

In this work, we discuss a recently popular type of recommender system: an LLM-based coding assistant. Connecting the task of providing code recommendations in multiple formats to traditional RecSys challenges, we outline several similarities and differences due to domain specifics. We emphasize the importance of providing relevant context to an LLM for this use case and discuss lessons learned from context enhancements & offline and online evaluation of such AI-assisted coding systems.

With advancements in self-supervised learning, the availability of trillions tokens in a pre-training corpus, instruction fine-tuning, and the development of large Transformers with billions of parameters, large language models (LLMs) are now capable of generating factual and coherent responses to human queries. However, the mixed quality of training data can lead to the generation of undesired responses, presenting a significant challenge. Over the past two years, various methods have been proposed from different perspectives to enhance LLMs, particularly in aligning them with human expectation. Despite these efforts, there has not been a comprehensive survey paper that categorizes and details these approaches. In this work, we aim to address this gap by categorizing these papers into distinct topics and providing detailed explanations of each alignment method, thereby helping readers gain a thorough understanding of the current state of the field.

Graph neural networks (GNNs) are widely applied in graph data modeling. However, existing GNNs are often trained in a task-driven manner that fails to fully capture the intrinsic nature of the graph structure, resulting in sub-optimal node and graph representations. To address this limitation, we propose a novel Graph structure Prompt Learning method (GPL) to enhance the training of GNNs, which is inspired by prompt mechanisms in natural language processing. GPL employs task-independent graph structure losses to encourage GNNs to learn intrinsic graph characteristics while simultaneously solving downstream tasks, producing higher-quality node and graph representations. In extensive experiments on eleven real-world datasets, after being trained by GPL, GNNs significantly outperform their original performance on node classification, graph classification, and edge prediction tasks (up to 10.28%, 16.5%, and 24.15%, respectively). By allowing GNNs to capture the inherent structural prompts of graphs in GPL, they can alleviate the issue of over-smooth and achieve new state-of-the-art performances, which introduces a novel and effective direction for GNN research with potential applications in various domains.

Despite the Graph Neural Networks' (GNNs) proficiency in analyzing graph data, achieving high-accuracy and interpretable predictions remains challenging. Existing GNN interpreters typically provide post-hoc explanations disjointed from GNNs' predictions, resulting in misrepresentations. Self-explainable GNNs offer built-in explanations during the training process. However, they cannot exploit the explanatory outcomes to augment prediction performance, and they fail to provide high-quality explanations of node features and require additional processes to generate explainable subgraphs, which is costly. To address the aforementioned limitations, we propose a self-explained and self-supervised graph neural network (SES) to bridge the gap between explainability and prediction. SES comprises two processes: explainable training and enhanced predictive learning. During explainable training, SES employs a global mask generator co-trained with a graph encoder and directly produces crucial structure and feature masks, reducing time consumption and providing node feature and subgraph explanations. In the enhanced predictive learning phase, mask-based positive-negative pairs are constructed utilizing the explanations to compute a triplet loss and enhance the node representations by contrastive learning., Comment: Accepted as a conference paper at ICDE 2024

This paper presents price prediction models using Machine Learning algorithms augmented with Superforecasters predictions, aimed at enhancing investment decisions. Five Machine Learning models are built, including Bidirectional LSTM, ARIMA, a combination of CNN and LSTM, GRU, and a model built using LSTM and GRU algorithms. The models are evaluated using the Mean Absolute Error to determine their predictive accuracy. Additionally, the paper suggests incorporating human intelligence by identifying Superforecasters and tracking their predictions to anticipate unpredictable shifts or changes in stock prices . The predictions made by these users can further enhance the accuracy of stock price predictions when combined with Machine Learning and Natural Language Processing techniques. Predicting the price of any commodity can be a significant task but predicting the price of a stock in the stock market deals with much more uncertainty. Recognising the limited knowledge and exposure to stocks among certain investors, this paper proposes price prediction models using Machine Learning algorithms. In this work, five Machine learning models are built using Bidirectional LSTM, ARIMA, a combination of CNN and LSTM, GRU and the last one is built using LSTM and GRU algorithms. Later these models are assessed using MAE scores to find which model is predicting with the highest accuracy. In addition to this, this paper also suggests the use of human intelligence to closely predict the shift in price patterns in the stock market The main goal is to identify Superforecasters and track their predictions to anticipate unpredictable shifts or changes in stock prices. By leveraging the combined power of Machine Learning and the Human Intelligence, predictive accuracy can be significantly increased., Comment: Keywords: Machine Learning, Artificial Intelligence, LSTM, GRU, ARMA, CNN, NLP, ANN, SVM, BSE, NIFTY, MAE, MSE, BiLSTM . Published in SSRN Journal

This paper studies privacy in the context of complex decision support queries composed of multiple conditions on different aggregate statistics combined using disjunction and conjunction operators. Utility requirements for such queries necessitate the need for private mechanisms that guarantee a bound on the false negative and false positive errors. This paper formally defines complex decision support queries and their accuracy requirements, and provides algorithms that proportion the existing budget to optimally minimize privacy loss while supporting a bounded guarantee on the accuracy. Our experimental results on multiple real-life datasets show that our algorithms successfully maintain such utility guarantees, while also minimizing privacy loss.

A common method to solve complex problems in software engineering, is to divide the problem into multiple sub-problems. Inspired by this, we propose a Modular Architecture for Software-engineering AI (MASAI) agents, where different LLM-powered sub-agents are instantiated with well-defined objectives and strategies tuned to achieve those objectives. Our modular architecture offers several advantages: (1) employing and tuning different problem-solving strategies across sub-agents, (2) enabling sub-agents to gather information from different sources scattered throughout a repository, and (3) avoiding unnecessarily long trajectories which inflate costs and add extraneous context. MASAI enabled us to achieve the highest performance (28.33% resolution rate) on the popular and highly challenging SWE-bench Lite dataset consisting of 300 GitHub issues from 11 Python repositories. We conduct a comprehensive evaluation of MASAI relative to other agentic methods and analyze the effects of our design decisions and their contribution to the success of MASAI.

Vertical take-off and landing (VTOL) aircraft do not require a prolonged runway, thus allowing them to land almost anywhere. In recent years, their flexibility has made them popular in development, research, and operation. When compared to traditional fixed-wing aircraft and rotorcraft, VTOLs bring unique challenges as they combine many maneuvers from both types of aircraft. Pilot workload is a critical factor for safe and efficient operation of VTOLs. In this work, we conduct a user study to collect multimodal data from 28 pilots while they perform a variety of VTOL flight tasks. We analyze and interpolate behavioral patterns related to their performance and perceived workload. Finally, we build machine learning models to estimate their workload from the collected data. Our results are promising, suggesting that quantitative and accurate VTOL pilot workload monitoring is viable. Such assistive tools would help the research field understand VTOL operations and serve as a stepping stone for the industry to ensure VTOL safe operations and further remote operations., Comment: 8 pages, 7 figures

Goemans and Williamson proposed a randomized rounding algorithm for the MAX-CUT problem with a 0.878 approximation bound in expectation. The 0.878 approximation bound remains the best-known approximation bound for this APX-hard problem. Their approach was subsequently applied to other related problems such as Max-DiCut, MAX-SAT, and Max-2SAT, etc. We show that the randomized rounding algorithm can also be used to achieve a 0.878 approximation bound for the robust and distributionally robust counterparts of the max-cut problem. We also show that the approximation bounds for the other problems are maintained for their robust and distributionally robust counterparts if the randomization projection framework is used.

The limited availability of psychologists necessitates efficient identification of individuals requiring urgent mental healthcare. This study explores the use of Natural Language Processing (NLP) pipelines to analyze text data from online mental health forums used for consultations. By analyzing forum posts, these pipelines can flag users who may require immediate professional attention. A crucial challenge in this domain is data privacy and scarcity. To address this, we propose utilizing readily available curricular texts used in institutes specializing in mental health for pre-training the NLP pipelines. This helps us mimic the training process of a psychologist. Our work presents CASE-BERT that flags potential mental health disorders based on forum text. CASE-BERT demonstrates superior performance compared to existing methods, achieving an f1 score of 0.91 for Depression and 0.88 for Anxiety, two of the most commonly reported mental health disorders. Our code and data are publicly available.

This paper explores the enhancement of creativity in Large Language Models (LLMs) like vGPT-4 through associative thinking, a cognitive process where creative ideas emerge from linking seemingly unrelated concepts. Associative thinking strategies have been found to effectively help humans boost creativity. However, whether the same strategies can help LLMs become more creative remains under-explored. In this work, we investigate whether prompting LLMs to connect disparate concepts can augment their creative outputs. Focusing on three domains -- Product Design, Storytelling, and Marketing -- we introduce creativity tasks designed to assess vGPT-4's ability to generate original and useful content. By challenging the models to form novel associations, we evaluate the potential of associative thinking to enhance the creative capabilities of LLMs. Our findings show that leveraging associative thinking techniques can significantly improve the originality of vGPT-4's responses.

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

In Generalized Category Discovery (GCD), we cluster unlabeled samples of known and novel classes, leveraging a training dataset of known classes. A salient challenge arises due to domain shifts between these datasets. To address this, we present a novel setting: Across Domain Generalized Category Discovery (AD-GCD) and bring forth CDAD-NET (Class Discoverer Across Domains) as a remedy. CDAD-NET is architected to synchronize potential known class samples across both the labeled (source) and unlabeled (target) datasets, while emphasizing the distinct categorization of the target data. To facilitate this, we propose an entropy-driven adversarial learning strategy that accounts for the distance distributions of target samples relative to source-domain class prototypes. Parallelly, the discriminative nature of the shared space is upheld through a fusion of three metric learning objectives. In the source domain, our focus is on refining the proximity between samples and their affiliated class prototypes, while in the target domain, we integrate a neighborhood-centric contrastive learning mechanism, enriched with an adept neighborsmining approach. To further accentuate the nuanced feature interrelation among semantically aligned images, we champion the concept of conditional image inpainting, underscoring the premise that semantically analogous images prove more efficacious to the task than their disjointed counterparts. Experimentally, CDAD-NET eclipses existing literature with a performance increment of 8-15% on three AD-GCD benchmarks we present., Comment: Accepted in L3D-IVU, CVPR Workshop, 2024

Graph-structured data is prevalent in domains such as social networks, financial transactions, brain networks, and protein interactions. As a result, the research community has produced new databases and analytics engines to process such data. Unfortunately, there is not yet widespread benchmark standardization in graph processing, and the heterogeneity of evaluations found in the literature can lead researchers astray. Evaluations frequently ignore datasets' statistical idiosyncrasies, which significantly affect system performance. Scalability studies often use datasets that fit easily in memory on a modest desktop. Some studies rely on synthetic graph generators, but these generators produce graphs with unnatural characteristics that also affect performance, producing misleading results. Currently, the community has no consistent and principled manner with which to compare systems and provide guidance to developers who wish to select the system most suited to their application. We provide three different systematizations of benchmarking practices. First, we present a 12-year literary review of graph processing benchmarking, including a summary of the prevalence of specific datasets and benchmarks used in these papers. Second, we demonstrate the impact of two statistical properties of datasets that drastically affect benchmark performance. We show how different assignments of IDs to vertices, called vertex orderings, dramatically alter benchmark performance due to the caching behavior they induce. We also show the impact of zero-degree vertices on the runtime of benchmarks such as breadth-first search and single-source shortest path. We show that these issues can cause performance to change by as much as 38% on several popular graph processing systems. Finally, we suggest best practices to account for these issues when evaluating graph systems.

We propose leveraging prosocial observations to cultivate new social norms to encourage prosocial behaviors toward delivery robots. With an online experiment, we quantitatively assess updates in norm beliefs regarding human-robot prosocial behaviors through observational learning. Results demonstrate the initially perceived normativity of helping robots is influenced by familiarity with delivery robots and perceptions of robots' social intelligence. Observing human-robot prosocial interactions notably shifts peoples' normative beliefs about prosocial actions; thereby changing their perceived obligations to offer help to delivery robots. Additionally, we found that observing robots offering help to humans, rather than receiving help, more significantly increased participants' feelings of obligation to help robots. Our findings provide insights into prosocial design for future mobility systems. Improved familiarity with robot capabilities and portraying them as desirable social partners can help foster wider acceptance. Furthermore, robots need to be designed to exhibit higher levels of interactivity and reciprocal capabilities for prosocial behavior., Comment: Accepted as a Late Breaking Work at CHI'24

The widespread sharing of consumers personal information with third parties raises significant privacy concerns. The California Consumer Privacy Act (CCPA) mandates that online businesses offer consumers the option to opt out of the sale and sharing of personal information. Our study automatically tracks the presence of the opt-out link longitudinally across multiple states after the California Privacy Rights Act (CPRA) went into effect. We categorize websites based on whether they are subject to CCPA and investigate cases of potential non-compliance. We find a number of websites that implement the opt-out link early and across all examined states but also find a significant number of CCPA-subject websites that fail to offer any opt-out methods even when CCPA is in effect. Our findings can shed light on how websites are reacting to the CCPA and identify potential gaps in compliance and opt-out method designs that hinder consumers from exercising CCPA opt-out rights.

Addressing inequities in the educational outcomes--particularly for students of color and students from low-income backgrounds--cannot happen without comparable data from statewide summative assessments. Statewide assessment results help schools and district leaders target state and local resources to the students and schools with the greatest need and track whether these resources are impacting student achievement. Despite this, many educators, students, and families say that federal assessment and accountability policies take away from instructional time without providing actionable data. Meanwhile, pandemic pauses in administering statewide assessments and changes in political dynamics at the state and federal levels have opened a window of opportunity to develop new statewide summative assessments that gauge how students are doing, highlight disparities, and show where interventions aren't measuring up to their promise and might be improved. This paper centers the lived experiences and perspectives of students, families, educators, and district and state leaders, so that they can be used to design assessments that provide data that will enable the Ed Trust to promote equitable learning opportunities and improve outcomes for all students. To better understand how directly impacted communities are experiencing statewide assessments, Ed Trust held focus groups with diverse stakeholders who are on the ground, focusing on students of color, students from low-income backgrounds, English learners, and those who work in a school or district in which the majority of students are members of these student groups. The focus group findings informed the creation of "equity pillars," which highlight key values and identify criteria for improving federal assessment policy, and federal policy recommendations for how this vision could be achieved.

Generative AI applications, such as ChatGPT or DALL-E, have shown the world their impressive capabilities in generating human-like text or image. Diving deeper, the science stakeholder for those AI applications are Deep Generative Models, a.k.a DGMs, which are designed to learn the underlying distribution of the data and generate new data points that are statistically similar to the original dataset. One critical question is raised: how can we leverage DGMs into morden retail supply chain realm? To address this question, this paper expects to provide a comprehensive review of DGMs and discuss their existing and potential usecases in retail supply chain, by (1) providing a taxonomy and overview of state-of-the-art DGMs and their variants, (2) reviewing existing DGM applications in retail supply chain from a end-to-end view of point, and (3) discussing insights and potential directions on how DGMs can be further utilized on solving retail supply chain problems.

In many predictive contexts (e.g., credit lending), true outcomes are only observed for samples that were positively classified in the past. These past observations, in turn, form training datasets for classifiers that make future predictions. However, such training datasets lack information about the outcomes of samples that were (incorrectly) negatively classified in the past and can lead to erroneous classifiers. We present an approach that trains a classifier using available data and comes with a family of exploration strategies to collect outcome data about subpopulations that otherwise would have been ignored. For any exploration strategy, the approach comes with guarantees that (1) all sub-populations are explored, (2) the fraction of false positives is bounded, and (3) the trained classifier converges to a ``desired'' classifier. The right exploration strategy is context-dependent; it can be chosen to improve learning guarantees and encode context-specific group fairness properties. Evaluation on real-world datasets shows that this approach consistently boosts the quality of collected outcome data and improves the fraction of true positives for all groups, with only a small reduction in predictive utility., Comment: Accepted for presentation at ICML 2024

The ClueWeb22 dataset containing nearly 10 billion documents was released in 2022 to support academic and industry research. The goal of this project was to build retrieval baselines for the English section of the "super head" part (category B) of this dataset. These baselines can then be used by the research community to compare their systems and also to generate data to train/evaluate new retrieval and ranking algorithms. The report covers sparse and dense first stage retrievals as well as neural rerankers that were implemented for this dataset. These systems are available as a service on a Carnegie Mellon University cluster.

Unmanned aerial vehicles, commonly known as drones, have emerged as a disruptive technology with the potential to revolutionize operations across various industries. Drones are the fast-growing internet-of-things technology and are estimated to have a $100 billion market value in the next decade. Exploring drone operations through research has the potential to yield innovative academic insights and create significant practical effects in diverse industries, offering a competitive edge. Drawing insights from both academic and industry literature, this article describes how technological advancements in UAVs may disrupt traditional operational practices in different industries (e.g., commercial last-mile delivery, commercial pickup and delivery, telecommunication, insurance, healthcare, humanitarian, environmental, urban planning, homeland security), identifies the value of this evolving disruptive technology from sustainability and operational innovation perspectives, argues the significance of this area for operations management by conceptualizing a research agenda. The current state of the art focuses on the computing aspect of analytical models to tackle a variety of synthetic drone-related problems, with mixed integer optimization being the primary tool. There is a very significant research gap that should focus on drone operations management with industry know-how by partnering with actual stakeholders and using a variety of tools (i.e., econometrics, field experiments, game theory, optimal control, utility functions). This article aims to promote research on UAVs from operations management and industry-specific point of view., Comment: 45 pages, 10 figures

Text-to-image diffusion models have demonstrated unprecedented capabilities for flexible and realistic image synthesis. Nevertheless, these models rely on a time-consuming sampling procedure, which has motivated attempts to reduce their latency. When improving efficiency, researchers often use the original diffusion model to train an additional network designed specifically for fast image generation. In contrast, our approach seeks to reduce latency directly, without any retraining, fine-tuning, or knowledge distillation. In particular, we find the repeated calculation of attention maps to be costly yet redundant, and instead suggest reusing them during sampling. Our specific reuse strategies are based on ODE theory, which implies that the later a map is reused, the smaller the distortion in the final image. We empirically compare these reuse strategies with few-step sampling procedures of comparable latency, finding that reuse generates images that are closer to those produced by the original high-latency diffusion model.

While Large Language Models (LLMs) display versatile functionality, they continue to generate harmful, biased, and toxic content, as demonstrated by the prevalence of human-designed jailbreaks. In this work, we present Tree of Attacks with Pruning (TAP), an automated method for generating jailbreaks that only requires black-box access to the target LLM. TAP utilizes an attacker LLM to iteratively refine candidate (attack) prompts until one of the refined prompts jailbreaks the target. In addition, before sending prompts to the target, TAP assesses them and prunes the ones unlikely to result in jailbreaks, reducing the number of queries sent to the target LLM. In empirical evaluations, we observe that TAP generates prompts that jailbreak state-of-the-art LLMs (including GPT4-Turbo and GPT4o) for more than 80% of the prompts. This significantly improves upon the previous state-of-the-art black-box methods for generating jailbreaks while using a smaller number of queries than them. Furthermore, TAP is also capable of jailbreaking LLMs protected by state-of-the-art guardrails, e.g., LlamaGuard., Comment: Accepted for presentation at NeurIPS 2024. Code: https://github.com/RICommunity/TAP

Practitioners who wish to build real-world applications that rely on ranking models, need to decide which modelling paradigm to follow. This is not an easy choice to make, as the research literature on this topic has been shifting in recent years. In particular, whilst Gradient Boosted Decision Trees (GBDTs) have reigned supreme for more than a decade, the flexibility of neural networks has allowed them to catch up, and recent works report accuracy metrics that are on par. Nevertheless, practical systems require considerations beyond mere accuracy metrics to decide on a modelling approach. This work describes our experiences in balancing some of the trade-offs that arise, presenting a case study on a short-video recommendation application. We highlight (1) neural networks' ability to handle large training data size, user- and item-embeddings allows for more accurate models than GBDTs in this setting, and (2) because GBDTs are less reliant on specialised hardware, they can provide an equally accurate model at a lower cost. We believe these findings are of relevance to researchers in both academia and industry, and hope they can inspire practitioners who need to make similar modelling choices in the future., Comment: Appearing in the Industry Track Proceedings of the Forum for Information Retrieval Evaluation (FIRE '23)