Recent LLMs have enabled significant advancements for conversational agents. However, they are also well-known to hallucinate, i.e., they often produce responses that seem plausible but are not factually correct. On the other hand, users tend to over-rely on LLM-based AI agents; they accept the AI's suggestion even when it is wrong. Adding good friction, such as explanations or getting user confirmations, has been proposed as a mitigation in AI-supported decision-making systems. In this paper, we propose an accountability model for LLM-based task-oriented dialogue agents to address user overreliance via friction turns in cases of model uncertainty and errors associated with dialogue state tracking (DST). The accountability model is an augmented LLM with an additional accountability head, which functions as a binary classifier to predict the slots of the dialogue states. We perform our experiments with three backbone LLMs (Llama, Mistral, Gemma) on two established task-oriented datasets (MultiWOZ and Snips). Our empirical findings demonstrate that this approach not only enables reliable estimation of AI agent errors but also guides the LLM decoder in generating more accurate actions. We observe around 3% absolute improvement in joint goal accuracy by incorporating accountability heads in modern LLMs for the MultiWOZ dataset. We also show that this method enables the agent to self-correct its actions, further boosting its performance by 3%. Finally, we discuss the application of accountability modeling to prevent user overreliance by introducing friction.

Autonomous surgical robots are a promising solution to the increasing demand for surgery amid a shortage of surgeons. Recent work has proposed learning-based approaches for the autonomous manipulation of soft tissue. However, due to variability in tissue geometries and stiffnesses, these methods do not always perform optimally, especially in out-of-distribution settings. We propose, develop, and test the first application of uncertainty quantification to learned surgical soft-tissue manipulation policies as an early identification system for task failures. We analyze two different methods of uncertainty quantification, deep ensembles and Monte Carlo dropout, and find that deep ensembles provide a stronger signal of future task success or failure. We validate our approach using the physical daVinci Research Kit (dVRK) surgical robot to perform physical soft-tissue manipulation. We show that we are able to successfully detect task failure and request human intervention when necessary while still enabling autonomous manipulation when possible. Our learned tissue manipulation policy with uncertainty-based early failure detection achieves a zero-shot sim2real performance improvement of 47.5% over the prior state of the art in learned soft-tissue manipulation. We also show that our method generalizes well to new types of tissue as well as to a bimanual soft tissue manipulation task.

Recent advances in Large Language Models (LLMs) have led to significant improvements in natural language processing tasks, but their ability to generate human-quality text raises significant ethical and operational concerns in settings where it is important to recognize whether or not a given text was generated by a human. Thus, recent work has focused on developing techniques for watermarking LLM-generated text, i.e., introducing an almost imperceptible signal that allows a provider equipped with a secret key to determine if given text was generated by their model. Current watermarking techniques are often not practical due to concerns with generation latency, detection time, degradation in text quality, or robustness. Many of these drawbacks come from the focus on token-level watermarking, which ignores the inherent structure of text. In this work, we introduce a new scheme, GaussMark, that is simple and efficient to implement, has formal statistical guarantees on its efficacy, comes at no cost in generation latency, and embeds the watermark into the weights of the model itself, providing a structural watermark. Our approach is based on Gaussian independence testing and is motivated by recent empirical observations that minor additive corruptions to LLM weights can result in models of identical (or even improved) quality. We show that by adding a small amount of Gaussian noise to the weights of a given LLM, we can watermark the model in a way that is statistically detectable by a provider who retains the secret key. We provide formal statistical bounds on the validity and power of our procedure. Through an extensive suite of experiments, we demonstrate that GaussMark is reliable, efficient, and relatively robust to corruptions such as insertions, deletions, substitutions, and roundtrip translations and can be instantiated with essentially no loss in model quality.

Tabular data is one of the most widely used formats across industries, driving critical applications in areas such as finance, healthcare, and marketing. In the era of data-centric AI, improving data quality and representation has become essential for enhancing model performance, particularly in applications centered around tabular data. This survey examines the key aspects of tabular data-centric AI, emphasizing feature selection and feature generation as essential techniques for data space refinement. We provide a systematic review of feature selection methods, which identify and retain the most relevant data attributes, and feature generation approaches, which create new features to simplify the capture of complex data patterns. This survey offers a comprehensive overview of current methodologies through an analysis of recent advancements, practical applications, and the strengths and limitations of these techniques. Finally, we outline open challenges and suggest future perspectives to inspire continued innovation in this field.

One approach to improving the running time of kernel-based methods is to build a small sketch of the kernel matrix and use it in lieu of the full matrix in the machine learning task of interest. Here, we describe a version of this approach that comes with running time guarantees as well as improved guarantees on its statistical performance.By extending the notion of \emph{statistical leverage scores} to the setting of kernel ridge regression, we are able to identify a sampling distribution that reduces the size of the sketch (i.e., the required number of columns to be sampled) to the \emph{effective dimensionality} of the problem. This latter quantity is often much smaller than previous bounds that depend on the \emph{maximal degrees of freedom}. We give an empirical evidence supporting this fact. Our second contribution is to present a fast algorithm to quickly compute coarse approximations to thesescores in time linear in the number of samples.

In the quest to improve services, companies offer customers the option to interact with agents via texting. Such contact centers face unique challenges compared to traditional call centers, as measuring customer experience proxies like abandonment and patience involves uncertainty. A key source of this uncertainty is silent abandonment, where customers leave without notifying the system, wasting agent time and leaving their status unclear. Silent abandonment also obscures whether a customer was served or left. Our goals are to measure the magnitude of silent abandonment and mitigate its effects. Classification models show that 3%-70% of customers across 17 companies abandon silently. In one study, 71.3% of abandoning customers did so silently, reducing agent efficiency by 3.2% and system capacity by 15.3%, incurring $5,457 in annual costs per agent. We develop an expectation-maximization (EM) algorithm to estimate customer patience under uncertainty and identify influencing covariates. We find that companies should use classification models to estimate abandonment scope and our EM algorithm to assess patience. We suggest strategies to operationally mitigate the impact of silent abandonment by predicting suspected silent-abandonment behavior or changing service design. Specifically, we show that while allowing customers to write while waiting in the queue creates a missing data challenge, it also significantly increases patience and reduces service time, leading to reduced abandonment and lower staffing requirements.

This paper describes the approach of the Unibuc - NLP team in tackling the Coling 2025 GenAI Workshop, Task 1: Binary Multilingual Machine-Generated Text Detection. We explored both masked language models and causal models. For Subtask A, our best model achieved first-place out of 36 teams when looking at F1 Micro (Auxiliary Score) of 0.8333, and second-place when looking at F1 Macro (Main Score) of 0.8301

The proliferation of Vision-Language Models (VLMs) in the past several years calls for rigorous and comprehensive evaluation methods and benchmarks. This work analyzes existing VLM evaluation techniques, including automated metrics, AIbased assessments, and human evaluations across diverse tasks. We first introduce Robin - a novel suite of VLMs that we built by combining Large Language Models (LLMs) and Vision Encoders (VEs) at multiple scales, and use Robin to identify shortcomings of current evaluation approaches across scales. Next, to overcome the identified limitations, we introduce CHIRP - a new long form response benchmark we developed for more robust and complete VLM evaluation. We provide open access to the Robin training code, model suite, and CHIRP benchmark to promote reproducibility and advance VLM research. Recently, a lot of significant advances have been made in Vision-Language Models (VLMs), driven by breakthroughs in computer vision and natural language processing Chen et al. (2022); Li et al. (2023b); Liu et al. (2023b); Sun et al. (2023). However, existing VLM benchmarks, often designed for specific tasks (e.g., VQAv2 Goyal et al. (2017)), struggle to accurately reflect real-world VLM performance and capture nuanced differences between models Hsieh et al. (2024). This is particularly evident when evaluating models with significant architectural variations, where standard benchmark scores remain similar despite noticeable differences in human-perceived model quality. To address this issue, we introduce CHIRP, a hybrid VLM benchmark that combines automated metrics' scalability with human evaluators' nuanced judgment. We argue that this approach is crucial for capturing the complexities of VLM behavior, which traditional benchmarks often fail to represent. To demonstrate the limitations of existing benchmarks and the efficacy of our proposed method, we introduce Robin, a suite of VLMs trained at various scales, inspired by the Pythia language model suite Biderman et al. (2023). By systematically varying the Vision Encoder (VE) and the Large Language Model (LLM) sizes, we will show that while benchmark scores remain largely unaffected, human evaluations reveal significant differences in the models' outputs quality. Our findings underscore the need for more robust and human-centric VLM evaluation methodologies. CHIRP paves the way for developing more reliable and informative VLM benchmarks, ultimately leading to the creation of more effective and impactful VLMs. Our Contributions: We investigate the drawbacks of relying on automatic metrics and show the benefits of AI-based and human-based evaluations of VLMs. We train and release an open-source collection of VLMs named Robin. Robin is a scaling suite based on LLMs and VEs of different sizes.

As artificial intelligence (AI) becomes increasingly embedded in healthcare delivery, this chapter explores the critical aspects of developing reliable and ethical Clinical Decision Support Systems (CDSS). Beginning with the fundamental transition from traditional statistical models to sophisticated machine learning approaches, this work examines rigorous validation strategies and performance assessment methods, including the crucial role of model calibration and decision curve analysis. The chapter emphasizes that creating trustworthy AI systems in healthcare requires more than just technical accuracy; it demands careful consideration of fairness, explainability, and privacy. The challenge of ensuring equitable healthcare delivery through AI is stressed, discussing methods to identify and mitigate bias in clinical predictive models. The chapter then delves into explainability as a cornerstone of human-centered CDSS. This focus reflects the understanding that healthcare professionals must not only trust AI recommendations but also comprehend their underlying reasoning. The discussion advances in an analysis of privacy vulnerabilities in medical AI systems, from data leakage in deep learning models to sophisticated attacks against model explanations. The text explores privacy-preservation strategies such as differential privacy and federated learning, while acknowledging the inherent trade-offs between privacy protection and model performance. This progression, from technical validation to ethical considerations, reflects the multifaceted challenges of developing AI systems that can be seamlessly and reliably integrated into daily clinical practice while maintaining the highest standards of patient care and data protection.

Text-to-SQL enables users to interact with databases through natural language, simplifying the retrieval and synthesis of information. Despite the success of large language models (LLMs) in converting natural language questions into SQL queries, their broader adoption is limited by two main challenges: achieving robust generalization across diverse queries and ensuring interpretative confidence in their predictions. To tackle these issues, our research investigates the integration of selective classifiers into Text-to-SQL systems. We analyse the trade-off between coverage and risk using entropy based confidence estimation with selective classifiers and assess its impact on the overall performance of Text-to-SQL models. Additionally, we explore the models' initial calibration and improve it with calibration techniques for better model alignment between confidence and accuracy. Our experimental results show that encoder-decoder T5 is better calibrated than in-context-learning GPT 4 and decoder-only Llama 3, thus the designated external entropy-based selective classifier has better performance. The study also reveal that, in terms of error detection, selective classifier with a higher probability detects errors associated with irrelevant questions rather than incorrect query generations.