Translating Natural Language Queries into Structured Query Language (Text-to-SQL or NLQ-to-SQL) is a critical task extensively studied by both the natural language processing and database communities, aimed at providing a natural language interface to databases (NLIDB) and lowering the barrier for non-experts. Despite recent advancements made through the use of Large Language Models (LLMs), significant challenges remain. These include handling complex database schemas, resolving ambiguity in user queries, and generating SQL queries with intricate structures that accurately reflect the user's intent. In this work, we introduce E-SQL, a novel pipeline specifically designed to address these challenges through direct schema linking and candidate predicate augmentation. E-SQL enhances the natural language query by incorporating relevant database items (i.e., tables, columns, and values) and conditions directly into the question, bridging the gap between the query and the database structure. The pipeline leverages candidate predicate augmentation to mitigate erroneous or incomplete predicates in generated SQLs. We further investigate the impact of schema filtering, a technique widely explored in previous work, and demonstrate its diminishing returns when applied alongside advanced large language models. Comprehensive evaluations on the BIRD benchmark illustrate that E-SQL achieves competitive performance, particularly excelling in complex queries with a 66.29% execution accuracy on the test set. All code required to reproduce the reported results is publicly available on our GitHub repository.

We test the robustness of debate as a method of scalable oversight by training models to debate with data generated via self-play. In a long-context reading comprehension task, we find that language model based evaluators answer questions more accurately when judging models optimized to win debates. By contrast, we find no such relationship for consultancy models trained to persuade a judge without an opposing debater present. In quantitative and qualitative comparisons between our debate models and novel consultancy baselines, we find evidence that debate training encourages stronger and more informative arguments, showing promise that it can help provide high-quality supervision for tasks that are difficult to directly evaluate.

Recent studies have evaluated the creativity/novelty of large language models (LLMs) primarily from a semantic perspective, using benchmarks from cognitive science. However, accessing the novelty in scholarly publications is a largely unexplored area in evaluating LLMs. In this paper, we introduce a scholarly novelty benchmark (SchNovel) to evaluate LLMs' ability to assess novelty in scholarly papers. SchNovel consists of 15000 pairs of papers across six fields sampled from the arXiv dataset with publication dates spanning 2 to 10 years apart. In each pair, the more recently published paper is assumed to be more novel. Additionally, we propose RAG-Novelty, which simulates the review process taken by human reviewers by leveraging the retrieval of similar papers to assess novelty. Extensive experiments provide insights into the capabilities of different LLMs to assess novelty and demonstrate that RAG-Novelty outperforms recent baseline models.

Recent developments in natural language generation have tremendous implications for healthcare. For instance, state-of-the-art systems could automate the generation of sections in clinical reports to alleviate physician workload and streamline hospital documentation. To explore these applications, we present a shared task consisting of two subtasks: (1) Radiology Report Generation (RRG24) and (2) Discharge Summary Generation ("Discharge Me!"). RRG24 involves generating the 'Findings' and 'Impression' sections of radiology reports given chest X-rays. "Discharge Me!" involves generating the 'Brief Hospital Course' and 'Discharge Instructions' sections of discharge summaries for patients admitted through the emergency department. "Discharge Me!" submissions were subsequently reviewed by a team of clinicians. Both tasks emphasize the goal of reducing clinician burnout and repetitive workloads by generating documentation. We received 201 submissions from across 8 teams for RRG24, and 211 submissions from across 16 teams for "Discharge Me!".

Recent work has shown that Transformers trained from scratch can successfully solve various arithmetic and algorithmic tasks, such as adding numbers and computing parity. While these Transformers generalize well on unseen inputs of the same length, they struggle with length generalization, i.e., handling inputs of unseen lengths. In this work, we demonstrate that looped Transformers with an adaptive number of steps significantly improve length generalization. We focus on tasks with a known iterative solution, involving multiple iterations of a RASP-L operation - a length-generalizable operation that can be expressed by a finite-sized Transformer. We train looped Transformers using our proposed learning algorithm and observe that they learn highly length-generalizable solutions for various tasks.

LLM-based applications are helping people write, and LLM-generated text is making its way into social media, journalism, and our classrooms. However, the differences between LLM-generated and human-written text remain unclear. To explore this, we hired professional writers to edit paragraphs in several creative domains. We first found these writers agree on undesirable idiosyncrasies in LLM-generated text, formalizing it into a seven-category taxonomy (e.g. cliches, unnecessary exposition). Second, we curated the LAMP corpus: 1,057 LLM-generated paragraphs edited by professional writers according to our taxonomy. Analysis of LAMP reveals that none of the LLMs used in our study (GPT4o, Claude-3.5-Sonnet, Llama-3.1-70b) outperform each other in terms of writing quality, revealing common limitations across model families. Third, we explored automatic editing methods to improve LLM-generated text. A large-scale preference annotation confirms that although experts largely prefer text edited by other experts, automatic editing methods show promise in improving alignment between LLM-generated and human-written text.

Assessment of proficiency of the learner is an essential part of Intelligent Tutoring Systems (ITS). We use Item Response Theory (IRT) in computer-aided language learning for assessment of student ability in two contexts: in test sessions, and in exercises during practice sessions. Exhaustive testing across a wide range of skills can provide a detailed picture of proficiency, but may be undesirable for a number of reasons. Therefore, we first aim to replace exhaustive tests with efficient but accurate adaptive tests. We use learner data collected from exhaustive tests under imperfect conditions, to train an IRT model to guide adaptive tests. Simulations and experiments with real learner data confirm that this approach is efficient and accurate. Second, we explore whether we can accurately estimate learner ability directly from the context of practice with exercises, without testing. We transform learner data collected from exercise sessions into a form that can be used for IRT modeling. This is done by linking the exercises to {\em linguistic constructs}; the constructs are then treated as "items" within IRT. We present results from large-scale studies with thousands of learners. Using teacher assessments of student ability as "ground truth," we compare the estimates obtained from tests vs. those from exercises. The experiments confirm that the IRT models can produce accurate ability estimation based on exercises.

Automatic Question Answering (QA) systems rely on contextual information to provide accurate answers. Commonly, contexts are prepared through either retrieval-based or generation-based methods. The former involves retrieving relevant documents from a corpus like Wikipedia, whereas the latter uses generative models such as Large Language Models (LLMs) to generate the context. In this paper, we introduce a novel context preparation approach called HINTQA, which employs Automatic Hint Generation (HG) techniques. Unlike traditional methods, HINTQA prompts LLMs to produce hints about potential answers for the question rather than generating relevant context. We evaluate our approach across three QA datasets including TriviaQA, NaturalQuestions, and Web Questions, examining how the number and order of hints impact performance. Our findings show that the HINTQA surpasses both retrieval-based and generation-based approaches. We demonstrate that hints enhance the accuracy of answers more than retrieved and generated contexts.

Classical neural ordinary differential equations (ODEs) are powerful tools for approximating the log-density functions in high-dimensional spaces along trajectories, where neural networks parameterize the velocity fields. This paper proposes a system of neural differential equations representing first-and second-order score functions along trajectories based on deep neural networks. We reformulate the mean field control (MFC) problem with individual noises into an unconstrained optimization problem framed by the proposed neural ODE system. Additionally, we introduce a novel regularization term to enforce characteristics of viscous Hamilton-Jacobi-Bellman (HJB) equations to be satisfied based on the evolution of the second-order score function. Examples include regularized Wasserstein proximal operators (RWPOs), probability flow matching of Fokker-Planck (FP) equations, and linear quadratic (LQ) MFC problems, which demonstrate the effectiveness and accuracy of the proposed method. Score functions have been widely used in modern machine learning algorithms, particularly generative models through time-reversible diffusion (Song et al., 2021). The score function can be viewed as a deterministic representation of diffusion in stochastic trajectories (Carrillo et al., 2019). These properties have inspired algorithms for simulating stochastic trajectories or sampling problems that converge to target distributions (Wang et al., 2022; Lu et al., 2024). Typical applications include modeling the time evolution of probability densities for stochastic dynamics and solving control problems constrained by such dynamics. While score functions provide powerful tools for modeling stochastic trajectories, their computations are often inefficient, especially in high-dimensional spaces. Classical methods, such as kernel density estimation (KDE) (Chen, 2017), tend to perform poorly in such settings due to the curse of dimensionality (Terrell & Scott, 1992). Recently, neural ODEs have emerged as efficient ways of estimating densities. In particular, one uses neural networks to parameterize the velocity fields and then approximates the logarithm of density function along trajectories. The time discretizations of neural ODEs can be viewed as normalization flows in generative models.

We focus on a relatively unexplored learning paradigm known as {\em Online Unsupervised Continual Learning} (O-UCL), where an agent receives a non-stationary, unlabeled data stream and progressively learns to identify an increasing number of classes. This paradigm is designed to model real-world applications where encountering novelty is the norm, such as exploring a terrain with several unknown and time-varying entities. Unlike prior work in unsupervised, continual, or online learning, O-UCL combines all three areas into a single challenging and realistic learning paradigm. In this setting, agents are frequently evaluated and must aim to maintain the best possible representation at any point of the data stream, rather than at the end of pre-specified offline tasks. The proposed approach, called \textbf{P}atch-based \textbf{C}ontrastive learning and \textbf{M}emory \textbf{C}onsolidation (PCMC), builds a compositional understanding of data by identifying and clustering patch-level features. Embeddings for these patch-level features are extracted with an encoder trained via patch-based contrastive learning. PCMC incorporates new data into its distribution while avoiding catastrophic forgetting, and it consolidates memory examples during ``sleep" periods. We evaluate PCMC's performance on streams created from the ImageNet and Places365 datasets. Additionally, we explore various versions of the PCMC algorithm and compare its performance against several existing methods and simple baselines.