Despite the success of large language models (LLMs) in Text-to-SQL tasks, open-source LLMs encounter challenges in contextual understanding and response coherence. To tackle these issues, we present \ours, a systematic methodology tailored for Text-to-SQL with open-source LLMs. Our contributions include a comprehensive evaluation of open-source LLMs in Text-to-SQL tasks, the \openprompt strategy for effective question representation, and novel strategies for supervised fine-tuning. We explore the benefits of Chain-of-Thought in step-by-step inference and propose the \openexample method for enhanced few-shot learning. Additionally, we introduce token-efficient techniques, such as \textbf{Variable-length Open DB Schema}, \textbf{Target Column Truncation}, and \textbf{Example Column Truncation}, addressing challenges in large-scale databases. Our findings emphasize the need for further investigation into the impact of supervised fine-tuning on contextual learning capabilities. Remarkably, our method significantly improved Llama2-7B from 2.54\% to 41.04\% and Code Llama-7B from 14.54\% to 48.24\% on the BIRD-Dev dataset. Notably, the performance of Code Llama-7B surpassed GPT-4 (46.35\%) on the BIRD-Dev dataset.

Humans can develop new theorems to explore broader and more complex mathematical results. While current generative language models (LMs) have achieved significant improvement in automatically proving theorems, their ability to generate new or reusable theorems is still under-explored. Without the new theorems, current LMs struggle to prove harder theorems that are distant from the given hypotheses with the exponentially growing search space. Therefore, this paper proposes an Automated Theorem Generation (ATG) benchmark that evaluates whether an agent can automatically generate valuable (and possibly brand new) theorems that are applicable for downstream theorem proving as reusable knowledge. Specifically, we construct the ATG benchmark by Figure 1: An example theorem generated by GPT-4 splitting the Metamath library into three sets: (OpenAI, 2023). GPT-4 wrongly refers to the intermediate axioms, library, and problem based on their theorem (A (B A) A A) as proving depth. We conduct extensive experiments ((A B) (A A)). In Step 4, it applies "ax-to investigate whether current LMs can 1" but obtains the wrong expression instead of correct generate theorems in the library and benefit (A (B A)) and can not derive (A A) even the problem theorems proving. The results with the incorrect Steps 4 and 5. demonstrate that high-quality ATG data facilitates models' performances on downstream

This work focuses on leveraging and selecting from vast, unlabeled, open data to pre-fine-tune a pre-trained language model. The goal is to minimize the need for costly domain-specific data for subsequent fine-tuning while achieving desired performance levels. While many data selection algorithms have been designed for small-scale applications, rendering them unsuitable for our context, some emerging methods do cater to language data scales. However, they often prioritize data that aligns with the target distribution. While this strategy may be effective when training a model from scratch, it can yield limited results when the model has already been pre-trained on a different distribution. Differing from prior work, our key idea is to select data that nudges the pre-training distribution closer to the target distribution. We show the optimality of this approach for fine-tuning tasks under certain conditions. We demonstrate the efficacy of our methodology across a diverse array of tasks (NLU, NLG, zero-shot) with models up to 2.7B, showing that it consistently surpasses other selection methods. Moreover, our proposed method is significantly faster than existing techniques, scaling to millions of samples within a single GPU hour. Our code is open-sourced (Code repository: https://anonymous.4open.science/r/DV4LLM-D761/ ). While fine-tuning offers significant potential for enhancing performance across diverse tasks, its associated costs often limit its widespread adoption; with this work, we hope to lay the groundwork for cost-effective fine-tuning, making its benefits more accessible.

Despite the longstanding adage "an image is worth a thousand words," creating accurate and hyper-detailed image descriptions for training Vision-Language models remains challenging. Current datasets typically have web-scraped descriptions that are short, low-granularity, and often contain details unrelated to the visual content. As a result, models trained on such data generate descriptions replete with missing information, visual inconsistencies, and hallucinations. To address these issues, we introduce ImageInWords (IIW), a carefully designed human-in-the-loop annotation framework for curating hyper-detailed image descriptions and a new dataset resulting from this process. We validate the framework through evaluations focused on the quality of the dataset and its utility for fine-tuning with considerations for readability, comprehensiveness, specificity, hallucinations, and human-likeness. Our dataset significantly improves across these dimensions compared to recently released datasets (+66%) and GPT-4V outputs (+48%). Furthermore, models fine-tuned with IIW data excel by +31% against prior work along the same human evaluation dimensions. Given our fine-tuned models, we also evaluate text-to-image generation and vision-language reasoning. Our model's descriptions can generate images closest to the original, as judged by both automated and human metrics. We also find our model produces more compositionally rich descriptions, outperforming the best baseline by up to 6% on ARO, SVO-Probes, and Winoground datasets.

The task of Information Retrieval (IR) requires a system to identify relevant documents based on users' information needs. In real-world scenarios, retrievers are expected to not only rely on the semantic relevance between the documents and the queries but also recognize the nuanced intents or perspectives behind a user query. For example, when asked to verify a claim, a retrieval system is expected to identify evidence from both supporting vs. contradicting perspectives, for the downstream system to make a fair judgment call. In this work, we study whether retrievers can recognize and respond to different perspectives of the queries -- beyond finding relevant documents for a claim, can retrievers distinguish supporting vs. opposing documents? We reform and extend six existing tasks to create a benchmark for retrieval, where we have diverse perspectives described in free-form text, besides root, neutral queries. We show that current retrievers covered in our experiments have limited awareness of subtly different perspectives in queries and can also be biased toward certain perspectives. Motivated by the observation, we further explore the potential to leverage geometric features of retriever representation space to improve the perspective awareness of retrievers in a zero-shot manner. We demonstrate the efficiency and effectiveness of our projection-based methods on the same set of tasks. Further analysis also shows how perspective awareness improves performance on various downstream tasks, with 4.2% higher accuracy on AmbigQA and 29.9% more correlation with designated viewpoints on essay writing, compared to non-perspective-aware baselines.

As generative artificial intelligence (AI), particularly Large Language Models (LLMs), continues to permeate healthcare, it remains crucial to supplement traditional automated evaluations with human expert evaluation. Understanding and evaluating the generated texts is vital for ensuring safety, reliability, and effectiveness. However, the cumbersome, time-consuming, and non-standardized nature of human evaluation presents significant obstacles to the widespread adoption of LLMs in practice. This study reviews existing literature on human evaluation methodologies for LLMs within healthcare. We highlight a notable need for a standardized and consistent human evaluation approach. Our extensive literature search, adhering to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, spans publications from January 2018 to February 2024. This review provides a comprehensive overview of the human evaluation approaches used in diverse healthcare applications.This analysis examines the human evaluation of LLMs across various medical specialties, addressing factors such as evaluation dimensions, sample types, and sizes, the selection and recruitment of evaluators, frameworks and metrics, the evaluation process, and statistical analysis of the results. Drawing from diverse evaluation strategies highlighted in these studies, we propose a comprehensive and practical framework for human evaluation of generative LLMs, named QUEST: Quality of Information, Understanding and Reasoning, Expression Style and Persona, Safety and Harm, and Trust and Confidence. This framework aims to improve the reliability, generalizability, and applicability of human evaluation of generative LLMs in different healthcare applications by defining clear evaluation dimensions and offering detailed guidelines.

Methods for relation extraction from text mostly focus on high precision, at the cost of limited recall. High recall is crucial, though, to populate long lists of object entities that stand in a specific relation with a given subject. Cues for relevant objects can be spread across many passages in long texts. This poses the challenge of extracting long lists from long texts. We present the L3X method which tackles the problem in two stages: (1) recall-oriented generation using a large language model (LLM) with judicious techniques for retrieval augmentation, and (2) precision-oriented scrutinization to validate or prune candidates. Our L3X method outperforms Figure 1: Example for extracting long lists from long LLM-only generations by a substantial texts. For the subject "Harry Potter", we aim to extract margin.

In today's technologically driven world, the rapid spread of fake news, particularly during critical events like elections, poses a growing threat to the integrity of information. To tackle this challenge head-on, we introduce FakeWatch, a comprehensive framework carefully designed to detect fake news. Leveraging a newly curated dataset of North American election-related news articles, we construct robust classification models. Our framework integrates a model hub comprising of both traditional machine learning (ML) techniques, and state-of-the-art Language Models (LMs) to discern fake news effectively. Our objective is to provide the research community with adaptable and precise classification models adept at identifying fake news for the elections agenda. Quantitative evaluations of fake news classifiers on our dataset reveal that, while state-of-the-art LMs exhibit a slight edge over traditional ML models, classical models remain competitive due to their balance of accuracy and computational efficiency. Additionally, qualitative analyses shed light on patterns within fake news articles. We provide our labeled data at https://huggingface.co/datasets/newsmediabias/fake_news_elections_labelled_data and model https://huggingface.co/newsmediabias/FakeWatch for reproducibility and further research.

Large language models (LLMs) have shown remarkable adaptability to diverse tasks, by leveraging context prompts containing instructions, or minimal input-output examples. However, recent work revealed they also exhibit label bias -- an undesirable preference toward predicting certain answers over others. Still, detecting and measuring this bias reliably and at scale has remained relatively unexplored. In this study, we evaluate different approaches to quantifying label bias in a model's predictions, conducting a comprehensive investigation across 279 classification tasks and ten LLMs. Our investigation reveals substantial label bias in models both before and after debiasing attempts, as well as highlights the importance of outcomes-based evaluation metrics, which were not previously used in this regard. We further propose a novel label bias calibration method tailored for few-shot prompting, which outperforms recent calibration approaches for both improving performance and mitigating label bias. Our results emphasize that label bias in the predictions of LLMs remains a barrier to their reliability.

Text Style Transfer (TST) seeks to alter the style of text while retaining its core content. Given the constraints of limited parallel datasets for TST, we propose CoTeX, a framework that leverages large language models (LLMs) alongside chain-of-thought (CoT) prompting to facilitate TST. CoTeX distills the complex rewriting and reasoning capabilities of LLMs into more streamlined models capable of working with both non-parallel and parallel data. Through experimentation across four TST datasets, CoTeX is shown to surpass traditional supervised fine-tuning and knowledge distillation methods, particularly in low-resource settings. We conduct a comprehensive evaluation, comparing CoTeX against current unsupervised, supervised, in-context learning (ICL) techniques, and instruction-tuned LLMs. Furthermore, CoTeX distinguishes itself by offering transparent explanations for its style transfer process.