Pre-trained large foundation models play a central role in the recent surge of artificial intelligence, resulting in fine-tuned models with remarkable abilities when measured on benchmark datasets, standard exams, and applications. Due to their inherent complexity, these models are not well understood. While small adversarial inputs to such models are well known, the structures of the representation space are not well characterized despite their fundamental importance. In this paper, using the vision transformers as an example due to the continuous nature of their input space, we show via analyses and systematic experiments that the representation space consists of large piecewise linear subspaces where there exist very different inputs sharing the same representations, and at the same time, local normal spaces where there are visually indistinguishable inputs having very different representations. The empirical results are further verified using the local directional estimations of the Lipschitz constants of the underlying models. Consequently, the resulting representations change the results of downstream models, and such models are subject to overgeneralization and with limited semantically meaningful generalization capability.

Table question answering is a popular task that assesses a model's ability to understand and interact with structured data. However, the given table often does not contain sufficient information for answering the question, necessitating the integration of external knowledge. Existing methods either convert both the table and external knowledge into text, which neglects the structured nature of the table; or they embed queries for external sources in the interaction with the table, which complicates the process. In this paper, we propose a simple yet effective method to integrate external information in a given table. Our method first constructs an augmenting table containing the missing information and then generates a SQL query over the two tables to answer the question. Experiments show that our method outperforms strong baselines on three table QA benchmarks. Our code is publicly available at https://github.com/UCSB-NLP-Chang/Augment_tableQA.

In recent times, a plethora of Large Code Generation Models (LCGMs) have been proposed, showcasing significant potential in assisting developers with complex programming tasks. Benchmarking LCGMs necessitates the creation of a set of diverse programming problems, and each problem comprises the prompt (including the task description), canonical solution, and test inputs. The existing methods for constructing such a problem set can be categorized into two main types: manual methods and perturbation-based methods. However, manual methods demand high effort and lack scalability, while also risking data integrity due to LCGMs' potentially contaminated data collection, and perturbation-based approaches mainly generate semantically homogeneous problems with the same canonical solutions and introduce typos that can be easily auto-corrected by IDE, making them ineffective and unrealistic. In this work, we propose the idea of programming problem merging (PPM) and provide two implementation of this idea, we utilize our tool on two widely-used datasets and compare it against nine baseline methods using eight code generation models. The results demonstrate the effectiveness of our tool in generating more challenging, diverse, and natural programming problems, comparing to the baselines.

In the domain of Natural Language Processing (NLP), many works have investigated social biases in terms of associations in the embeddings space. Early works [1, 2] introduced methods to measure and mitigate social biases based on cosine similarity in word embeddigs. With NLP research progressing to large language models and contextualized embeddings, doubts have been raised whether these methods are still suitable for fairness evaluation [3] and other works criticize that for instance the Word Embedding Association Test (WEAT) [2] fails to detect some kinds of biases [4, 5]. Overall there exists a great deal of bias measures in the literature, which not necessarily detect the same biases [6, 4, 5]. In general, researchers are questioning the usability of model intrinsic bias measures, such as cosine based methods [7, 8, 9]. There exist few papers that compare the performance of different bias scores [10, 11] and works that evaluate experimental setups for bias measurement [12]. However, to our knowledge, only two works investigate the properties of intrinsic bias scores on a theoretical level [5, 13]. To further close this gap, we evaluate the semantic properties of cosine based bias scores, focusing on bias quantification as opposed to bias detection. We make the following contributions: (i) We formalize the properties of trustworthiness and comparability as requirements for cosine based bias scores.

This paper investigates the potential benefits of language-specific fact-checking models, focusing on the case of Chinese. We demonstrate the limitations of methods such as translating Chinese claims and evidence into English or directly using multilingual large language models (e.g. GPT4), highlighting the need for language-specific systems. We further develop a state-of-the-art Chinese fact-checking system that, in contrast to previous approaches which treat evidence selection as a pairwise sentence classification task, considers the context of sentences. We also create an adversarial dataset to identify biases in our model, and while they are present as in English language datasets and models, they are often specific to the Chinese culture. Our study emphasizes the importance of language-specific fact-checking models to effectively combat misinformation.

While large language models (LLMs) are extremely capable at text generation, their outputs are still distinguishable from human-authored text. We explore this separation across many metrics over text, many sampling techniques, many types of text data, and across two popular LLMs, LLaMA and Vicuna. Along the way, we introduce a new metric, recoverability, to highlight differences between human and machine text; and we propose a new sampling technique, burst sampling, designed to close this gap. We find that LLaMA and Vicuna have distinct distributions under many of the metrics, and that this influences our results: Recoverability separates real from fake text better than any other metric when using LLaMA. When using Vicuna, burst sampling produces text which is distributionally closer to real text compared to other sampling techniques.

Mastering commonsense understanding and reasoning is a pivotal skill essential for conducting engaging conversations. While there have been several attempts to create datasets that facilitate commonsense inferences in dialogue contexts, existing datasets tend to lack in-depth details, restate information already present in the conversation, and often fail to capture the multifaceted nature of commonsense reasoning. In response to these limitations, we compile a new synthetic dataset for commonsense reasoning in dialogue contexts using GPT, ConvoSense, that boasts greater contextual novelty, offers a higher volume of inferences per example, and substantially enriches the detail conveyed by the inferences. Our dataset contains over 500,000 inferences across 12,000 dialogues with 10 popular inference types, which empowers the training of generative commonsense models for dialogue that are superior in producing plausible inferences with high novelty when compared to models trained on the previous datasets. To the best of our knowledge, ConvoSense is the first of its kind to provide such a multitude of novel inferences at such a large scale.

This paper introduces DataFrame question answering (QA), a novel task that utilizes large language models (LLMs) to generate Pandas queries for information retrieval and data analysis on dataframes, emphasizing safe and non-revealing data handling. Our method, which solely relies on dataframe column names, not only ensures data privacy but also significantly reduces the context window in the prompt, streamlining information processing and addressing major challenges in LLM-based data analysis. We propose DataFrame QA as a comprehensive framework that includes safe Pandas query generation and code execution. Various LLMs, notably GPT-4, are evaluated using the pass@1 metric on the renowned WikiSQL and our newly developed 'UCI-DataFrameQA', tailored for complex data analysis queries. Our findings indicate that GPT-4 achieves pass@1 rates of 86% on WikiSQL and 97% on UCI-DataFrameQA, underscoring its capability in securely retrieving and aggregating dataframe values and conducting sophisticated data analyses. This approach, deployable in a zero-shot manner without prior training or adjustments, proves to be highly adaptable and secure for diverse applications.

We evaluate the ability of Large Language Models (LLMs) to discern and express their internal knowledge state, a key factor in countering factual hallucination and ensuring reliable application of LLMs. We observe a robust self-awareness of internal knowledge state in LLMs, evidenced by over 85% accuracy in knowledge probing. However, LLMs often fail to express their internal knowledge during generation, leading to factual hallucinations. We develop an automated hallucination annotation tool, Dreamcatcher, which merges knowledge probing and consistency checking methods to rank factual preference data. Using knowledge preference as reward, We propose a Reinforcement Learning from Knowledge Feedback (RLKF) training framework, leveraging reinforcement learning to enhance the factuality and honesty of LLMs. Our experiments across multiple models show that RLKF training effectively enhances the ability of models to utilize their internal knowledge state, boosting performance in a variety of knowledge-based and honesty-related tasks.

Multiword expressions (MWEs) are composed of multiple words and exhibit variable degrees of compositionality. As such, their meanings are notoriously difficult to model, and it is unclear to what extent this issue affects transformer architectures. Addressing this gap, we provide the first in-depth survey of MWE processing with transformer models. We overall find that they capture MWE semantics inconsistently, as shown by reliance on surface patterns and memorized information. MWE meaning is also strongly localized, predominantly in early layers of the architecture. Representations benefit from specific linguistic properties, such as lower semantic idiosyncrasy and ambiguity of target expressions. Our findings overall question the ability of transformer models to robustly capture fine-grained semantics. Furthermore, we highlight the need for more directly comparable evaluation setups.