We introduce a novel writing method called Probing Chain of Thought (ProCoT), which prevents students from cheating using a Large Language Model (LLM), such as ChatGPT, while enhancing their active learning through such models. LLMs have disrupted education and many other feilds. For fear of students cheating, many educationists have resorted to banning their use, as their outputs can be human-like and hard to detect in some cases. These LLMs are also known for hallucinations (i.e. fake facts). We conduct studies with ProCoT in two different courses with a combined total of about 66 students. The students in each course were asked to prompt an LLM of their choice with one question from a set of four and required to affirm or refute statements in the LLM output by using peer reviewed references. The results show two things: (1) ProCoT stimulates creative/critical thinking and writing of students through engagement with LLMs when we compare the LLM solely output to ProCoT output and (2) ProCoT can prevent cheating because of clear limitations in existing LLMs when we compare students ProCoT output to LLM ProCoT output. We also discover that most students prefer to give answers in fewer words than LLMs, which are typically verbose. The average word counts for students, ChatGPT (v3.5) and Phind (v8) are 208, 391 and 383, respectively.

Invariant representation learning (IRL) encourages the prediction from invariant causal features to labels de-confounded from the environments, advancing the technical roadmap of out-of-distribution (OOD) generalization. Despite spotlights around, recent theoretical results verified that some causal features recovered by IRLs merely pretend domain-invariantly in the training environments but fail in unseen domains. The \emph{fake invariance} severely endangers OOD generalization since the trustful objective can not be diagnosed and existing causal surgeries are invalid to rectify. In this paper, we review a IRL family (InvRat) under the Partially and Fully Informative Invariant Feature Structural Causal Models (PIIF SCM /FIIF SCM) respectively, to certify their weaknesses in representing fake invariant features, then, unify their causal diagrams to propose ReStructured SCM (RS-SCM). RS-SCM can ideally rebuild the spurious and the fake invariant features simultaneously. Given this, we further develop an approach based on conditional mutual information with respect to RS-SCM, then rigorously rectify the spurious and fake invariant effects. It can be easily implemented by a small feature selection subnet introduced in the IRL family, which is alternatively optimized to achieve our goal. Experiments verified the superiority of our approach to fight against the fake invariant issue across a variety of OOD generalization benchmarks.

Since Pretrained Language Models (PLMs) are the cornerstone of the most recent Information Retrieval (IR) models, the way they encode semantic knowledge is particularly important. However, little attention has been given to studying the PLMs' capability to capture hierarchical semantic knowledge. Traditionally, evaluating such knowledge encoded in PLMs relies on their performance on a task-dependent evaluation approach based on proxy tasks, such as hypernymy detection. Unfortunately, this approach potentially ignores other implicit and complex taxonomic relations. In this work, we propose a task-agnostic evaluation method able to evaluate to what extent PLMs can capture complex taxonomy relations, such as ancestors and siblings. The evaluation is based on intrinsic properties that capture the hierarchical nature of taxonomies. Our experimental evaluation shows that the lexico-semantic knowledge implicitly encoded in PLMs does not always capture hierarchical relations. We further demonstrate that the proposed properties can be injected into PLMs to improve their understanding of hierarchy. Through evaluations on taxonomy reconstruction, hypernym discovery and reading comprehension tasks, we show that the knowledge about hierarchy is moderately but not systematically transferable across tasks.

This research study delves into the conceptualization, development, and deployment of an innovative learning analytics tool, leveraging the capabilities of OpenAI's GPT-4 model. This tool is designed to quantify student engagement, map learning progression, and evaluate the efficacy of diverse instructional strategies within an educational context. Through the analysis of various critical data points such as students' stress levels, curiosity, confusion, agitation, topic preferences, and study methods, the tool offers a rich, multi-dimensional view of the learning environment. Furthermore, it employs Bloom's taxonomy as a framework to gauge the cognitive levels addressed by students' questions, thereby elucidating their learning progression. The information gathered from these measurements can empower educators by providing valuable insights to enhance teaching methodologies, pinpoint potential areas for improvement, and craft personalized interventions for individual students. The study articulates the design intricacies, implementation strategy, and thorough evaluation of the learning analytics tool, underscoring its prospective contributions to enhancing educational outcomes and bolstering student success. Moreover, the practicalities of integrating the tool within existing educational platforms and the requisite robust, secure, and scalable technical infrastructure are addressed. This research opens avenues for harnessing AI's potential in shaping the future of education, facilitating data-driven pedagogical decisions, and ultimately fostering a more conducive, personalized learning environment.

Without an agreed-upon definition of intelligence, asking "is this system intelligent?"" is an untestable question. This lack of consensus hinders research, and public perception, on Artificial Intelligence (AI), particularly since the rise of generative- and large-language models. Most work on precisely capturing what we mean by "intelligence" has come from the fields of philosophy, psychology, and cognitive science. Because these perspectives are intrinsically linked to intelligence as it is demonstrated by natural creatures, we argue such fields cannot, and will not, provide a sufficiently rigorous definition that can be applied to artificial means. Thus, we present an argument for a purely functional, black-box definition of intelligence, distinct from how that intelligence is actually achieved; focusing on the "what", rather than the "how". To achieve this, we first distinguish other related concepts (sentience, sensation, agency, etc.) from the notion of intelligence, particularly identifying how these concepts pertain to artificial intelligent systems. As a result, we achieve a formal definition of intelligence that is conceptually testable from only external observation, that suggests intelligence is a continuous variable. We conclude by identifying challenges that still remain towards quantifiable measurement. This work provides a useful perspective for both the development of AI, and for public perception of the capabilities and risks of AI.

As Large Language Models (LLMs) become increasingly integrated into everyday life, their capabilities to understand and emulate human cognition are under steady examination. This study investigates the ability of LLMs to comprehend and interpret linguistic pragmatics, an aspect of communication that considers context and implied meanings. Using Grice's communication principles, LLMs and human subjects (N=76) were evaluated based on their responses to various dialogue-based tasks. The findings revealed the superior performance and speed of LLMs, particularly GPT4, over human subjects in interpreting pragmatics. GPT4 also demonstrated accuracy in the pre-testing of human-written samples, indicating its potential in text analysis. In a comparative analysis of LLMs using human individual and average scores, the models exhibited significant chronological improvement. The models were ranked from lowest to highest score, with GPT2 positioned at 78th place, GPT3 ranking at 23rd, Bard at 10th, GPT3.5 placing 5th, Best Human scoring 2nd, and GPT4 achieving the top spot. The findings highlight the remarkable progress made in the development and performance of these LLMs. Future studies should consider diverse subjects, multiple languages, and other cognitive aspects to fully comprehend the capabilities of LLMs. This research holds significant implications for the development and application of AI-based models in communication-centered sectors.

Building general-purpose robots that can operate seamlessly, in any environment, with any object, and utilizing various skills to complete diverse tasks has been a long-standing goal in Artificial Intelligence. Unfortunately, however, most existing robotic systems have been constrained - having been designed for specific tasks, trained on specific datasets, and deployed within specific environments. These systems usually require extensively-labeled data, rely on task-specific models, have numerous generalization issues when deployed in real-world scenarios, and struggle to remain robust to distribution shifts. Motivated by the impressive open-set performance and content generation capabilities of web-scale, large-capacity pre-trained models (i.e., foundation models) in research fields such as Natural Language Processing (NLP) and Computer Vision (CV), we devote this survey to exploring (i) how these existing foundation models from NLP and CV can be applied to the field of robotics, and also exploring (ii) what a robotics-specific foundation model would look like. We begin by providing an overview of what constitutes a conventional robotic system and the fundamental barriers to making it universally applicable. Next, we establish a taxonomy to discuss current work exploring ways to leverage existing foundation models for robotics and develop ones catered to robotics. Finally, we discuss key challenges and promising future directions in using foundation models for enabling general-purpose robotic systems. We encourage readers to view our living GitHub repository of resources, including papers reviewed in this survey as well as related projects and repositories for developing foundation models for robotics.

Differentiable particle filters are an emerging class of sequential Bayesian inference techniques that use neural networks to construct components in state space models. Existing approaches are mostly based on offline supervised training strategies. This leads to the delay of the model deployment and the obtained filters are susceptible to distribution shift of test-time data. In this paper, we propose an online learning framework for differentiable particle filters so that model parameters can be updated as data arrive. The technical constraint is that there is no known ground truth state information in the online inference setting. We address this by adopting an unsupervised loss to construct the online model updating procedure, which involves a sequence of filtering operations for online maximum likelihood-based parameter estimation. We empirically evaluate the effectiveness of the proposed method, and compare it with supervised learning methods in simulation settings including a multivariate linear Gaussian state-space model and a simulated object tracking experiment.

Recent advances in large language models (LLMs) like GPT-3 and GPT-4 have opened up new opportunities for text analysis in political science. They promise automation with better results and less programming. In this study, we evaluate LLMs on three original coding tasks of non-English political science texts, and we provide a detailed description of a general workflow for using LLMs for text coding in political science research. Our use case offers a practical guide for researchers looking to incorporate LLMs into their research on text analysis. We find that, when provided with detailed label definitions and coding examples, an LLM can be as good as or even better than a human annotator while being much faster (up to hundreds of times), considerably cheaper (costing up to 60% less than human coding), and much easier to scale to large amounts of text. Overall, LLMs present a viable option for most text coding projects.

As the use of large language models (LLMs) increases within society, as does the risk of their misuse. Appropriate safeguards must be in place to ensure LLM outputs uphold the ethical standards of society, highlighting the positive role that artificial intelligence technologies can have. Recent events indicate ethical concerns around conventionally trained LLMs, leading to overall unsafe user experiences. This motivates our research question: how do we ensure LLM alignment? In this work, we introduce a test suite of unique prompts to foster the development of aligned LLMs that are fair, safe, and robust. We show that prompting LLMs at every step of the development pipeline, including data curation, pre-training, and fine-tuning, will result in an overall more responsible model. Our test suite evaluates outputs from four state-of-the-art language models: GPT-3.5, GPT-4, OPT, and LLaMA-2. The assessment presented in this paper highlights a gap between societal alignment and the capabilities of current LLMs. Additionally, implementing a test suite such as ours lowers the environmental overhead of making models safe and fair.