LLMs trained in the understanding of programming syntax are now providing effective assistance to developers and are being used in programming education such as in generation of coding problem examples or providing code explanations. A key aspect of programming education is understanding and dealing with error message. However, 'logical errors' in which the program operates against the programmer's intentions do not receive error messages from the compiler. In this study, building on existing research on programming errors, we first define the types of logical errors that can occur in programming in general. Based on the definition, we propose an effective approach for detecting logical errors with LLMs that makes use of relations among error types in the Chain-of-Thought and Tree-of-Thought prompts. The experimental results indicate that when such logical error descriptions in the prompt are used, the average classification performance is about 21% higher than the ones without them. We also conducted an experiment for exploiting the relations among errors in generating a new logical error dataset using LLMs. As there is very limited dataset for logical errors such benchmark dataset can be very useful for various programming related applications. We expect that our work can assist novice programmers in identifying the causes of code errors and correct them more effectively.

This study presents a targeted model editing analysis focused on the latest large language model, Llama-3. We explore the efficacy of popular model editing techniques - ROME, MEMIT, and EMMET, which are designed for precise layer interventions. We identify the most effective layers for targeted edits through an evaluation that encompasses up to 4096 edits across three distinct strategies: sequential editing, batch editing, and a hybrid approach we call as sequential-batch editing. Our findings indicate that increasing edit batch-sizes may degrade model performance more significantly than using smaller edit batches sequentially for equal number of edits. With this, we argue that sequential model editing is an important component for scaling model editing methods and future research should focus on methods that combine both batched and sequential editing. This observation suggests a potential limitation in current model editing methods which push towards bigger edit batch sizes, and we hope it paves way for future investigations into optimizing batch sizes and model editing performance.

Large language model pre-training has become increasingly expensive, with most practitioners relying on scaling laws to allocate compute budgets for model size and training tokens, commonly referred to as Compute-Optimal or Chinchilla Optimal. In this paper, we hypothesize a new scaling law that suggests model performance depends mostly on the amount of compute spent for transformer-based models, independent of the specific allocation to model size and dataset size. Using this unified scaling law, we predict that (a) for inference efficiency, training should prioritize smaller model sizes and larger training datasets, and (b) assuming the exhaustion of available web datasets, scaling the model size might be the only way to further improve model performance. Figure 1: Model performance scales linearly with the log scale of compute.

We introduce WorkBench: a benchmark dataset for evaluating agents' ability to execute tasks in a workplace setting. WorkBench contains a sandbox environment with five databases, 26 tools, and 690 tasks. These tasks represent common business activities, such as sending emails and scheduling meetings. The tasks in WorkBench are challenging as they require planning, tool selection, and often multiple actions. If a task has been successfully executed, one (or more) of the database values may change. The correct outcome for each task is unique and unambiguous, which allows for robust, automated evaluation. We call this key contribution outcome-centric evaluation. We evaluate five existing ReAct agents on WorkBench, finding they successfully complete as few as 3% of tasks (Llama2-70B), and just 43% for the best-performing (GPT-4). We further find that agents' errors can result in the wrong action being taken, such as an email being sent to the wrong person. WorkBench reveals weaknesses in agents' ability to undertake common business activities, raising questions about their use in high-stakes workplace settings. WorkBench is publicly available as a free resource at https://github.com/olly-styles/WorkBench.

Questions combine our mastery of language with our remarkable facility for reasoning about uncertainty. How do people navigate vast hypothesis spaces to pose informative questions given limited cognitive resources? We study these tradeoffs in a classic grounded question-asking task based on the board game Battleship. Our language-informed program sampling (LIPS) model uses large language models (LLMs) to generate natural language questions, translate them into symbolic programs, and evaluate their expected information gain. We find that with a surprisingly modest resource budget, this simple Monte Carlo optimization strategy yields informative questions that mirror human performance across varied Battleship board scenarios. In contrast, LLM-only baselines struggle to ground questions in the board state; notably, GPT-4V provides no improvement over non-visual baselines. Our results illustrate how Bayesian models of question-asking can leverage the statistics of language to capture human priors, while highlighting some shortcomings of pure LLMs as grounded reasoners.

The prevalence of unwarranted beliefs, spanning pseudoscience, logical fallacies, and conspiracy theories, presents substantial societal hurdles and the risk of disseminating misinformation. Utilizing established psychometric assessments, this study explores the capabilities of large language models (LLMs) vis-a-vis the average human in detecting prevalent logical pitfalls. We undertake a philosophical inquiry, juxtaposing the rationality of humans against that of LLMs. Furthermore, we propose methodologies for harnessing LLMs to counter misconceptions, drawing upon psychological models of persuasion such as cognitive dissonance theory and elaboration likelihood theory. Through this endeavor, we highlight the potential of LLMs as personalized misinformation debunking agents.

Gender bias research has been pivotal in revealing undesirable behaviors in large language models, exposing serious gender stereotypes associated with occupations, and emotions. A key observation in prior work is that models reinforce stereotypes as a consequence of the gendered correlations that are present in the training data. In this paper, we focus on bias where the effect from training data is unclear, and instead address the question: Do language models still exhibit gender bias in non-stereotypical settings? To do so, we introduce UnStereoEval (USE), a novel framework tailored for investigating gender bias in stereotype-free scenarios. USE defines a sentence-level score based on pretraining data statistics to determine if the sentence contain minimal word-gender associations. To systematically benchmark the fairness of popular language models in stereotype-free scenarios, we utilize USE to automatically generate benchmarks without any gender-related language. By leveraging USE's sentence-level score, we also repurpose prior gender bias benchmarks (Winobias and Winogender) for non-stereotypical evaluation. Surprisingly, we find low fairness across all 28 tested models. Concretely, models demonstrate fair behavior in only 9%-41% of stereotype-free sentences, suggesting that bias does not solely stem from the presence of gender-related words. These results raise important questions about where underlying model biases come from and highlight the need for more systematic and comprehensive bias evaluation. We release the full dataset and code at https://ucinlp.github.io/unstereo-eval.

Building on the methodology outlined (NLP) has seen significant advancements, beginning by Kanakarajan and Sankarasubbu (2023), with the introduction of word embeddings we assessed the zero-shot performance of various (Mikolov et al., 2013), followed by transformer instruction-tuned LLMs to identify the most effective architectures like BERT (Vaswani et al., 2017; Devlin model. Upon selecting the best LLM, we introduced et al., 2019), and specialized language models an auxiliary module during the fine-tuning (LMs) such as BioBERT (Lee et al., 2020) and process, which emphasized learning "hard" examples. PubMedBERT (Gu et al., 2021) tailored for the Taking inspiration from Korakakis and Vlachos biomedical domain. The advent of large language (2023), who experimented with various configurations models (LLMs) like GPT-3 (Brown et al., 2020), for the auxiliary module and highlighted commonly known as Chat-GPT, has further pushed its substantial impact on the final NLI system's the boundaries of NLP, showcasing capabilities performance, we explored various architectures for in diverse NLP tasks and even reasoning.

Automatic grading and feedback have been long studied using traditional machine learning and deep learning techniques using language models. With the recent accessibility to high performing large language models (LLMs) like LLaMA-2, there is an opportunity to investigate the use of these LLMs for automatic grading and feedback generation. Despite the increase in performance, LLMs require significant computational resources for fine-tuning and additional specific adjustments to enhance their performance for such tasks. To address these issues, Parameter Efficient Fine-tuning (PEFT) methods, such as LoRA and QLoRA, have been adopted to decrease memory and computational requirements in model fine-tuning. This paper explores the efficacy of PEFT-based quantized models, employing classification or regression head, to fine-tune LLMs for automatically assigning continuous numerical grades to short answers and essays, as well as generating corresponding feedback. We conducted experiments on both proprietary and open-source datasets for our tasks. The results show that prediction of grade scores via finetuned LLMs are highly accurate, achieving less than 3% error in grade percentage on average. For providing graded feedback fine-tuned 4-bit quantized LLaMA-2 13B models outperform competitive base models and achieve high similarity with subject matter expert feedback in terms of high BLEU and ROUGE scores and qualitatively in terms of feedback. The findings from this study provide important insights into the impacts of the emerging capabilities of using quantization approaches to fine-tune LLMs for various downstream tasks, such as automatic short answer scoring and feedback generation at comparatively lower costs and latency.

Multiple choice questions (MCQs) are a popular method for evaluating students' knowledge due to their efficiency in administration and grading. Crafting high-quality math MCQs is a labor-intensive process that requires educators to formulate precise stems and plausible distractors. Recent advances in large language models (LLMs) have sparked interest in automating MCQ creation, but challenges persist in ensuring mathematical accuracy and addressing student errors. This paper introduces a prototype tool designed to facilitate collaboration between LLMs and educators for streamlining the math MCQ generation process. We conduct a pilot study involving math educators to investigate how the tool can help them simplify the process of crafting high-quality math MCQs. We found that while LLMs can generate well-formulated question stems, their ability to generate distractors that capture common student errors and misconceptions is limited. Nevertheless, a human-AI collaboration has the potential to enhance the efficiency and effectiveness of MCQ generation.