State-of-the-art natural language processing (NLP) models are trained on massive training corpora, and report a superlative performance on evaluation datasets. This survey delves into an important attribute of these datasets: the dialect of a language. Motivated by the performance degradation of NLP models for dialectic datasets and its implications for the equity of language technologies, we survey past research in NLP for dialects in terms of datasets, and approaches. We describe a wide range of NLP tasks in terms of two categories: natural language understanding (NLU) (for tasks such as dialect classification, sentiment analysis, parsing, and NLU benchmarks) and natural language generation (NLG) (for summarisation, machine translation, and dialogue systems). The survey is also broad in its coverage of languages which include English, Arabic, German among others. We observe that past work in NLP concerning dialects goes deeper than mere dialect classification, and . This includes early approaches that used sentence transduction that lead to the recent approaches that integrate hypernetworks into LoRA. We expect that this survey will be useful to NLP researchers interested in building equitable language technologies by rethinking LLM benchmarks and model architectures.

In this paper, we introduce Concise Chain-of-Thought (CCoT) prompting. We compared standard CoT and CCoT prompts to see how conciseness impacts response length and correct-answer accuracy. We evaluated this using GPT-3.5 and GPT-4 with a multiple-choice question-and-answer (MCQA) benchmark. CCoT reduced average response length by 48.70% for both GPT-3.5 and GPT-4 while having a negligible impact on problem-solving performance. However, on math problems, GPT-3.5 with CCoT incurs a performance penalty of 27.69%. Overall, CCoT leads to an average per-token cost reduction of 22.67%. These results have practical implications for AI systems engineers using LLMs to solve real-world problems with CoT prompt-engineering techniques. In addition, these results provide more general insight for AI researchers studying the emergent behavior of step-by-step reasoning in LLMs.

We study and quantify the problem of forgetting when fine-tuning pre-trained large language models (LLMs) on a downstream task. We find that parameter-efficient fine-tuning (PEFT) strategies, such as Low-Rank Adapters (LoRA), still suffer from catastrophic forgetting. In particular, we identify a strong inverse linear relationship between the fine-tuning performance and the amount of forgetting when fine-tuning LLMs with LoRA. We further obtain precise scaling laws that show forgetting increases as a shifted power law in the number of parameters fine-tuned and the number of update steps. We also examine the impact of forgetting on knowledge, reasoning, and the safety guardrails trained into Llama 2 7B chat. Our study suggests that forgetting cannot be avoided through early stopping or by varying the number of parameters fine-tuned. We believe this opens up an important safety-critical direction for future research to evaluate and develop fine-tuning schemes which mitigate forgetting.

We propose a new benchmark evaluating the performance of multimodal large language models on rebus puzzles. The dataset covers 333 original examples of image-based wordplay, cluing 13 categories such as movies, composers, major cities, and food. To achieve good performance on the benchmark of identifying the clued word or phrase, models must combine image recognition and string manipulation with hypothesis testing, multi-step reasoning, and an understanding of human cognition, making for a complex, multimodal evaluation of capabilities. We find that proprietary models such as GPT-4V and Gemini Pro significantly outperform all other tested models. However, even the best model has a final accuracy of just 24%, highlighting the need for substantial improvements in reasoning. Further, models rarely understand all parts of a puzzle, and are almost always incapable of retroactively explaining the correct answer. Our benchmark can therefore be used to identify major shortcomings in the knowledge and reasoning of multimodal large language models.

In this paper, we introduce "InfiAgent-DABench", the first benchmark specifically designed to evaluate LLM-based agents in data analysis tasks. This benchmark contains DAEval, a dataset consisting of 311 data analysis questions derived from 55 CSV files, and an agent framework to evaluate LLMs as data analysis agents. We adopt a format-prompting technique, ensuring questions to be closed-form that can be automatically evaluated. Our extensive benchmarking of 23 state-of-the-art LLMs uncovers the current challenges encountered in data analysis tasks. In addition, we have developed DAAgent, a specialized agent trained on instruction-tuning datasets. Evaluation datasets and toolkits for InfiAgent-DABench are released at https://github.com/InfiAgent/InfiAgent.

This paper introduces CADgpt, an innovative plugin integrating Natural Language Processing (NLP) with Rhino3D for enhancing 3D modelling in computer-aided design (CAD) environments. Leveraging OpenAI's GPT-4, CADgpt simplifies the CAD interface, enabling users, particularly beginners, to perform complex 3D modelling tasks through intuitive natural language commands. This approach significantly reduces the learning curve associated with traditional CAD software, fostering a more inclusive and engaging educational environment. The paper discusses CADgpt's technical architecture, including its integration within Rhino3D and the adaptation of GPT-4 capabilities for CAD tasks. It presents case studies demonstrating CADgpt's efficacy in various design scenarios, highlighting its potential to democratise design education by making sophisticated design tools accessible to a broader range of students. The discussion further explores CADgpt's implications for pedagogy and curriculum development, emphasising its role in enhancing creative exploration and conceptual thinking in design education. Keywords: Natural Language Processing, Computer-Aided Design, 3D Modelling, Design Automation, Design Education, Architectural Education

The recent advances in large language models (LLMs) have led to the creation of many modular AI agents. These agents employ LLMs as zero-shot learners to perform sub-tasks in order to solve complex tasks set forth by human users. We propose an approach to enhance the robustness and performance of modular AI agents that utilize LLMs as zero-shot learners. Our iterative machine teaching method offers an efficient way to teach AI agents over time with limited human feedback, addressing the limit posed by the quality of zero-shot learning. We advocate leveraging the data traces from initial deployments and outputs or annotations from the zero-shot learners to train smaller and task-specific substitute models which can reduce both the monetary costs and environmental impact. Our machine teaching process avails human expertise to correct examples with a high likelihood of misannotations. Results on three tasks, common to conversational AI agents, show that close-to-oracle performance can be achieved with supervision on 20-70% of the dataset depending upon the complexity of the task and performance of zero-shot learners.

Large language models (LLMs) have made significant progress in code generation tasks, but their performance in tackling programming problems with complex data structures and algorithms remains suboptimal. To address this issue, we propose an in-context learning approach that guides LLMs to debug by using a "print debugging" method, which involves inserting print statements to trace and analysing logs for fixing the bug. We collect a Leetcode problem dataset and evaluate our method using the Leetcode online judging system. Experiments with GPT-4 demonstrate the effectiveness of our approach, outperforming rubber duck debugging in easy and medium-level Leetcode problems by 1.5% and 17.9%.

Statements involving metalinguistic self-reference ("This paper has six sections.") are prevalent in many domains. Can large language models (LLMs) handle such language? In this paper, we present "I am a Strange Dataset", a new dataset for addressing this question. There are two subtasks: generation and verification. In generation, models continue statements like "The penultimate word in this sentence is" (where a correct continuation is "is"). In verification, models judge the truth of statements like "The penultimate word in this sentence is sentence." (false). We also provide minimally different metalinguistic non-self-reference examples to complement the main dataset by probing for whether models can handle metalinguistic language at all. The dataset is hand-crafted by experts and validated by non-expert annotators. We test a variety of open-source LLMs (7B to 70B parameters) as well as closed-source LLMs through APIs. All models perform close to chance across both subtasks and even on the non-self-referential metalinguistic control data, though we find some steady improvement with model scale. GPT 4 is the only model to consistently do significantly better than chance, and it is still only in the 60% range, while our untrained human annotators score well in the 89-93% range. The dataset and evaluation toolkit are available at https://github.com/TristanThrush/i-am-a-strange-dataset.

The argument sufficiency assessment task aims to determine if the premises of a given argument support its conclusion. To tackle this task, existing works often train a classifier on data annotated by humans. However, annotating data is laborious, and annotations are often inconsistent due to subjective criteria. Motivated by the probability of sufficiency (PS) definition in the causal literature, we propose CASA, a zero-shot causality-driven argument sufficiency assessment framework. PS measures how likely introducing the premise event would lead to the conclusion, when both the premise and conclusion events are absent. To estimate this probability, we propose to use large language models (LLMs) to generate contexts that are inconsistent with the premise and conclusion, and revise them by injecting the premise event. Experiments on two logical fallacy detection datasets demonstrate that CASA accurately identifies insufficient arguments. We further deploy CASA in a writing assistance application, and find that suggestions generated by CASA enhance the sufficiency of student-written arguments. Code and data are available at https://github.com/xxxiaol/CASA.