Recent work has shown that prompting language models with code-like representations of natural language leads to performance improvements on structured reasoning tasks. However, such tasks comprise only a small subset of all natural language tasks. In our work, we seek to answer whether or not code-prompting is the preferred way of interacting with language models in general. We compare code and text prompts across three popular GPT models (davinci, code-davinci-002, and text-davinci-002) on a broader selection of tasks (e.g., QA, sentiment, summarization) and find that with few exceptions, code prompts do not consistently outperform text prompts. Furthermore, we show that the style of code prompt has a large effect on performance for some but not all tasks and that fine-tuning on text instructions leads to better relative performance of code prompts.
相關內容
This paper explores how AI-owners can develop safeguards for AI-generated content by drawing from established codes of conduct and ethical standards in other content-creation industries. It delves into the current state of ethical awareness on Large Language Models (LLMs). By dissecting the mechanism of content generation by LLMs, four key areas (upstream/downstream and at user prompt/answer), where safeguards could be effectively applied, are identified. A comparative analysis of these four areas follows and includes an evaluation of the existing ethical safeguards in terms of cost, effectiveness, and alignment with established industry practices. The paper's key argument is that existing IT-related ethical codes, while adequate for traditional IT engineering, are inadequate for the challenges posed by LLM-based content generation. Drawing from established practices within journalism, we propose potential standards for businesses involved in distributing and selling LLM-generated content. Finally, potential conflicts of interest between dataset curation at upstream and ethical benchmarking downstream are highlighted to underscore the need for a broader evaluation beyond mere output. This study prompts a nuanced conversation around ethical implications in this rapidly evolving field of content generation.
Vision-Language (VL) models with the Two-Tower architecture have dominated visual-language representation learning in recent years. Current VL models either use lightweight uni-modal encoders and learn to extract, align and fuse both modalities simultaneously in a deep cross-modal encoder, or feed the last-layer uni-modal representations from the deep pre-trained uni-modal encoders into the top cross-modal encoder. Both approaches potentially restrict vision-language representation learning and limit model performance. In this paper, we propose BridgeTower, which introduces multiple bridge layers that build a connection between the top layers of uni-modal encoders and each layer of the cross-modal encoder. This enables effective bottom-up cross-modal alignment and fusion between visual and textual representations of different semantic levels of pre-trained uni-modal encoders in the cross-modal encoder. Pre-trained with only 4M images, BridgeTower achieves state-of-the-art performance on various downstream vision-language tasks. In particular, on the VQAv2 test-std set, BridgeTower achieves an accuracy of 78.73%, outperforming the previous state-of-the-art model METER by 1.09% with the same pre-training data and almost negligible additional parameters and computational costs. Notably, when further scaling the model, BridgeTower achieves an accuracy of 81.15%, surpassing models that are pre-trained on orders-of-magnitude larger datasets. Code and checkpoints are available at //github.com/microsoft/BridgeTower.
With the ever-increasing utilization of natural language processing (NLP), we started to witness over the past few years a significant transformation in our interaction with legal texts. This technology has advanced the analysis and enhanced the understanding of complex legal terminology and contexts. The development of recent large language models (LLMs), particularly ChatGPT, has also introduced a revolutionary contribution to the way that legal texts can be processed and comprehended. In this paper, we present our work on a cooperative-legal question-answering LLM-based chatbot, where we developed a set of legal questions about Palestinian cooperatives, associated with their regulations and compared the auto-generated answers by the chatbot to their correspondences that are designed by a legal expert. To evaluate the proposed chatbot, we have used 50 queries generated by the legal expert and compared the answers produced by the chart to their relevance judgments. Finding demonstrated that an overall accuracy rate of 82% has been achieved when answering the queries, while exhibiting an F1 score equivalent to 79%.
Language instructions and demonstrations are two natural ways for users to teach robots personalized tasks. Recent progress in Large Language Models (LLMs) has shown impressive performance in translating language instructions into code for robotic tasks. However, translating demonstrations into task code continues to be a challenge due to the length and complexity of both demonstrations and code, making learning a direct mapping intractable. This paper presents Demo2Code, a novel framework that generates robot task code from demonstrations via an extended chain-of-thought and defines a common latent specification to connect the two. Our framework employs a robust two-stage process: (1) a recursive summarization technique that condenses demonstrations into concise specifications, and (2) a code synthesis approach that expands each function recursively from the generated specifications. We conduct extensive evaluation on various robot task benchmarks, including a novel game benchmark Robotouille, designed to simulate diverse cooking tasks in a kitchen environment. The project's website is available at //portal-cornell.github.io/demo2code-webpage
Current multimodal models, aimed at solving Vision and Language (V+L) tasks, predominantly repurpose Vision Encoders (VE) as feature extractors. While many VEs -- of different architectures, trained on different data and objectives -- are publicly available, they are not designed for the downstream V+L tasks. Nonetheless, most current work assumes that a \textit{single} pre-trained VE can serve as a general-purpose encoder. In this work, we focus on analysis and aim to understand whether the information stored within different VEs is complementary, i.e. if providing the model with features from multiple VEs can improve the performance on a target task, and how they are combined. We exhaustively experiment with three popular VEs on six downstream V+L tasks and analyze the attention and VE-dropout patterns. Our analyses suggest that diverse VEs complement each other, resulting in improved downstream V+L task performance, where the improvements are not due to simple ensemble effects (i.e. the performance does not always improve when increasing the number of encoders). We demonstrate that future VEs, which are not \textit{repurposed}, but explicitly \textit{designed} for V+L tasks, have the potential of improving performance on the target V+L tasks.
Story generation and understanding -- as with all NLG/NLU tasks -- has seen a surge in neurosymbolic work. Researchers have recognized that, while large language models (LLMs) have tremendous utility, they can be augmented with symbolic means to be even better and to make up for any flaws that the neural networks might have. However, symbolic methods are extremely costly in terms of the amount of time and expertise needed to create them. In this work, we capitalize on state-of-the-art Code-LLMs, such as Codex, to bootstrap the use of symbolic methods for tracking the state of stories and aiding in story understanding. We show that our CoRRPUS system and abstracted prompting procedures can beat current state-of-the-art structured LLM techniques on pre-existing story understanding tasks (bAbI Task 2 and Re^3) with minimal hand engineering. We hope that this work can help highlight the importance of symbolic representations and specialized prompting for LLMs as these models require some guidance for performing reasoning tasks properly.
In this work we explore recent advances in instruction-tuning language models on a range of open instruction-following datasets. Despite recent claims that open models can be on par with state-of-the-art proprietary models, these claims are often accompanied by limited evaluation, making it difficult to compare models across the board and determine the utility of various resources. We provide a large set of instruction-tuned models from 6.7B to 65B parameters in size, trained on 12 instruction datasets ranging from manually curated (e.g., OpenAssistant) to synthetic and distilled (e.g., Alpaca) and systematically evaluate them on their factual knowledge, reasoning, multilinguality, coding, and open-ended instruction following abilities through a collection of automatic, model-based, and human-based metrics. We further introduce T\"ulu, our best performing instruction-tuned model suite finetuned on a combination of high-quality open resources. Our experiments show that different instruction-tuning datasets can uncover or enhance specific skills, while no single dataset (or combination) provides the best performance across all evaluations. Interestingly, we find that model and human preference-based evaluations fail to reflect differences in model capabilities exposed by benchmark-based evaluations, suggesting the need for the type of systemic evaluation performed in this work. Our evaluations show that the best model in any given evaluation reaches on average 83% of ChatGPT performance, and 68% of GPT-4 performance, suggesting that further investment in building better base models and instruction-tuning data is required to close the gap. We release our instruction-tuned models, including a fully finetuned 65B T\"ulu, along with our code, data, and evaluation framework at //github.com/allenai/open-instruct to facilitate future research.
Code LLMs are being rapidly deployed and there is evidence that they can make professional programmers more productive. Current benchmarks for code generation measure whether models generate correct programs given an expert prompt. In this paper, we present a new benchmark containing multiple prompts per problem, written by a specific population of non-expert prompters: beginning programmers. StudentEval contains 1,749 prompts for 48 problems, written by 80 students who have only completed one semester of Python programming. Our students wrote these prompts while working interactively with a Code LLM, and we observed very mixed success rates. We use StudentEval to evaluate 5 Code LLMs and find that StudentEval is a better discriminator of model performance than existing benchmarks. We analyze the prompts and find significant variation in students' prompting techniques. We also find that nondeterministic LLM sampling could mislead students into thinking that their prompts are more (or less) effective than they actually are, which has implications for how to teach with Code LLMs.
Large Language Models (LLMs) have made remarkable advancements in the field of artificial intelligence, significantly reshaping the human-computer interaction. We not only focus on the performance of LLMs, but also explore their features from a psychological perspective, acknowledging the importance of understanding their behavioral characteristics. Our study examines the behavioral patterns displayed by LLMs by employing trait theory, a psychological framework. We first focus on evaluating the consistency of personality types exhibited by ChatGPT. Furthermore, experiments include cross-lingual effects on seven additional languages, and the investigation of six other LLMs. Moreover, the study investigates whether ChatGPT can exhibit personality changes in response to instructions or contextual cues. The findings show that ChatGPT consistently maintains its ENFJ personality regardless of instructions or contexts. By shedding light on the personalization of LLMs, we anticipate that our study will serve as a catalyst for further research in this field.
This paper surveys and organizes research works in a new paradigm in natural language processing, which we dub "prompt-based learning". Unlike traditional supervised learning, which trains a model to take in an input x and predict an output y as P(y|x), prompt-based learning is based on language models that model the probability of text directly. To use these models to perform prediction tasks, the original input x is modified using a template into a textual string prompt x' that has some unfilled slots, and then the language model is used to probabilistically fill the unfilled information to obtain a final string x, from which the final output y can be derived. This framework is powerful and attractive for a number of reasons: it allows the language model to be pre-trained on massive amounts of raw text, and by defining a new prompting function the model is able to perform few-shot or even zero-shot learning, adapting to new scenarios with few or no labeled data. In this paper we introduce the basics of this promising paradigm, describe a unified set of mathematical notations that can cover a wide variety of existing work, and organize existing work along several dimensions, e.g.the choice of pre-trained models, prompts, and tuning strategies. To make the field more accessible to interested beginners, we not only make a systematic review of existing works and a highly structured typology of prompt-based concepts, but also release other resources, e.g., a website //pretrain.nlpedia.ai/ including constantly-updated survey, and paperlist.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191