Large pre-trained neural language models have brought immense progress to both NLP and software engineering. Models in OpenAI's GPT series now dwarf Google's BERT and Meta's RoBERTa, which previously set new benchmarks on a wide range of NLP applications. These models are trained on massive corpora of heterogeneous data from web crawls, which enables them to learn general language patterns and semantic relationships. However, the largest models are both expensive to train and deploy and are often closed-source, so we lack access to their data and design decisions. We argue that this trend towards large, general-purpose models should be complemented with single-purpose, more modestly sized pre-trained models. In this work, we take StackOverflow (SO) as a domain example in which large volumes of rich aligned code and text data is available. We adopt standard practices for pre-training large language models, including using a very large context size (2,048 tokens), batch size (0.5M tokens) and training set (27B tokens), coupled with a powerful toolkit (Megatron-LM), to train two models: SOBertBase, with 109M parameters, and SOBertLarge with 762M parameters, at a budget of just $\$187$ and $\$800$ each. We compare the performance of our models with both the previous SOTA model trained on SO data exclusively as well general-purpose BERT models and OpenAI's ChatGPT on four SO-specific downstream tasks - question quality prediction, closed question prediction, named entity recognition and obsoletion prediction (a new task we introduce). Not only do our models consistently outperform all baselines, the smaller model is often sufficient for strong results. Both models are released to the public. These results demonstrate that pre-training both extensively and properly on in-domain data can yield a powerful and affordable alternative to leveraging closed-source general-purpose models.
相關內容
ACM/IEEE第23屆模型驅動工程語言和系統國際會議，是模型驅動軟件和系統工程的首要會議系列，由ACM-SIGSOFT和IEEE-TCSE支持組織。自1998年以來，模型涵蓋了建模的各個方面，從語言和方法到工具和應用程序。模特的參加者來自不同的背景，包括研究人員、學者、工程師和工業專業人士。MODELS 2019是一個論壇，參與者可以圍繞建模和模型驅動的軟件和系統交流前沿研究成果和創新實踐經驗。今年的版本將為建模社區提供進一步推進建模基礎的機會，并在網絡物理系統、嵌入式系統、社會技術系統、云計算、大數據、機器學習、安全、開源等新興領域提出建模的創新應用以及可持續性。
官網鏈接： ·
MoDELS
·
大語言模型
·
ChatGPT
·
知識 (knowledge)
·
Large language models (LLMs) with chat-based capabilities, such as ChatGPT, are widely used in various workflows. However, due to a limited understanding of these large-scale models, users struggle to use this technology and experience different kinds of dissatisfaction. Researchers have introduced several methods, such as prompt engineering, to improve model responses. However, they focus on enhancing the model's performance in specific tasks, and little has been investigated on how to deal with the user dissatisfaction resulting from the model's responses. Therefore, with ChatGPT as the case study, we examine users' dissatisfaction along with their strategies to address the dissatisfaction. After organizing users' dissatisfaction with LLM into seven categories based on a literature review, we collected 511 instances of dissatisfactory ChatGPT responses from 107 users and their detailed recollections of dissatisfactory experiences, which we released as a publicly accessible dataset. Our analysis reveals that users most frequently experience dissatisfaction when ChatGPT fails to grasp their intentions, while they rate the severity of dissatisfaction related to accuracy the highest. We also identified four tactics users employ to address their dissatisfaction and their effectiveness. We found that users often do not use any tactics to address their dissatisfaction, and even when using tactics, 72% of dissatisfaction remained unresolved. Moreover, we found that users with low knowledge of LLMs tend to face more dissatisfaction on accuracy while they often put minimal effort in addressing dissatisfaction. Based on these findings, we propose design implications for minimizing user dissatisfaction and enhancing the usability of chat-based LLM.
Large language models (LLMs) such as ChatGPT are increasingly proficient in understanding and generating a mixture of code and text. Evaluation based on such $\textit{mixture}$ can lead to a more comprehensive understanding of the models' abilities in solving coding problems. However, in this context, current evaluation methods are either limited in task coverage or lack standardization. To address this issue, we propose using category theory as a framework for evaluation. Specifically, morphisms within a code category can represent code debugging and transformation, functors between two categories represent code translation, and functors between a code category and a natural language category represent code generation, explanation, and reproduction. We present an automatic evaluation framework called $\textbf{CatCode}$ ($\textbf{Cat}$egory $\textbf{Code}$) that can comprehensively assess the coding abilities of LLMs, including ChatGPT, Text-Davinci, and CodeGeeX.
Foundation models (FMs), such as Large Language Models (LLMs), have revolutionized software development by enabling new use cases and business models. We refer to software built using FMs as FMware. The unique properties of FMware (e.g., prompts, agents, and the need for orchestration), coupled with the intrinsic limitations of FMs (e.g., hallucination) lead to a completely new set of software engineering challenges. Based on our industrial experience, we identified 10 key SE4FMware challenges that have caused enterprise FMware development to be unproductive, costly, and risky. In this paper, we discuss these challenges in detail and state the path for innovation that we envision. Next, we present FMArts, which is our long-term effort towards creating a cradle-to-grave platform for the engineering of trustworthy FMware. Finally, we (i) show how the unique properties of FMArts enabled us to design and develop a complex FMware for a large customer in a timely manner and (ii) discuss the lessons that we learned in doing so. We hope that the disclosure of the aforementioned challenges and our associated efforts to tackle them will not only raise awareness but also promote deeper and further discussions, knowledge sharing, and innovative solutions across the software engineering discipline.
Recent breakthroughs in Large-scale language models (LLMs) have demonstrated impressive performance on various tasks. The immense sizes of LLMs have led to very high resource demand and cost for running the models. Though the models are largely served using uniform high-caliber GPUs nowadays, utilizing a heterogeneous cluster with a mix of available high- and low-capacity GPUs can potentially substantially reduce the serving cost. There is a lack of designs to support efficient LLM serving using a heterogeneous cluster, while the current solutions focus on model partition and uniform compression among homogeneous devices. This paper proposes LLM-PQ, a system that advocates adaptive model quantization and phase-aware partition to improve LLM serving efficiency on heterogeneous GPU clusters. We carefully decide on mixed-precision model quantization together with phase-aware model partition and micro-batch sizing in distributed LLM serving with an efficient algorithm, to greatly enhance inference throughput while fulfilling user-specified model quality targets. Extensive experiments on production inference workloads in 11 different clusters demonstrate that LLM-PQ achieves up to 2.88x (2.26x on average) throughput improvement in inference, showing great advantages over state-of-the-art works.
This work aims at decreasing the end-to-end generation latency of large language models (LLMs). One of the major causes of the high generation latency is the sequential decoding approach adopted by almost all state-of-the-art LLMs. In this work, motivated by the thinking and writing process of humans, we propose Skeleton-of-Thought (SoT), which first guides LLMs to generate the skeleton of the answer, and then conducts parallel API calls or batched decoding to complete the contents of each skeleton point in parallel. Not only does SoT provide considerable speed-ups across 12 LLMs, but it can also potentially improve the answer quality on several question categories. SoT is an initial attempt at data-centric optimization for inference efficiency, and showcases the potential of eliciting high-quality answers by explicitly planning the answer structure in language.
Growth in computational materials science and initiatives such as the Materials Genome Initiative (MGI) and the European Materials Modelling Council (EMMC) has motivated the development and application of ontologies. A key factor has been increased adoption of the FAIR principles, making research data findable, accessible, interoperable, and reusable (Wilkinson et al. 2016). This paper characterizes semantic interoperability among a subset of materials science ontologies in the MatPortal repository. Background context covers semantic interoperability, ontological commitment, and the materials science ontology landscape. The research focused on MatPortal's two interoperability protocols: LOOM term matching and URI matching. Results report the degree of overlap and demonstrate the different types of ambiguity among ontologies. The discussion considers implications for FAIR and AI, and the conclusion highlight key findings and next steps.
Sample efficiency remains a crucial challenge in applying Reinforcement Learning (RL) to real-world tasks. While recent algorithms have made significant strides in improving sample efficiency, none have achieved consistently superior performance across diverse domains. In this paper, we introduce EfficientZero V2, a general framework designed for sample-efficient RL algorithms. We have expanded the performance of EfficientZero to multiple domains, encompassing both continuous and discrete actions, as well as visual and low-dimensional inputs. With a series of improvements we propose, EfficientZero V2 outperforms the current state-of-the-art (SOTA) by a significant margin in diverse tasks under the limited data setting. EfficientZero V2 exhibits a notable advancement over the prevailing general algorithm, DreamerV3, achieving superior outcomes in 50 of 66 evaluated tasks across diverse benchmarks, such as Atari 100k, Proprio Control, and Vision Control.
Prompt engineering in LLMs has shown potential for improving translation quality. However, the potential of incorporating translation concepts in prompt design remains largely underexplored. Against this backdrop, this paper discusses the effectiveness of incorporating the conceptual tool of translation brief and the personas of translator and author into prompt design for translation tasks in ChatGPT. Findings suggest that, although certain elements are constructive in facilitating human to human communication for translation tasks, their effectiveness is limited for improving translation quality in ChatGPT. This accentuates the need for more explorative research on how translation theorists and practitioners can develop the current set of conceptual tools rooted in the human to human communication paradigm for translation purposes in this emerging workflow involving human machine interaction.
Background: Modern Code Review (MCR) is a lightweight alternative to traditional code inspections. While secondary studies on MCR exist, it is unknown whether the research community has targeted themes that practitioners consider important. Objectives: The objectives are to provide an overview of MCR research, analyze the practitioners' opinions on the importance of MCR research, investigate the alignment between research and practice, and propose future MCR research avenues. Method: We conducted a systematic mapping study to survey state of the art until and including 2021, employed the Q-Methodology to analyze the practitioners' perception of the relevance of MCR research, and analyzed the primary studies' research impact. Results: We analyzed 244 primary studies, resulting in five themes. As a result of the 1,300 survey data points, we found that the respondents are positive about research investigating the impact of MCR on product quality and MCR process properties. In contrast, they are negative about human factor- and support systems-related research. Conclusion: These results indicate a misalignment between the state of the art and the themes deemed important by most survey respondents. Researchers should focus on solutions that can improve the state of MCR practice. We provide an MCR research agenda that can potentially increase the impact of MCR research.
The emergence of large language models (LLMs) has marked a significant breakthrough in natural language processing (NLP), leading to remarkable advancements in text understanding and generation. Nevertheless, alongside these strides, LLMs exhibit a critical tendency to produce hallucinations, resulting in content that is inconsistent with real-world facts or user inputs. This phenomenon poses substantial challenges to their practical deployment and raises concerns over the reliability of LLMs in real-world scenarios, which attracts increasing attention to detect and mitigate these hallucinations. In this survey, we aim to provide a thorough and in-depth overview of recent advances in the field of LLM hallucinations. We begin with an innovative taxonomy of LLM hallucinations, then delve into the factors contributing to hallucinations. Subsequently, we present a comprehensive overview of hallucination detection methods and benchmarks. Additionally, representative approaches designed to mitigate hallucinations are introduced accordingly. Finally, we analyze the challenges that highlight the current limitations and formulate open questions, aiming to delineate pathways for future research on hallucinations in LLMs.
小貼士
登錄享
相關主題
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191