A significant amount of research is focused on developing and evaluating large language models for a variety of code synthesis tasks. These include synthesizing code from natural language instructions, synthesizing tests from code, and synthesizing explanations of code. In contrast, the behavior of instructional code editing with LLMs is understudied. These are tasks in which the model is instructed to update a block of code provided in a prompt. The editing instruction may ask for a feature to added or removed, describe a bug and ask for a fix, ask for a different kind of solution, or many other common code editing tasks. We introduce a carefully crafted benchmark of code editing tasks and use it evaluate several cutting edge LLMs. Our evaluation exposes a significant gap between the capabilities of state-of-the-art open and closed models. For example, even GPT-3.5-Turbo is 8.8% better than the best open model at editing code. We also introduce a new, carefully curated, permissively licensed training set of code edits coupled with natural language instructions. Using this training set, we show that we can fine-tune open Code LLMs to significantly improve their code editing capabilities.
相關內容
RF data-driven device fingerprinting through the use of deep learning has recently surfaced as a possible method for enabling secure device identification and authentication. Traditional approaches are commonly susceptible to the domain adaptation problem where a model trained on data collected under one domain performs badly when tested on data collected under a different domain. Some examples of a domain change include varying the location or environment of the device and varying the time or day of the data collection. In this work, we propose using multifractal analysis and the variance fractal dimension trajectory (VFDT) as a data representation input to the deep neural network to extract device fingerprints that are domain generalizable. We analyze the effectiveness of the proposed VFDT representation in detecting device-specific signatures from hardware-impaired IQ (in-phase and quadrature) signals, and we evaluate its robustness in real-world settings, using an experimental testbed of 30 WiFi-enabled Pycom devices. Our experimental results show that the proposed VFDT representation improves the scalability, robustness and generalizability of the deep learning models significantly compared to when using IQ data samples.
Recently, the potential of large language models (LLMs) has been widely used in assisting programming. However, current research does not explore the artist potential of LLMs in creative coding within artist and AI collaboration. Our work probes the reflection type of artists in the creation process with such collaboration. We compare two common collaboration approaches: invoking the entire program and multiple subtasks. Our findings exhibit artists' different stimulated reflections in two different methods. Our finding also shows the correlation of reflection type with user performance, user satisfaction, and subjective experience in two collaborations through conducting two methods, including experimental data and qualitative interviews. In this sense, our work reveals the artistic potential of LLM in creative coding. Meanwhile, we provide a critical lens of human-AI collaboration from the artists' perspective and expound design suggestions for future work of AI-assisted creative tasks.
The remarkable instruction-following capability of large language models (LLMs) has sparked a growing interest in automatically learning suitable prompts. However, while many effective methods have been proposed, the cost incurred during the learning process (e.g., accessing LLM and evaluating the responses) has not been considered. To overcome this limitation, this work explicitly incorporates a finite budget constraint into prompt learning. Towards developing principled solutions, a novel connection is established between prompt learning and fixed-budget best arm identification (BAI-FB) in multi-armed bandits (MAB). Based on this connection, a general framework TRIPLE (besT aRm Identification for Prompt LEarning) is proposed to harness the power of BAI-FB in prompt learning systematically. Unique characteristics of prompt learning further lead to two embedding-based enhancements of TRIPLE by exploiting the ideas of clustering and function approximation. Extensive experiments on multiple well-adopted tasks using both GPT 3.5 and Llama2 demonstrate the significant performance improvement of TRIPLE over the previous baselines while satisfying the limited budget constraints.
The problem of repairing inconsistent knowledge bases has a long history within the communities of database theory and knowledge representation and reasoning, especially from the perspective of structured data. However, as the data available in real-world domains becomes more complex and interconnected, the need naturally arises for developing new types of repositories, representation languages, and semantics, to allow for more suitable ways to query and reason about it. Graph databases provide an effective way to represent relationships among semi-structured data, and allow processing and querying these connections efficiently. In this work, we focus on the problem of computing prioritized repairs over graph databases with data values, using a notion of consistency based on Reg-GXPath expressions as integrity constraints. We present several preference criteria based on the standard subset repair semantics, incorporating weights, multisets, and set-based priority levels. We study the most common repairing tasks, showing that it is possible to maintain the same computational complexity as in the case where no preference criterion is available for exploitation. To complete the picture, we explore the complexity of consistent query answering in this setting and obtain tight lower and upper bounds for all the preference criteria introduced.
Prior studies on the emergence in large models have primarily focused on how the functional capabilities of large language models (LLMs) scale with model size. Our research, however, transcends this traditional paradigm, aiming to deepen our understanding of the emergence within LLMs by placing a special emphasis not just on the model size but more significantly on the complex behavior of neuron interactions during the training process. By introducing the concepts of "self-organization" and "multifractal analysis," we explore how neuron interactions dynamically evolve during training, leading to "emergence," mirroring the phenomenon in natural systems where simple micro-level interactions give rise to complex macro-level behaviors. To quantitatively analyze the continuously evolving interactions among neurons in large models during training, we propose the Neuron-based Multifractal Analysis (NeuroMFA). Utilizing NeuroMFA, we conduct a comprehensive examination of the emergent behavior in LLMs through the lens of both model size and training process, paving new avenues for research into the emergence in large models.
Large language models (LLMs) have performed well on several reasoning benchmarks, including ones that test analogical reasoning abilities. However, it has been debated whether they are actually performing humanlike abstract reasoning or instead employing less general processes that rely on similarity to what has been seen in their training data. Here we investigate the generality of analogy-making abilities previously claimed for LLMs (Webb, Holyoak, & Lu, 2023). We take one set of analogy problems used to evaluate LLMs and create a set of "counterfactual" variants-versions that test the same abstract reasoning abilities but that are likely dissimilar from any pre-training data. We test humans and three GPT models on both the original and counterfactual problems, and show that, while the performance of humans remains high for all the problems, the GPT models' performance declines sharply on the counterfactual set. This work provides evidence that, despite previously reported successes of LLMs on analogical reasoning, these models lack the robustness and generality of human analogy-making.
Transformer-based models have dominated natural language processing and other areas in the last few years due to their superior (zero-shot) performance on benchmark datasets. However, these models are poorly understood due to their complexity and size. While probing-based methods are widely used to understand specific properties, the structures of the representation space are not systematically characterized; consequently, it is unclear how such models generalize and overgeneralize to new inputs beyond datasets. In this paper, based on a new gradient descent optimization method, we are able to explore the embedding space of a commonly used vision-language model. Using the Imagenette dataset, we show that while the model achieves over 99\% zero-shot classification performance, it fails systematic evaluations completely. Using a linear approximation, we provide a framework to explain the striking differences. We have also obtained similar results using a different model to support that our results are applicable to other transformer models with continuous inputs. We also propose a robust way to detect the modified images.
To evaluate code large language models (LLMs), research has relied on a few small manually curated benchmarks, such as HumanEval and MBPP, which represent a narrow part of the real-world software domains. In this work, we introduce round-trip correctness (RTC) as an alternative evaluation method. RTC allows Code LLM evaluation on a broader spectrum of real-world software domains without the need for costly human curation. RTC rests on the idea that we can ask a model to make a prediction (e.g., describe some code using natural language), feed that prediction back (e.g., synthesize code from the predicted description), and check if this round-trip leads to code that is semantically equivalent to the original input. We show how to employ RTC to evaluate code synthesis and editing. We find that RTC strongly correlates with model performance on existing narrow-domain code synthesis benchmarks while allowing us to expand to a much broader set of domains and tasks which was not previously possible without costly human annotations.
Transfomer-based models have significantly advanced natural language processing, in particular the performance in text classification tasks. Nevertheless, these models face challenges in processing large files, primarily due to their input constraints, which are generally restricted to hundreds or thousands of tokens. Attempts to address this issue in existing models usually consist in extracting only a fraction of the essential information from lengthy inputs, while often incurring high computational costs due to their complex architectures. In this work, we address the challenge of classifying large files from the perspective of correlated multiple instance learning. We introduce LaFiCMIL, a method specifically designed for large file classification. LaFiCMIL is optimized for efficient operation on a single GPU, making it a versatile solution for binary, multi-class, and multi-label classification tasks. We conducted extensive experiments using seven diverse and comprehensive benchmark datasets to assess LaFiCMIL's effectiveness. By integrating BERT for feature extraction, LaFiCMIL demonstrates exceptional performance, setting new benchmarks across all datasets. A notable achievement of our approach is its ability to scale BERT to handle nearly 20,000 tokens while operating on a single GPU with 32GB of memory. This efficiency, coupled with its state-of-the-art performance, highlights LaFiCMIL's potential as a groundbreaking approach in the field of large file classification.
While large language models (LLMs) have demonstrated remarkable capabilities across a range of downstream tasks, a significant concern revolves around their propensity to exhibit hallucinations: LLMs occasionally generate content that diverges from the user input, contradicts previously generated context, or misaligns with established world knowledge. This phenomenon poses a substantial challenge to the reliability of LLMs in real-world scenarios. In this paper, we survey recent efforts on the detection, explanation, and mitigation of hallucination, with an emphasis on the unique challenges posed by LLMs. We present taxonomies of the LLM hallucination phenomena and evaluation benchmarks, analyze existing approaches aiming at mitigating LLM hallucination, and discuss potential directions for future research.
小貼士
登錄享
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191