As the breadth and depth of language model applications continue to expand rapidly, it is increasingly important to build efficient frameworks for measuring and mitigating the learned or inherited social biases of these models. In this paper, we present our work on evaluating instruction fine-tuned language models' ability to identify bias through zero-shot prompting, including Chain-of-Thought (CoT) prompts. Across LLaMA and its two instruction fine-tuned versions, Alpaca 7B performs best on the bias identification task with an accuracy of 56.7%. We also demonstrate that scaling up LLM size and data diversity could lead to further performance gain. This is a work-in-progress presenting the first component of our bias mitigation framework. We will keep updating this work as we get more results.
相關內容
With the development of blockchain technology, smart contracts have become an important component of blockchain applications. Despite their crucial role, the development of smart contracts may introduce vulnerabilities and potentially lead to severe consequences, such as financial losses. Meanwhile, large language models, represented by ChatGPT, have gained great attentions, showcasing great capabilities in code analysis tasks. In this paper, we presented an empirical study to investigate the performance of ChatGPT in identifying smart contract vulnerabilities. Initially, we evaluated ChatGPT's effectiveness using a publicly available smart contract dataset. Our findings discover that while ChatGPT achieves a high recall rate, its precision in pinpointing smart contract vulnerabilities is limited. Furthermore, ChatGPT's performance varies when detecting different vulnerability types. We delved into the root causes for the false positives generated by ChatGPT, and categorized them into four groups. Second, by comparing ChatGPT with other state-of-the-art smart contract vulnerability detection tools, we found that ChatGPT's F-score is lower than others for 3 out of the 7 vulnerabilities. In the case of the remaining 4 vulnerabilities, ChatGPT exhibits a slight advantage over these tools. Finally, we analyzed the limitation of ChatGPT in smart contract vulnerability detection, revealing that the robustness of ChatGPT in this field needs to be improved from two aspects: its uncertainty in answering questions; and the limited length of the detected code. In general, our research provides insights into the strengths and weaknesses of employing large language models, specifically ChatGPT, for the detection of smart contract vulnerabilities.
Recent publications suggest using natural language analysis on database schema elements to guide tuning and profiling efforts. The underlying hypothesis is that state-of-the-art language processing methods, so-called language models, are able to extract information on data properties from schema text. This paper examines that hypothesis in the context of data correlation analysis: is it possible to find column pairs with correlated data by analyzing their names via language models? First, the paper introduces a novel benchmark for data correlation analysis, created by analyzing thousands of Kaggle data sets (and available for download). Second, it uses that data to study the ability of language models to predict correlation, based on column names. The analysis covers different language models, various correlation metrics, and a multitude of accuracy metrics. It pinpoints factors that contribute to successful predictions, such as the length of column names as well as the ratio of words. Finally, \rev{the study analyzes the impact of column types on prediction performance.} The results show that schema text can be a useful source of information and inform future research efforts, targeted at NLP-enhanced database tuning and data profiling.
Recent times have witnessed an increasing number of applications of deep neural networks towards solving tasks that require superior cognitive abilities, e.g., playing Go, generating art, ChatGPT, etc. Such a dramatic progress raises the question: how generalizable are neural networks in solving problems that demand broad skills? To answer this question, we propose SMART: a Simple Multimodal Algorithmic Reasoning Task and the associated SMART-101 dataset, for evaluating the abstraction, deduction, and generalization abilities of neural networks in solving visuo-linguistic puzzles designed specifically for children in the 6--8 age group. Our dataset consists of 101 unique puzzles; each puzzle comprises a picture and a question, and their solution needs a mix of several elementary skills, including arithmetic, algebra, and spatial reasoning, among others. To scale our dataset towards training deep neural networks, we programmatically generate entirely new instances for each puzzle, while retaining their solution algorithm. To benchmark performances on SMART-101, we propose a vision and language meta-learning model using varied state-of-the-art backbones. Our experiments reveal that while powerful deep models offer reasonable performances on puzzles in a supervised setting, they are not better than random accuracy when analyzed for generalization. We also evaluate the recent ChatGPT and other large language models on a subset of SMART-101 and find that while these models show convincing reasoning abilities, the answers are often incorrect.
Large language models (LLMs) have led to a surge in collaborative writing with model assistance. As different users incorporate suggestions from the same model, there is a risk of decreased diversity in the produced content, potentially limiting diverse perspectives in public discourse. In this work, we measure the impact of co-writing on diversity via a controlled experiment, where users write argumentative essays in three setups -- using a base LLM (GPT3), a feedback-tuned LLM (InstructGPT), and writing without model help. We develop a set of diversity metrics and find that writing with InstructGPT (but not the GPT3) results in a statistically significant reduction in diversity. Specifically, it increases the similarity between the writings of different authors and reduces the overall lexical and content diversity. We additionally find that this effect is mainly attributable to InstructGPT contributing less diverse text to co-written essays. In contrast, the user-contributed text remains unaffected by model collaboration. This suggests that the recent improvement in generation quality from adapting models to human feedback might come at the cost of more homogeneous and less diverse content.
Short-packet communications are applied to various scenarios where transmission covertness and reliability are crucial due to the open wireless medium and finite blocklength. Although intelligent reflection surface (IRS) has been widely utilized to enhance transmission covertness and reliability, the question of how many reflection elements at IRS are required remains unanswered, which is vital to system design and practical deployment. The inherent strong coupling exists between the transmission covertness and reliability by IRS, leading to the question of intractability. To address this issue, the detection error probability at the warder and its approximation are derived first to reveal the relation between covertness performance and the number of reflection elements. Besides, to evaluate the reliability performance of the system, the decoding error probability at the receiver is also derived. Subsequently, the asymptotic reliability performance in high covertness regimes is investigated, which provides theoretical predictions about the number of reflection elements at IRS required to achieve a decoding error probability close to 0 with given covertness requirements. Furthermore, Monte-Carlo simulations verify the accuracy of the derived results for detection (decoding) error probabilities and the validity of the theoretical predictions for reflection elements. Moreover, results show that more reflection elements are required to achieve high reliability with tighter covertness requirements, longer blocklength and higher transmission rates.
We propose a novel unsupervised object localization method that allows us to explain the predictions of the model by utilizing self-supervised pre-trained models without additional finetuning. Existing unsupervised and self-supervised object localization methods often utilize class-agnostic activation maps or self-similarity maps of a pre-trained model. Although these maps can offer valuable information for localization, their limited ability to explain how the model makes predictions remains challenging. In this paper, we propose a simple yet effective unsupervised object localization method based on representer point selection, where the predictions of the model can be represented as a linear combination of representer values of training points. By selecting representer points, which are the most important examples for the model predictions, our model can provide insights into how the model predicts the foreground object by providing relevant examples as well as their importance. Our method outperforms the state-of-the-art unsupervised and self-supervised object localization methods on various datasets with significant margins and even outperforms recent weakly supervised and few-shot methods.
Non-convex optimization is ubiquitous in modern machine learning. Researchers devise non-convex objective functions and optimize them using off-the-shelf optimizers such as stochastic gradient descent and its variants, which leverage the local geometry and update iteratively. Even though solving non-convex functions is NP-hard in the worst case, the optimization quality in practice is often not an issue -- optimizers are largely believed to find approximate global minima. Researchers hypothesize a unified explanation for this intriguing phenomenon: most of the local minima of the practically-used objectives are approximately global minima. We rigorously formalize it for concrete instances of machine learning problems.
Compared with cheap addition operation, multiplication operation is of much higher computation complexity. The widely-used convolutions in deep neural networks are exactly cross-correlation to measure the similarity between input feature and convolution filters, which involves massive multiplications between float values. In this paper, we present adder networks (AdderNets) to trade these massive multiplications in deep neural networks, especially convolutional neural networks (CNNs), for much cheaper additions to reduce computation costs. In AdderNets, we take the $\ell_1$-norm distance between filters and input feature as the output response. The influence of this new similarity measure on the optimization of neural network have been thoroughly analyzed. To achieve a better performance, we develop a special back-propagation approach for AdderNets by investigating the full-precision gradient. We then propose an adaptive learning rate strategy to enhance the training procedure of AdderNets according to the magnitude of each neuron's gradient. As a result, the proposed AdderNets can achieve 74.9% Top-1 accuracy 91.7% Top-5 accuracy using ResNet-50 on the ImageNet dataset without any multiplication in convolution layer.
Language model pre-training has proven to be useful in learning universal language representations. As a state-of-the-art language model pre-training model, BERT (Bidirectional Encoder Representations from Transformers) has achieved amazing results in many language understanding tasks. In this paper, we conduct exhaustive experiments to investigate different fine-tuning methods of BERT on text classification task and provide a general solution for BERT fine-tuning. Finally, the proposed solution obtains new state-of-the-art results on eight widely-studied text classification datasets.
We propose a new method for event extraction (EE) task based on an imitation learning framework, specifically, inverse reinforcement learning (IRL) via generative adversarial network (GAN). The GAN estimates proper rewards according to the difference between the actions committed by the expert (or ground truth) and the agent among complicated states in the environment. EE task benefits from these dynamic rewards because instances and labels yield to various extents of difficulty and the gains are expected to be diverse -- e.g., an ambiguous but correctly detected trigger or argument should receive high gains -- while the traditional RL models usually neglect such differences and pay equal attention on all instances. Moreover, our experiments also demonstrate that the proposed framework outperforms state-of-the-art methods, without explicit feature engineering.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191