AI regulations are expected to prohibit machine learning models from using sensitive attributes during training. However, the latest Natural Language Processing (NLP) classifiers, which rely on deep learning, operate as black-box systems, complicating the detection and remediation of such misuse. Traditional bias mitigation methods in NLP aim for comparable performance across different groups based on attributes like gender or race but fail to address the underlying issue of reliance on protected attributes. To partly fix that, we introduce NLPGuard, a framework for mitigating the reliance on protected attributes in NLP classifiers. NLPGuard takes an unlabeled dataset, an existing NLP classifier, and its training data as input, producing a modified training dataset that significantly reduces dependence on protected attributes without compromising accuracy. NLPGuard is applied to three classification tasks: identifying toxic language, sentiment analysis, and occupation classification. Our evaluation shows that current NLP classifiers heavily depend on protected attributes, with up to $23\%$ of the most predictive words associated with these attributes. However, NLPGuard effectively reduces this reliance by up to $79\%$, while slightly improving accuracy.

Large Language Models (LLMs) trained on extensive corpora inevitably retain sensitive data, such as personal privacy information and copyrighted material. Recent advancements in knowledge unlearning involve updating LLM parameters to erase specific knowledge. However, current unlearning paradigms are mired in vague forgetting boundaries, often erasing knowledge indiscriminately. In this work, we introduce KnowUnDo, a benchmark containing copyrighted content and user privacy domains to evaluate if the unlearning process inadvertently erases essential knowledge. Our findings indicate that existing unlearning methods often suffer from excessive unlearning. To address this, we propose a simple yet effective method, MemFlex, which utilizes gradient information to precisely target and unlearn sensitive parameters. Experimental results show that MemFlex is superior to existing methods in both precise knowledge unlearning and general knowledge retaining of LLMs. Code and dataset will be released at https://github.com/zjunlp/KnowUnDo.

This paper summarizes Team SCaLAR's work on SemEval-2024 Task 5: Legal Argument Reasoning in Civil Procedure. To address this Binary Classification task, which was daunting due to the complexity of the Legal Texts involved, we propose a simple yet novel similarity and distance-based unsupervised approach to generate labels. Further, we explore the Multi-level fusion of Legal-Bert embeddings using ensemble features, including CNN, GRU, and LSTM. To address the lengthy nature of Legal explanation in the dataset, we introduce T5-based segment-wise summarization, which successfully retained crucial information, enhancing the model's performance. Our unsupervised system witnessed a 20-point increase in macro F1-score on the development set and a 10-point increase on the test set, which is promising given its uncomplicated architecture.

Existing efforts in safeguarding LLMs are limited in actively exposing the vulnerabilities of the target LLM and readily adapting to newly emerging safety risks. To address this, we present Purple-teaming LLMs with Adversarial Defender training (PAD), a pipeline designed to safeguard LLMs by novelly incorporating the red-teaming (attack) and blue-teaming (safety training) techniques. In PAD, we automatically collect conversational data that cover the vulnerabilities of an LLM around specific safety risks in a self-play manner, where the attacker aims to elicit unsafe responses and the defender generates safe responses to these attacks. We then update both modules in a generative adversarial network style by training the attacker to elicit more unsafe responses and updating the defender to identify them and explain the unsafe reason. Experimental results demonstrate that PAD significantly outperforms existing baselines in both finding effective attacks and establishing a robust safe guardrail. Furthermore, our findings indicate that PAD excels in striking a balance between safety and overall model quality. We also reveal key challenges in safeguarding LLMs, including defending multi-turn attacks and the need for more delicate strategies to identify specific risks.

Data-driven software solutions have significantly been used in critical domains with significant socio-economic, legal, and ethical implications. The rapid adoptions of data-driven solutions, however, pose major threats to the trustworthiness of automated decision-support software. A diminished understanding of the solution by the developer and historical/current biases in the data sets are primary challenges. To aid data-driven software developers and end-users, we present \toolname, a debugging tool to test and explain the fairness implications of data-driven solutions. \toolname visualizes the logic of datasets, trained models, and decisions for a given data point. In addition, it trains various models with varying fairness-accuracy trade-offs. Crucially, \toolname incorporates counterfactual fairness testing that finds bugs beyond the development datasets. We conducted two studies through \toolname that allowed us to measure false positives/negatives in prevalent counterfactual testing and understand the human perception of counterfactual test cases in a class survey. \toolname and its benchmarks are publicly available at~\url{https://github.com/Pennswood/FairLay-ML}. The live version of the tool is available at~\url{https://fairlayml-v2.streamlit.app/}. We provide a video demo of the tool at https://youtu.be/wNI9UWkywVU?t=127

Forensic authorship profiling uses linguistic markers to infer characteristics about an author of a text. This task is paralleled in dialect classification, where a prediction is made about the linguistic variety of a text based on the text itself. While there have been significant advances in recent years in variety classification, forensic linguistics rarely relies on these approaches due to their lack of transparency, among other reasons. In this paper we therefore explore the explainability of machine learning approaches considering the forensic context. We focus on variety classification as a means of geolinguistic profiling of unknown texts based on social media data from the German-speaking area. For this, we identify the lexical items that are the most impactful for the variety classification. We find that the extracted lexical features are indeed representative of their respective varieties and note that the trained models also rely on place names for classifications.

Large Language Models (LLMs) are increasingly adopted in financial analysis for interpreting complex market data and trends. However, their use is challenged by intrinsic biases (e.g., risk-preference bias) and a superficial understanding of market intricacies, necessitating a thorough assessment of their financial insight. To address these issues, we introduce Financial Bias Indicators (FBI), a framework with components like Bias Unveiler, Bias Detective, Bias Tracker, and Bias Antidote to identify, detect, analyze, and eliminate irrational biases in LLMs. By combining behavioral finance principles with bias examination, we evaluate 23 leading LLMs and propose a de-biasing method based on financial causal knowledge. Results show varying degrees of financial irrationality among models, influenced by their design and training. Models trained specifically on financial datasets may exhibit more irrationality, and even larger financial language models (FinLLMs) can show more bias than smaller, general models. We utilize four prompt-based methods incorporating causal debiasing, effectively reducing financial biases in these models. This work enhances the understanding of LLMs' bias in financial applications, laying the foundation for developing more reliable and rational financial analysis tools.

Large Language Models (LLMs) have significantly advanced various NLP tasks. However, these models often risk generating unsafe text that perpetuates biases. Current approaches to produce unbiased outputs from LLMs can reduce biases but at the expense of knowledge retention. In this research, we address the question of whether producing safe (unbiased) outputs through LLMs can retain knowledge and language understanding. In response, we developed the Safety and Responsible Large Language Model (\textbf{SR}$_{\text{LLM}}$), an LLM that has been instruction fine-tuned on top of already safe LLMs (e.g., Llama2 or related) to diminish biases in generated text. To achieve our goals, we compiled a specialized dataset designed to train our model in identifying and correcting biased text. We conduct experiments, both on this custom data and out-of-distribution test sets, to show the bias reduction and knowledge retention. The results confirm that \textbf{SR}$_{\text{LLM}}$ outperforms traditional fine-tuning and prompting methods in both reducing biases and preserving the integrity of language knowledge. The significance of our findings lies in demonstrating that instruction fine-tuning can provide a more robust solution for bias reduction in LLMs. We have made our code and data available at \href{https://github.com/shainarazavi/Safe-Responsible-LLM}{Safe-LLM}.

This paper investigates the relationships between hyperparameters of machine learning and fairness. Data-driven solutions are increasingly used in critical socio-technical applications where ensuring fairness is important. Rather than explicitly encoding decision logic via control and data structures, the ML developers provide input data, perform some pre-processing, choose ML algorithms, and tune hyperparameters (HPs) to infer a program that encodes the decision logic. Prior works report that the selection of HPs can significantly influence fairness. However, tuning HPs to find an ideal trade-off between accuracy, precision, and fairness has remained an expensive and tedious task. Can we predict fairness of HP configuration for a given dataset? Are the predictions robust to distribution shifts? We focus on group fairness notions and investigate the HP space of 5 training algorithms. We first find that tree regressors and XGBoots significantly outperformed deep neural networks and support vector machines in accurately predicting the fairness of HPs. When predicting the fairness of ML hyperparameters under temporal distribution shift, the tree regressors outperforms the other algorithms with reasonable accuracy. However, the precision depends on the ML training algorithm, dataset, and protected attributes. For example, the tree regressor model was robust for training data shift from 2014 to 2018 on logistic regression and discriminant analysis HPs with sex as the protected attribute; but not for race and other training algorithms. Our method provides a sound framework to efficiently perform fine-tuning of ML training algorithms and understand the relationships between HPs and fairness.

Large language models (LLMs) trained on web-scale datasets raise substantial concerns regarding permissible data usage. One major question is whether these models "memorize" all their training data or they integrate many data sources in some way more akin to how a human would learn and synthesize information. The answer hinges, to a large degree, on how we define memorization. In this work, we propose the Adversarial Compression Ratio (ACR) as a metric for assessing memorization in LLMs. A given string from the training data is considered memorized if it can be elicited by a prompt (much) shorter than the string itself--in other words, if these strings can be "compressed" with the model by computing adversarial prompts of fewer tokens. The ACR overcomes the limitations of existing notions of memorization by (i) offering an adversarial view of measuring memorization, especially for monitoring unlearning and compliance; and (ii) allowing for the flexibility to measure memorization for arbitrary strings at a reasonably low compute. Our definition serves as a practical tool for determining when model owners may be violating terms around data usage, providing a potential legal tool and a critical lens through which to address such scenarios. Find the Minimal Prompt PROMPT: urgesTOBE quote!