SafeInfer: Context Adaptive Decoding Time Safety Alignment for Large Language Models

Safety-aligned language models often exhibit fragile and imbalanced safety mechanisms, increasing the likelihood of generating unsafe content. In addition, incorporating new knowledge through editing techniques to language models can further compromise safety. To address these issues, we propose SafeInfer, a context-adaptive, decoding-time safety alignment strategy for generating safe responses to user queries. SafeInfer comprises two phases: the safety amplification phase, which employs safe demonstration examples to adjust the model's hidden states and increase the likelihood of safer outputs, and the safety-guided decoding phase, which influences token selection based on safety-optimized distributions, ensuring the generated content complies with ethical guidelines. Further, we present HarmEval, a novel benchmark for extensive safety evaluations, designed to address potential misuse scenarios in accordance with the policies of leading AI tech giants.

DiTTO: A Feature Representation Imitation Approach for Improving Cross-Lingual Transfer

Zero-shot cross-lingual transfer is promising, however has been shown to be sub-optimal, with inferior transfer performance across low-resource languages. In this work, we envision languages as domains for improving zero-shot transfer by jointly reducing the feature incongruity between the source and the target language and increasing the generalization capabilities of pre-trained multilingual transformers. We show that our approach, DiTTO, significantly outperforms the standard zero-shot fine-tuning method on multiple datasets across all languages using solely unlabeled instances in the target language. Empirical results show that jointly reducing feature incongruity for multiple target languages is vital for successful cross-lingual transfer. Moreover, our model enables better cross-lingual transfer than standard fine-tuning methods, even in the few-shot setting.

"Diversity and Uncertainty in Moderation" are the Key to Data Selection for Multilingual Few-shot Transfer

Few-shot transfer often shows substantial gain over zero-shot transfer~\cite{lauscher2020zero}, which is a practically useful trade-off between fully supervised and unsupervised learning approaches for multilingual pretrained model-based systems. This paper explores various strategies for selecting data for annotation that can result in a better few-shot transfer. The proposed approaches rely on multiple measures such as data entropy using $n$-gram language model, predictive entropy, and gradient embedding. We propose a loss embedding method for sequence labeling tasks, which induces diversity and uncertainty sampling similar to gradient embedding. The proposed data selection strategies are evaluated and compared for POS tagging, NER, and NLI tasks for up to 20 languages. Our experiments show that the gradient and loss embedding-based strategies consistently outperform random data selection baselines, with gains varying with the initial performance of the zero-shot transfer. Furthermore, the proposed method shows similar trends in improvement even when the model is fine-tuned using a lower proportion of the original task-specific labeled training data for zero-shot transfer.

Multi Task Learning For Zero Shot Performance Prediction of Multilingual Models

Massively Multilingual Transformer based Language Models have been observed to be surprisingly effective on zero-shot transfer across languages, though the performance varies from language to language depending on the pivot language(s) used for fine-tuning. In this work, we build upon some of the existing techniques for predicting the zero-shot performance on a task, by modeling it as a multi-task learning problem. We jointly train predictive models for different tasks which helps us build more accurate predictors for tasks where we have test data in very few languages to measure the actual performance of the model. Our approach also lends us the ability to perform a much more robust feature selection and identify a common set of features that influence zero-shot performance across a variety of tasks.

Mitigating Uncertainty of Classifier for Unsupervised Domain Adaptation

Understanding unsupervised domain adaptation has been an important task that has been well explored. However, the wide variety of methods have not analyzed the role of a classifier's performance in detail. In this paper, we thoroughly examine the role of a classifier in terms of matching source and target distributions. We specifically investigate the classifier ability by matching a) the distribution of features, b) probabilistic uncertainty for samples and c) certainty activation mappings. Our analysis suggests that using these three distributions does result in a consistently improved performance on all the datasets. Our work thus extends present knowledge on the role of the various distributions obtained from the classifier towards solving unsupervised domain adaptation.

Attending to Discriminative Certainty for Domain Adaptation

In this paper, we aim to solve for unsupervised domain adaptation of classifiers where we have access to label information for the source domain while these are not available for a target domain. While various methods have been proposed for solving these including adversarial discriminator based methods, most approaches have focused on the entire image based domain adaptation. In an image, there would be regions that can be adapted better, for instance, the foreground object may be similar in nature. To obtain such regions, we propose methods that consider the probabilistic certainty estimate of various regions and specify focus on these during classification for adaptation. We observe that just by incorporating the probabilistic certainty of the discriminator while training the classifier, we are able to obtain state of the art results on various datasets as compared against all the recent methods. We provide a thorough empirical analysis of the method by providing ablation analysis, statistical significance test, and visualization of the attention maps and t-SNE embeddings. These evaluations convincingly demonstrate the effectiveness of the proposed approach.

Adversarial Adaptation of Scene Graph Models for Understanding Civic Issues

Citizen engagement and technology usage are two emerging trends driven by smart city initiatives. Governments around the world are adopting technology for faster resolution of civic issues. Typically, citizens report issues, such as broken roads, garbage dumps, etc. through web portals and mobile apps, in order for the government authorities to take appropriate actions. Several mediums -- text, image, audio, video -- are used to report these issues. Through a user study with 13 citizens and 3 authorities, we found that image is the most preferred medium to report civic issues. However, analyzing civic issue related images is challenging for the authorities as it requires manual effort. Moreover, previous works have been limited to identifying a specific set of issues from images. In this work, given an image, we propose to generate a Civic Issue Graph consisting of a set of objects and the semantic relations between them, which are representative of the underlying civic issue. We also release two multi-modal (text and images) datasets, that can help in further analysis of civic issues from images. We present a novel approach for adversarial training of existing scene graph models that enables the use of scene graphs for new applications in the absence of any labelled training data. We conduct several experiments to analyze the efficacy of our approach, and using human evaluation, we establish the appropriateness of our model at representing different civic issues.