In many text-generation problems, users may prefer not only a single response, but a diverse range of high-quality outputs from which to choose. Quality-diversity (QD) search algorithms aim at such outcomes, by continually improving and diversifying a population of candidates. However, the applicability of QD to qualitative domains, like creative writing, has been limited by the difficulty of algorithmically specifying measures of quality and diversity. Interestingly, recent developments in language models (LMs) have enabled guiding search through AI feedback, wherein LMs are prompted in natural language to evaluate qualitative aspects of text. Leveraging this development, we introduce Quality-Diversity through AI Feedback (QDAIF), wherein an evolutionary algorithm applies LMs to both generate variation and evaluate the quality and diversity of candidate text. When assessed on creative writing domains, QDAIF covers more of a specified search space with high-quality samples than do non-QD controls. Further, human evaluation of QDAIF-generated creative texts validates reasonable agreement between AI and human evaluation. Our results thus highlight the potential of AI feedback to guide open-ended search for creative and original solutions, providing a recipe that seemingly generalizes to many domains and modalities. In this way, QDAIF is a step towards AI systems that can independently search, diversify, evaluate, and improve, which are among the core skills underlying human society's capacity for innovation.

The rapid development of Large Language Models (LLMs) has substantially expanded the range of tasks they can address. In the field of Natural Language Processing (NLP), researchers have shifted their focus from conventional NLP tasks (e.g., sequence tagging and parsing) towards tasks that revolve around aligning with human needs (e.g., brainstorming and email writing). This shift in task distribution imposes new requirements on evaluating these aligned models regarding generality (i.e., assessing performance across diverse scenarios), flexibility (i.e., examining under different protocols), and interpretability (i.e., scrutinizing models with explanations). In this paper, we propose a generative judge with 13B parameters, Auto-J, designed to address these challenges. Our model is trained on user queries and LLM-generated responses under massive real-world scenarios and accommodates diverse evaluation protocols (e.g., pairwise response comparison and single-response evaluation) with well-structured natural language critiques. To demonstrate the efficacy of our approach, we construct a new testbed covering 58 different scenarios. Experimentally, Auto-J outperforms a series of strong competitors, including both open-source and closed-source models, by a large margin. We also provide detailed analysis and case studies to further reveal the potential of our method and make a variety of resources public at https://github.com/GAIR-NLP/auto-j.

In this work, we instantiate a novel perturbation-based multi-class explanation framework, LIPEx (Locally Interpretable Probabilistic Explanation). We demonstrate that LIPEx not only locally replicates the probability distributions output by the widely used complex classification models but also provides insight into how every feature deemed to be important affects the prediction probability for each of the possible classes. We achieve this by defining the explanation as a matrix obtained via regression with respect to the Hellinger distance in the space of probability distributions. Ablation tests on text and image data, show that LIPEx-guided removal of important features from the data causes more change in predictions for the underlying model than similar tests based on other saliency-based or feature importance-based Explainable AI (XAI) methods. It is also shown that compared to LIME, LIPEx is more data efficient in terms of using a lesser number of perturbations of the data to obtain a reliable explanation. This data-efficiency is seen to manifest as LIPEx being able to compute its explanation matrix around 53% faster than all-class LIME, for classification experiments with text data.

Large Language Models (LLMs) presents significant priority in text understanding and generation. However, LLMs suffer from the risk of generating harmful contents especially while being employed to applications. There are several black-box attack methods, such as Prompt Attack, which can change the behaviour of LLMs and induce LLMs to generate unexpected answers with harmful contents. Researchers are interested in Prompt Attack and Defense with LLMs, while there is no publicly available dataset with high successful attacking rate to evaluate the abilities of defending prompt attack. In this paper, we introduce a pipeline to construct high-quality prompt attack samples, along with a Chinese prompt attack dataset called CPAD. Our prompts aim to induce LLMs to generate unexpected outputs with several carefully designed prompt attack templates and widely concerned attacking contents. Different from previous datasets involving safety estimation, we construct the prompts considering three dimensions: contents, attacking methods and goals. Especially, the attacking goals indicate the behaviour expected after successfully attacking the LLMs, thus the responses can be easily evaluated and analysed. We run several popular Chinese LLMs on our dataset, and the results show that our prompts are significantly harmful to LLMs, with around 70% attack success rate to GPT-3.5. CPAD is publicly available at https://github.com/liuchengyuan123/CPAD.

Diffusion models and large language models have emerged as leading-edge generative models, revolutionizing various aspects of human life. However, the practical implementations of these models have also exposed inherent risks, bringing to the forefront their evil sides and sparking concerns regarding their trustworthiness. Despite the wealth of literature on this subject, a comprehensive survey specifically delving into the intersection of large-scale generative models and their trustworthiness remains largely absent. To bridge this gap, this paper investigates both the long-standing and emerging threats associated with these models across four fundamental dimensions: 1) privacy, 2) security, 3) fairness, and 4) responsibility. Based on the investigation results, we develop an extensive map outlining the trustworthiness of large generative models. After that, we provide practical recommendations and potential research directions for future secure applications equipped with large generative models, ultimately promoting the trustworthiness of the models and benefiting the society as a whole.

Large Language Models (LLMs) are capable of successfully performing many language processing tasks zero-shot (without training data). If zero-shot LLMs can also reliably classify and explain social phenomena like persuasiveness and political ideology, then LLMs could augment the Computational Social Science (CSS) pipeline in important ways. This work provides a road map for using LLMs as CSS tools. Towards this end, we contribute a set of prompting best practices and an extensive evaluation pipeline to measure the zero-shot performance of 13 language models on 25 representative English CSS benchmarks. On taxonomic labeling tasks (classification), LLMs fail to outperform the best fine-tuned models but still achieve fair levels of agreement with humans. On free-form coding tasks (generation), LLMs produce explanations that often exceed the quality of crowdworkers' gold references. We conclude that the performance of today's LLMs can augment the CSS research pipeline in two ways: (1) serving as zero-shot data annotators on human annotation teams, and (2) bootstrapping challenging creative generation tasks (e.g., explaining the underlying attributes of a text). In summary, LLMs are posed to meaningfully participate in} social science analysis in partnership with humans.

This document serves as an overview of the different mechanisms and areas of governance in the BigCode project. It aims to support transparency by providing relevant information about choices that were made during the project to the broader public, and to serve as an example of intentional governance of an open research project that future endeavors can leverage to shape their own approach. The first section, Project Structure, covers the project organization, its stated goals and values, its internal decision processes, and its funding and resources. The second section, Data and Model Governance, covers decisions relating to the questions of data subject consent, privacy, and model release.

As language models (LMs) advance, interest is growing in applying them to high-stakes societal decisions, such as determining financing or housing eligibility. However, their potential for discrimination in such contexts raises ethical concerns, motivating the need for better methods to evaluate these risks. We present a method for proactively evaluating the potential discriminatory impact of LMs in a wide range of use cases, including hypothetical use cases where they have not yet been deployed. Specifically, we use an LM to generate a wide array of potential prompts that decision-makers may input into an LM, spanning 70 diverse decision scenarios across society, and systematically vary the demographic information in each prompt. Applying this methodology reveals patterns of both positive and negative discrimination in the Claude 2.0 model in select settings when no interventions are applied. While we do not endorse or permit the use of language models to make automated decisions for the high-risk use cases we study, we demonstrate techniques to significantly decrease both positive and negative discrimination through careful prompt engineering, providing pathways toward safer deployment in use cases where they may be appropriate. Our work enables developers and policymakers to anticipate, measure, and address discrimination as language model capabilities and applications continue to expand. We release our dataset and prompts at https://huggingface.co/datasets/Anthropic/discrim-eval

We introduce Hyperbard, a dataset of diverse relational data representations derived from Shakespeare's plays. Our representations range from simple graphs capturing character co-occurrence in single scenes to hypergraphs encoding complex communication settings and character contributions as hyperedges with edge-specific node weights. By making multiple intuitive representations readily available for experimentation, we facilitate rigorous representation robustness checks in graph learning, graph mining, and network analysis, highlighting the advantages and drawbacks of specific representations. Leveraging the data released in Hyperbard, we demonstrate that many solutions to popular graph mining problems are highly dependent on the representation choice, thus calling current graph curation practices into question. As an homage to our data source, and asserting that science can also be art, we present all our points in the form of a play.

In policy research, one of the most critical analytic tasks is to estimate the causal effect of a policy-relevant shift to the distribution of a continuous exposure/treatment on an outcome of interest. We call this problem shift-response function (SRF) estimation. Existing neural network methods involving robust causal-effect estimators lack theoretical guarantees and practical implementations for SRF estimation. Motivated by a key policy-relevant question in public health, we develop a neural network method and its theoretical underpinnings to estimate SRFs with robustness and efficiency guarantees. We then apply our method to data consisting of 68 million individuals and 27 million deaths across the U.S. to estimate the causal effect from revising the US National Ambient Air Quality Standards (NAAQS) for PM 2.5 from 12 $\mu g/m^3$ to 9 $\mu g/m^3$. This change has been recently proposed by the US Environmental Protection Agency (EPA). Our goal is to estimate, for the first time, the reduction in deaths that would result from this anticipated revision using causal methods for SRFs. Our proposed method, called {T}argeted {R}egularization for {E}xposure {S}hifts with Neural {Net}works (TRESNET), contributes to the neural network literature for causal inference in two ways: first, it proposes a targeted regularization loss with theoretical properties that ensure double robustness and achieves asymptotic efficiency specific for SRF estimation; second, it enables loss functions from the exponential family of distributions to accommodate non-continuous outcome distributions (such as hospitalization or mortality counts). We complement our application with benchmark experiments that demonstrate TRESNET's broad applicability and competitiveness.