Large language models (LLMs) have shown promising results in a wide array of generative NLP tasks, such as summarization and machine translation. In the context of narrative generation, however, existing models still do not capture factors that contribute to producing consistent text. For instance, it is logical that a piece of text or a story should be uniformly readable throughout and that this form of complexity should be controllable. As such, if the complexity of an input text prompt is rated first-grade reading level in the Flesch Reading Ease test, then the generated text continuing the plot should also be within this range of complexity. With this in mind, we introduce Uniform Complexity for Text Generation (UCTG), a new benchmark test which raises the challenge of making generative models observe uniform linguistic properties with respect to prompts. We experiment with over 150+ linguistically and cognitively motivated features for evaluating text complexity in humans and generative models. From our results, we find that models such as GPT-2 struggle to preserve the complexity of input prompts used in its generations, even if finetuned with professionally written texts.
论文检索
输入标题、作者或关键词，从 7,591 篇学术成果中精准定位
While Neural Machine Translation (NMT) represents the leading approach to Machine Translation (MT), the outputs of NMT models still require translation post-editing to rectify errors and enhance quality under critical settings. In this work, we formalize the task of direct translation post-editing with Large Language Models (LLMs) and explore the use of GPT-4 to automatically post-edit NMT outputs across several language pairs. Our results demonstrate that GPT-4 is adept at translation post-editing, producing meaningful and trustworthy edits to translations that help improve its general quality as well as remove different classes of major errors in translations. In particular, human evaluations on assessing edit trustworthiness show that GPT-4 exhibits a large improvement over the prior state-of-the-art LLM. Notably, we improve upon state-of-the-art performance on WMT-22 English-Chinese, English-German, Chinese-English and German-English language pairs using GPT-4 based post-editing, as evaluated by state-of-the-art MT quality metrics. However, we also show that GPT-4 could produce hallucinated edits, thereby urging caution in its use as an expert translation post-editor.
Although existing neural retrieval models reveal promising results when training data is abundant and the performance keeps improving as training data increases, collecting high-quality annotated data is prohibitively costly. To this end, we introduce a novel noisy self-training framework combined with synthetic queries, showing that neural retrievers can be improved in a self-evolution manner with no reliance on any external models. Experimental results show that our method improves consistently over existing methods on both general-domain (e.g., MS-MARCO) and out-of-domain (i.e., BEIR) retrieval benchmarks. Extra analysis on low-resource settings reveals that our method is data efficient and outperforms competitive baselines, with as little as 30% of labelled training data. Further extending the framework for reranker training demonstrates that the proposed method is general and yields additional gains on tasks of diverse domains.
Referring Expression Comprehension (ReC) is a task that involves localizing objects in images based on natural language expressions. Most ReC methods typically approach the task as a supervised learning problem. However, the need for costly annotations, such as clear image-text pairs or region-text pairs, hinders the scalability of existing approaches. In this work, we propose a novel scene graph-based framework that automatically generates high-quality pseudo region-query pairs. Our method harnesses scene graphs to capture the relationships between objects in images and generate expressions enriched with relation information. To ensure accurate mapping between visual regions and text, we introduce an external module that employs a calibration algorithm to filter out ambiguous queries. Additionally, we employ a rewriter module to enhance the diversity of our generated pseudo queries through rewriting. Extensive experiments demonstrate that our method outperforms previous pseudo-labeling methods by about 10%, 12%, and 11% on RefCOCO, RefCOCO+, and RefCOCOg, respectively. Furthermore, it surpasses the state-of-the-art unsupervised approach by more than 15% on the RefCOCO dataset.
Instance attribution (IA) aims to identify the training instances leading to the prediction of a test example, helping researchers understand the dataset better and optimize data processing. While many IA methods have been proposed recently, how to evaluate them still remains open. Previous evaluations of IA only focus on one or two dimensions and are not comprehensive. In this work, we introduce IAEval for IA methods, a systematic and comprehensive evaluation scheme covering four significant requirements: sufficiency, completeness, stability and plausibility. We elaborately design novel metrics to measure these requirements for the first time. Three representative IA methods are evaluated under IAEval on four natural language understanding datasets. Extensive experiments confirmed the effectiveness of IAEval and exhibited its ability to provide comprehensive comparison among IA methods. With IAEval, researchers can choose the most suitable IA methods for applications like model debugging.
Consider a scenario where an author (e.g., activist, whistle-blower) with many public writings wishes to write “anonymously” when attackers may have already built an authorship attribution (AA) model based off of public writings including those of the author. To enable her wish, we ask a question “can one make the publicly released writings, T , unattributable so that AA models trained on T cannot attribute its authorship well?” Toward this question, we present a novel solution, UPTON, that exploits black-box data poisoning methods to weaken the authorship features in training samples and make released texts unlearnable. It is different from previous obfuscation works (e.g., adversarial attacks that modify test samples or backdoor works that only change the model outputs when triggering words occur). Using four authorship datasets (IMDb10, IMDb64, Enron and WJO), we present empirical validation where UPTON successfully downgrades the accuracy of AA models to the impractical level (e.g., ~ 35%) while keeping texts still readable (e.g., > 0.9 in BERTScore). UPTON remains effective to AA models that are already trained on available clean writings of authors.
The recent explosion of performance of large language models (LLMs) has changed the field of Natural Language Processing (NLP) more abruptly and seismically than any other shift in the field’s 80 year history. This has resulted in concerns that the field will become homogenized and resource-intensive. This new status quo has put many academic researchers, especially PhD students, at a disadvantage. This paper aims to define a new NLP playground by proposing 20+ PhD-dissertation-worthy research directions, covering theoretical analysis, new and challenging problems, learning paradigms and interdisciplinary applications.
Current abstractive summarization models often generate inconsistent content, i.e. texts that are not directly inferable from the source document, are not consistent with respect to world knowledge, or are self-contradictory. These inconsistencies motivate a new consistency taxonomy that we define as faithfulness, factuality, and self-supportiveness. However, most recent work on reducing inconsistency in document summarization only focuses on faithfulness detection and correction while ignoring other inconsistency phenomena, which limits the model’s scalability. To improve the general consistency we introduce EnergySum, where we apply the Residual Energy-based Model by designing energy scorers that reflect each type of consistency. These energy scores are utilized in candidate re-ranking during the sampling process. Experiments on XSUM and CNN/DM datasets show that EnergySum mitigates the trade-off between accuracy and consistency.
Video-grounded Dialogue (VGD) aims to answer questions regarding a given multi-modal input comprising video, audio, and dialogue history. Although there have been numerous efforts in developing VGD systems to improve the quality of their responses, existing systems are competent only to incorporate the information in the video and text and tend to struggle in extracting the necessary information from the audio when generating appropriate responses to the question. The VGD system seems to be deaf, and thus, we coin this symptom of current systems’ ignoring audio data as a deaf response. To overcome the deaf response problem, Hearing Enhanced Audio Response (HEAR) framework is proposed to perform sensible listening by selectively attending to audio whenever the question requires it. The HEAR framework enhances the accuracy and audibility of VGD systems in a model-agnostic manner. HEAR is validated on VGD datasets (i.e., AVSD@DSTC7 and AVSD@DSTC8) and shows effectiveness with various VGD systems.
While large neural-based conversational models have become increasingly proficient dialogue agents, recent work has highlighted safety issues with these systems. For example, these systems can be goaded into generating toxic content, often perpetuating social biases or stereotypes. We investigate a retrieval-based approach for reducing bias and toxicity in responses from chatbots. It uses in-context learning to steer a model towards safer generations. Concretely, to generate a response to an unsafe dialogue context, we retrieve demonstrations of safe responses to similar dialogue contexts. We find our method performs competitively with existing approaches to dialogue safety without requiring training. We also show, using automatic and human evaluation, that reductions in toxicity obtained using our approach are not at the cost engagingness or coherency. Finally, we note our method can be used in compliment to existing dialogue safety approaches, such as RLHF.
Large language models (LLMs) such as ChatGPT and GPT-4 have shown impressive performance in complex reasoning tasks. However, it is difficult to know whether the models are reasoning based on deep understandings of truth and logic, or leveraging their memorized patterns in a relatively superficial way. In this work, we explore testing LLMs’ reasoning by engaging with them in a debate-like conversation, where given a question, the LLM and the user need to discuss to make the correct decision starting from opposing arguments. Upon mitigating the Clever Hans effect, our task requires the LLM to not only achieve the correct answer on its own, but also be able to hold and defend its belief instead of blindly believing or getting misled by the user’s (invalid) arguments and critiques, thus testing in greater depth whether the LLM grasps the essence of the reasoning required to solve the problem. Across a range of complex reasoning benchmarks spanning math, commonsense, logic and BIG-Bench tasks, we find that despite their impressive performance as reported in existing work on generating correct step-by-step solutions in the beginning, LLMs like ChatGPT cannot maintain their beliefs in truth for a significant portion of examples when challenged by oftentimes absurdly invalid arguments. Our work points to danger zones of model alignment, and also suggests more careful treatments and interpretations of the recent findings that LLMs can improve their responses based on feedback.
Spoken Language Understanding (SLU), a crucial component of task-oriented dialogue systems, has consistently garnered attention from both academic and industrial communities. Although incorporating syntactic information into models has the potential to enhance the comprehension of user utterances and yield impressive results, its application in SLU systems remains largely unexplored. In this paper, we propose a carefully designed model termed Syntax-aware attention (SAT) to enhance SLU, where attention scopes are constrained based on relationships within the syntactic structure. Experimental results on three datasets show that our model achieves substantial improvements and excellent performance. Moreover, SAT can be integrated into other BERT-based language models to further boost their performance.
Augmenting pretrained language models (LMs) with a vision encoder (e.g., Flamingo) has obtained state-of-the-art results in image-to-text generation. However, these models store all the knowledge within their parameters, thus often requiring enormous model parameters to model the abundant visual concepts and very rich text descriptions. Additionally, they are inefficient in incorporating new data, requiring a computational-expensive fine-tuning process. In this work, we introduce a Retrieval-augmented Visual Language Model, Re-ViLM, built upon the Flamingo, that supports retrieving the relevant knowledge from the external database for zero and in-context few-shot image-to-text generations. By storing certain knowledge explicitly in the external database, our approach reduces the number of model parameters and can easily accommodate new data during evaluation by simply updating the database. We also construct an interleaved image and text data that facilitates in-context few-shot learning capabilities.We demonstrate that Re-ViLM significantly boosts performance for image-to-text generation tasks, especially for zero-shot and few-shot generation in out-of-domain settings with 4x less parameters compared with baseline methods.
In Psychotherapy, maladaptive schemas– negative perceptions that an individual has of the self, others, or the world that endure despite objective reality–often lead to resistance to treatments and relapse of mental health issues such as depression, anxiety, panic attacks etc. Identification of early maladaptive schemas (EMS) is thus a crucial step during Schema Therapy-based counseling sessions, where patients go through a detailed and lengthy EMS questionnaire. However, such an approach is not practical in ‘offline’ counseling scenarios, such as community QA forums which are gaining popularity for people seeking mental health support. In this paper, we investigate both LLM (Large Language Models) and non-LLM approaches for identifying EMS labels using resources from Schema Therapy. Our evaluation indicates that recent LLMs can be effective for identifying EMS but their predictions lack explainability and are too sensitive to precise ‘prompts’. Both LLM and non-LLM methods are unable to reliably address the null cases, i.e. cases with no EMS labels. However, we posit that the two approaches show complementary properties and together, they can be used to further devise techniques for EMS identification.
Let’s Synthesize Step by Step: Iterative Dataset Synthesis with Large Language Models by Extrapolating Errors from Small Models
PDF ↗*Data Synthesis* is a promising way to train a small model with very little labeled data. One approach for data synthesis is to leverage the rich knowledge from large language models to synthesize pseudo training examples for small models, making it possible to achieve both data and compute efficiency at the same time. However, a key challenge in data synthesis is that the synthesized dataset often suffers from a large distributional discrepancy from the *real task* data distribution. Thus, in this paper, we propose *Synthesis Step by Step* (**S3**), a data synthesis framework that shrinks this distribution gap by iteratively extrapolating the errors made by a small model trained on the synthesized dataset on a small real-world validation dataset using a large language model. Extensive experiments on multiple NLP tasks show that our approach improves the performance of a small model by reducing the gap between the synthetic dataset and the real data, resulting in significant improvement compared to several baselines: 9.48% improvement compared to ZeroGen and 2.73% compared to GoldGen, and at most 15.17% improvement compared to the small model trained on human-annotated data.
Clickbait posts tend to spread inaccurate or misleading information to manipulate people’s attention and emotions, which greatly harms the credibility of social media. Existing clickbait detection models rely on analyzing the objective semantics in posts or correlating posts with article content only. However, these models fail to identify and exploit the manipulation intention of clickbait from a user’s subjective perspective, leading to limited capability to explore comprehensive clues of clickbait. To address such a issue, we propose a multiview clickbait detection model, named MCDM, to model subjective and objective preferences simultaneously. MCDM introduces two novel complementary modules for modeling subjective feeling and objective content relevance, respectively. The subjective feeling module adopts a user-centric approach to capture subjective features of posts, such as language patterns and emotional inclinations. The objective module explores news elements from posts and models article content correlations to capture objective clues for clickbait detection. Extensive experimental results on two real-world datasets show that our proposed MCDM outperforms state-of-the-art approaches for clickbait detection, verifying the effectiveness of integrating subjective and objective preferences for detecting clickbait.
Storytelling’s captivating potential makes it a fascinating research area, with implications for entertainment, education, therapy, and cognitive studies. In this paper, we propose Affective Story Generator (AffGen) for generating interesting narratives. AffGen introduces ‘intriguing twists’ in narratives by employing two novel techniques—Dynamic Beam Sizing and Affective Reranking. Dynamic Beam Sizing encourages less predictable, more captivating word choices using a contextual multi-arm bandit model. Affective Reranking prioritizes sentence candidates based on affect intensity. Our empirical evaluations, both automatic and human, demonstrate AffGen’s superior performance over existing baselines in generating affectively charged and interesting narratives. Our ablation study and analysis provide insights into the strengths and weaknesses of AffGen.
The vast majority of the popular English named entity recognition (NER) datasets contain American or British English data, despite the existence of many global varieties of English. As such, it is unclear whether they generalize for analyzing use of English globally. To test this, we build a newswire dataset, the Worldwide English NER Dataset, to analyze NER model performance on low-resource English variants from around the world. We test widely used NER toolkits and transformer models, including models using the pre-trained contextual models RoBERTa and ELECTRA, on three datasets: a commonly used British English newswire dataset, CoNLL 2003, a more American focused dataset OntoNotes, and our global dataset. All models trained on the CoNLL or OntoNotes datasets experienced significant performance drops—over 10 F1 in some cases—when tested on the Worldwide English dataset. Upon examination of region-specific errors, we observe the greatest performance drops for Oceania and Africa, while Asia and the Middle East had comparatively strong performance. Lastly, we find that a combined model trained on the Worldwide dataset and either CoNLL or OntoNotes lost only 1-2 F1 on both test sets.
Multilingual Coarse Political Stance Classification of Media. The Editorial Line of a ChatGPT and Bard Newspaper
PDF ↗Neutrality is difficult to achieve and, in politics, subjective. Traditional media typically adopt an editorial line that can be used by their potential readers as an indicator of the media bias. Several platforms currently rate news outlets according to their political bias. The editorial line and the ratings help readers in gathering a balanced view of news. But in the advent of instruction-following language models, tasks such as writing a newspaper article can be delegated to computers. Without imposing a biased persona, where would an AI-based news outlet lie within the bias ratings? In this work, we use the ratings of authentic news outlets to create a multilingual corpus of news with coarse stance annotations (Left and Right) along with automatically extracted topic annotations. We show that classifiers trained on this data are able to identify the editorial line of most unseen newspapers in English, German, Spanish and Catalan. We then apply the classifiers to 101 newspaper-like articles written by ChatGPT and Bard in the 4 languages at different time periods. We observe that, similarly to traditional newspapers, ChatGPT editorial line evolves with time and, being a data-driven system, the stance of the generated articles differs among languages.
Ultra-Fine Entity Typing with Prior Knowledge about Labels: A Simple Clustering Based Strategy
PDF ↗Ultra-fine entity typing (UFET) is the task of inferring the semantic types from a large set of fine-grained candidates that apply to a given entity mention. This task is especially challenging because we only have a small number of training examples for many types, even with distant supervision strategies. State-of-the-art models, therefore, have to rely on prior knowledge about the type labels in some way. In this paper, we show that the performance of existing methods can be improved using a simple technique: we use pre-trained label embeddings to cluster the labels into semantic domains and then treat these domains as additional types. We show that this strategy consistently leads to improved results as long as high-quality label embeddings are used. Furthermore, we use the label clusters as part of a simple post-processing technique, which results in further performance gains. Both strategies treat the UFET model as a black box and can thus straightforwardly be used to improve a wide range of existing models.