Previous studies disclose that Large Language Models (LLMs) suffer from hallucinations when generating texts, bringing a novel and challenging research topic to the public, which centers on enabling LLMs to generate texts with citations. Existing work exposes two limitations when using LLMs to generate answers to questions with provided documents: unsatisfactory answer correctness and poor citation quality. To tackle the above issues, we investigate using Chain-of-Thought (CoT) to elicit LLMs’ ability to synthesize correct answers from multiple documents, as well as properly cite these documents. Moreover, we propose a Citation Insurance Mechanism, which enables LLMs to detect and cite those missing citations. We conduct experiments on the ALCE benchmark with six open-source LLMs. Experimental results demonstrate that: (1) the CoT prompting strategy significantly improves the quality of text generation with citations; (2) the Citation Insurance Mechanism delivers impressive gains in citation quality at a low cost; (3) our best approach performs comparably as previous best ChatGPT-based baselines. Extensive analyses further validate the effectiveness of the proposed approach.
论文检索
输入标题、作者或关键词，从 2,866 篇学术成果中精准定位
Zero-shot multi-speaker TTS aims to synthesize speech with the voice of a chosen target speaker without any fine-tuning. Prevailing methods, however, encounter limitations at adapting to new speakers of out-of-domain settings, primarily due to inadequate speaker disentanglement and content leakage. To overcome these constraints, we propose an innovative negation feature learning paradigm that models decoupled speaker attributes as deviations from the complete audio representation by utilizing the subtraction operation. By eliminating superfluous content information from the speaker representation, our negation scheme not only mitigates content leakage, thereby enhancing synthesis robustness, but also improves speaker fidelity. In addition, to facilitate the learning of diverse speaker attributes, we leverage multi-stream Transformers, which retain multiple hypotheses and instigate a training paradigm akin to ensemble learning. To unify these hypotheses and realize the final speaker representation, we employ attention pooling. Finally, in light of the imperative to generate target text utterances in the desired voice, we adopt adaptive layer normalizations to effectively fuse the previously generated speaker representation with the target text representations, as opposed to mere concatenation of the text and audio modalities. Extensive experiments and validations substantiate the efficacy of our proposed approach in preserving and harnessing speaker-specific attributes vis-à-vis alternative baseline models.
Document-level relation extraction (DocRE) poses the challenge of identifying relationships between entities within a document. Existing approaches rely on logical reasoning or contextual cues from entities. This paper reframes document-level RE as link prediction over a Knowledge Graph (KG) with distinct benefits: 1) Our approach amalgamates entity context and document-derived logical reasoning, enhancing link prediction quality. 2) Predicted links between entities offer interpretability, elucidating employed reasoning. We evaluate our approach on benchmark datasets - DocRED, ReDocRED, and DWIE. The results indicate that our proposed method outperforms the state-of-the-art models and suggests that incorporating context-based Knowledge Graph link prediction techniques can enhance the performance of document-level relation extraction models.
The prompt-based method has been proven effective in improving the performance of pre-trained language models (PLMs) on sentence-level few-shot tasks. However, when applying prompting to token-level tasks such as Named Entity Recognition (NER), specific templates need to be designed, and all possible segments of the input text need to be enumerated. These methods have high computational complexity in both training and inference processes, making them difficult to apply in real-world scenarios. To address these issues, we redefine the NER task as a Machine Reading Comprehension (MRC) task and incorporate prompting into the MRC framework. Specifically, we sequentially insert boundary markers for various entity types into the templates and use these markers as anchors during the inference process to differentiate entity types. In contrast to the traditional multi-turn question-answering extraction in the MRC framework, our method can extract all spans of entity types in one round. Furthermore, we propose word-based template and example-based template that enhance the MRC framework's perception of entity start and end positions while significantly reducing the manual effort required for template design. It is worth noting that in cross-domain scenarios, PMRC does not require redesigning the model architecture and can continue training by simply replacing the templates to recognize entity types in the target domain. Experimental results demonstrate that our approach outperforms state-of-the-art models in low-resource settings, achieving an average performance improvement of +5.2% in settings where access to source domain data is limited. Particularly, on the ATIS dataset with a large number of entity types and 10-shot setting, PMRC achieves a performance improvement of +15.7%. Moreover, our method achieves a decoding speed 40.56 times faster than the template-based cloze-style approach.
Leveraging vast and continually updated knowledge from the Internet has been considered an important ability for a dialogue system. Therefore, the dialogue query generation task is proposed for generating search queries from dialogue histories, which will be submitted to a search engine for retrieving relevant websites on the Internet. In this regard, previous efforts were devoted to collecting conversations with annotated queries and training a query producer (QP) via standard supervised learning. However, these studies still face the challenges of data scarcity and domain adaptation. To address these issues, in this paper, we propose a semi-supervised learning framework -- SemiDQG, to improve model performance with unlabeled conversations. Based on the observation that the search query is typically related to the topic of dialogue response, we train a response-augmented query producer (RA) to provide rich and effective training signals for QP. We first apply a similarity-based query selection strategy to select high-quality RA-generated pseudo queries, which are used to construct pseudo instances for training QP and RA. Then, we adopt the REINFORCE algorithm to further enhance QP, with RA-provided rewards as fine-grained training signals. Experimental results and in-depth analysis of three benchmarks show the effectiveness of our framework in cross-domain and low-resource scenarios. Particularly, SemiDQG significantly surpasses ChatGPT and competitive baselines. Our code is available at \url{https://github.com/DeepLearnXMU/SemiDQG}.
Current image-text retrieval methods have demonstrated impressive performance in recent years. However, they still face two problems: the inter-modal matching missing problem and the intra-modal semantic loss problem. These problems can significantly affect the accuracy of image-text retrieval. To address these challenges, we propose a novel method called Cross-modal and Uni-modal Soft-label Alignment (CUSA). Our method leverages the power of uni-modal pre-trained models to provide soft-label supervision signals for the image-text retrieval model. Additionally, we introduce two alignment techniques, Cross-modal Soft-label Alignment (CSA) and Uni-modal Soft-label Alignment (USA), to overcome false negatives and enhance similarity recognition between uni-modal samples. Our method is designed to be plug-and-play, meaning it can be easily applied to existing image-text retrieval models without changing their original architectures. Extensive experiments on various image-text retrieval models and datasets, we demonstrate that our method can consistently improve the performance of image-text retrieval and achieve new state-of-the-art results. Furthermore, our method can also boost the uni-modal retrieval performance of image-text retrieval models, enabling it to achieve universal retrieval. The code and supplementary files can be found at https://github.com/lerogo/aaai24_itr_cusa.
One of the key factors in language productivity and human cognition is the ability of Systematic Compositionality, which refers to understanding composed, unseen examples of seen primitives. However, recent evidence reveals that the Transformers have difficulty in generalizing the composed context based on the seen primitives. To this end, we take the first step to propose a compositionality-aware Transformer called CAT and two novel pre-training tasks to facilitate the systematic compositionality. We tentatively provide a successful implementation of a multi-layer CAT on the basis of the especially popular BERT. The experimental results demonstrate that CAT outperforms baselines on compositionality-aware tasks with minimal impact on effectiveness on standardized language understanding tasks.
Controllable text generation is a challenging and meaningful field in natural language generation (NLG). Especially, poetry generation is a typical one with well-defined and strict conditions for text generation which is an ideal playground for the assessment of current methodologies. While prior works succeeded in controlling either semantic or metrical aspects of poetry generation, simultaneously addressing both remains a challenge. In this paper, we pioneer the use of the Diffusion model for generating sonnets and Chinese SongCi poetry to tackle such challenges. In terms of semantics, our PoetryDiffusion model, built upon the Diffusion model, generates entire sentences or poetry by comprehensively considering the entirety of sentence information. This approach enhances semantic expression, distinguishing it from autoregressive and large language models (LLMs). For metrical control, its constraint control module which can be trained individually enables us to flexibly incorporate a novel metrical controller to manipulate and evaluate metrics (format and rhythm). The denoising process in PoetryDiffusion allows for the gradual enhancement of semantics and flexible integration of the metrical controller which can calculate and impose penalties on states that stray significantly from the target control distribution. Experimental results on two datasets demonstrate that our model outperforms existing models in terms of automatic evaluation of semantic, metrical, and overall performance as well as human evaluation. Codes are released to https://github.com/ChorlingLau/PoetryDiffusion.
Uncovering and Mitigating the Hidden Chasm: A Study on the Text-Text Domain Gap in Euphemism Identification
PDF ↗Euphemisms are commonly used on social media and darknet marketplaces to evade platform regulations by masking their true meanings with innocent ones. For instance, “weed” is used instead of “marijuana” for illicit transactions. Thus, euphemism identification, i.e., mapping a given euphemism (“weed”) to its specific target word (“marijuana”), is essential for improving content moderation and combating underground markets. Existing methods employ self-supervised schemes to automatically construct labeled training datasets for euphemism identification. However, they overlook the text-text domain gap caused by the discrepancy between the constructed training data and the test data, leading to performance deterioration. In this paper, we present the text-text domain gap and explain how it forms in terms of the data distribution and the cone effect. Moreover, to bridge this gap, we introduce a feature alignment network (FA-Net), which can both align the in-domain and cross-domain features, thus mitigating the domain gap from training data to test data and improving the performance of the base models for euphemism identification. We apply this FA-Net to the base models, obtaining markedly better results, and creating a state-of-the-art model which beats the large language models.
Three Heads Are Better than One: Improving Cross-Domain NER with Progressive Decomposed Network
PDF ↗Cross-domain named entity recognition (NER) tasks encourage NER models to transfer knowledge from data-rich source domains to sparsely labeled target domains. Previous works adopt the paradigms of pre-training on the source domain followed by fine-tuning on the target domain. However, these works ignore that general labeled NER source domain data can be easily retrieved in the real world, and soliciting more source domains could bring more benefits. Unfortunately, previous paradigms cannot efficiently transfer knowledge from multiple source domains. In this work, to transfer multiple source domains' knowledge, we decouple the NER task into the pipeline tasks of mention detection and entity typing, where the mention detection unifies the training object across domains, thus providing the entity typing with higher-quality entity mentions. Additionally, we request multiple general source domain models to suggest the potential named entities for sentences in the target domain explicitly, and transfer their knowledge to the target domain models through the knowledge progressive networks implicitly. Furthermore, we propose two methods to analyze in which source domain knowledge transfer occurs, thus helping us judge which source domain brings the greatest benefit. In our experiment, we develop a Chinese cross-domain NER dataset. Our model improved the F1 score by an average of 12.50% across 8 Chinese and English datasets compared to models without source domain data.
Separate the Wheat from the Chaff: Model Deficiency Unlearning via Parameter-Efficient Module Operation
PDF ↗Large language models (LLMs) have been widely used in various applications but are known to suffer from issues related to untruthfulness and toxicity. While parameter-efficient modules (PEMs) have demonstrated their effectiveness in equipping models with new skills, leveraging PEMs for deficiency unlearning remains underexplored. In this work, we propose a PEMs operation approach, namely Extraction-before-Subtraction (Ext-Sub), to enhance the truthfulness and detoxification of LLMs through the integration of ``expert'' PEM and ``anti-expert'' PEM. Remarkably, even anti-expert PEM possess valuable capabilities due to their proficiency in generating fabricated content, which necessitates language modeling and logical narrative competence. Rather than merely negating the parameters, our approach involves extracting and eliminating solely the deficiency capability within anti-expert PEM while preserving the general capabilities. To evaluate the effectiveness of our approach in terms of truthfulness and detoxification, we conduct extensive experiments on LLMs, encompassing additional abilities such as language modelling and mathematical reasoning. Our empirical results demonstrate that our approach effectively improves truthfulness and detoxification, while largely preserving the fundamental abilities of LLMs.
Selective rationalization can be regarded as a straightforward self-explaining approach for enhancing model explainability in natural language processing tasks. It aims to provide explanations that are more accessible and understandable to non-technical users by first selecting subsets of input texts as rationales and then predicting based on chosen subsets. However, existing methods that follow this select-then-predict framework may suffer from the rationalization degeneration problem, resulting in sub-optimal or unsatisfactory rationales that do not align with human judgments. This problem may further lead to rationalization failure, resulting in meaningless rationales that ultimately undermine people's trust in the rationalization model. To address these challenges, we propose a Guidance-based Rationalization method (G-RAT) that effectively improves robustness against failure situations and the quality of rationales by using a guidance module to regularize selections and distributions. Experimental results on two synthetic settings prove that our method is robust to the rationalization degeneration and failure problems, while the results on two real datasets show its effectiveness in providing rationales in line with human judgments. The source code is available at https://github.com/shuaibo919/g-rat.
LLM vs Small Model? Large Language Model Based Text Augmentation Enhanced Personality Detection Model
PDF ↗Personality detection aims to detect one's personality traits underlying in social media posts. One challenge of this task is the scarcity of ground-truth personality traits which are collected from self-report questionnaires. Most existing methods learn post features directly by fine-tuning the pre-trained language models under the supervision of limited personality labels. This leads to inferior quality of post features and consequently affects the performance. In addition, they treat personality traits as one-hot classification labels, overlooking the semantic information within them. In this paper, we propose a large language model (LLM) based text augmentation enhanced personality detection model, which distills the LLM's knowledge to enhance the small model for personality detection, even when the LLM fails in this task. Specifically, we enable LLM to generate post analyses (augmentations) from the aspects of semantic, sentiment, and linguistic, which are critical for personality detection. By using contrastive learning to pull them together in the embedding space, the post encoder can better capture the psycho-linguistic information within the post representations, thus improving personality detection. Furthermore, we utilize the LLM to enrich the information of personality labels for enhancing the detection performance. Experimental results on the benchmark datasets demonstrate that our model outperforms the state-of-the-art methods on personality detection.
The deployment of Pre-trained Language Models in memory-limited devices is hindered by their massive number of parameters, which motivated the interest in developing smaller architectures. Established works in the model compression literature showcased that small models often present a noticeable performance degradation and need to be paired with transfer learning methods, such as Knowledge Distillation. In this work, we propose a parameter-sharing method that consists of sharing parameters between embeddings and the hidden layers, enabling the design of near-zero parameter encoders. To demonstrate its effectiveness, we present an architecture design called ShareBERT, which can preserve up to 95.5% of BERT Base performances, using only 5M parameters (21.9× fewer parameters) without the help of Knowledge Distillation. We demonstrate empirically that our proposal does not negatively affect the model learning capabilities and that it is even beneficial for representation learning. Code will be available at https://github.com/jchenghu/sharebert.
Textual style transfer is the task of transforming stylistic properties of text while preserving meaning. Target "styles" can be defined in numerous ways, ranging from single attributes (e.g. formality) to authorship (e.g. Shakespeare). Previous unsupervised style-transfer approaches generally rely on significant amounts of labeled data for only a fixed set of styles or require large language models. In contrast, we introduce a novel diffusion-based framework for general-purpose style transfer that can be flexibly adapted to arbitrary target styles at inference time. Our parameter-efficient approach, ParaGuide, leverages paraphrase-conditioned diffusion models alongside gradient-based guidance from both off-the-shelf classifiers and strong existing style embedders to transform the style of text while preserving semantic information. We validate the method on the Enron Email Corpus, with both human and automatic evaluations, and find that it outperforms strong baselines on formality, sentiment, and even authorship style transfer.
Text2Analysis: A Benchmark of Table Question Answering with Advanced Data Analysis and Unclear Queries
PDF ↗Tabular data analysis is crucial in various fields, and large language models show promise in this area. However, current research mostly focuses on rudimentary tasks like Text2SQL and TableQA, neglecting advanced analysis like forecasting and chart generation. To address this gap, we developed the Text2Analysis benchmark, incorporating advanced analysis tasks that go beyond the SQL-compatible operations and require more in-depth analysis. We also develop five innovative and effective annotation methods, harnessing the capabilities of large language models to enhance data quality and quantity. Additionally, we include unclear queries that resemble real-world user questions to test how well models can understand and tackle such challenges. Finally, we collect 2249 query-result pairs with 347 tables. We evaluate five state-of-the-art models using three different metrics and the results show that our benchmark presents introduces considerable challenge in the field of tabular data analysis, paving the way for more advanced research opportunities.
Factual error correction (FEC) aims to revise factual errors in false claims with minimal editing, making them faithful to the provided evidence. This task is crucial for alleviating the hallucination problem encountered by large language models. Given the lack of paired data (i.e., false claims and their corresponding correct claims), existing methods typically adopt the ‘mask-then-correct’ paradigm. This paradigm relies solely on unpaired false claims and correct claims, thus being referred to as distantly supervised methods. These methods require a masker to explicitly identify factual errors within false claims before revising with a corrector. However, the absence of paired data to train the masker makes accurately pinpointing factual errors within claims challenging. To mitigate this, we propose to improve FEC by Learning to Inject Factual Errors (LIFE), a three-step distantly supervised method: ‘mask-corrupt-correct’. Specifically, we first train a corruptor using the ‘mask-then-corrupt’ procedure, allowing it to deliberately introduce factual errors into correct text. The corruptor is then applied to correct claims, generating a substantial amount of paired data. After that, we filter out low-quality data, and use the remaining data to train a corrector. Notably, our corrector does not require a masker, thus circumventing the bottleneck associated with explicit factual error identification. Our experiments on a public dataset verify the effectiveness of LIFE in two key aspects: Firstly, it outperforms the previous best-performing distantly supervised method by a notable margin of 10.59 points in SARI Final (19.3% improvement). Secondly, even compared to ChatGPT prompted with in-context examples, LIFE achieves a superiority of 7.16 points in SARI Final.
Large language models (LLMs) can understand human instructions, showing their potential for pragmatic applications beyond traditional NLP tasks. However, they still struggle with complex instructions, which can be either complex task descriptions that require multiple tasks and constraints, or complex input that contains long context, noise, heterogeneous information and multi-turn format. Due to these features, LLMs often ignore semantic constraints from task descriptions, generate incorrect formats, violate length or sample count constraints, and be unfaithful to the input text. Existing benchmarks are insufficient to assess LLMs’ ability to understand complex instructions, as they are close-ended and simple. To bridge this gap, we propose CELLO, a benchmark for evaluating LLMs' ability to follow complex instructions systematically. We design eight features for complex instructions and construct a comprehensive evaluation dataset from real-world scenarios. We also establish four criteria and develop corresponding metrics, as current ones are inadequate, biased or too strict and coarse-grained. We compare the performance of representative Chinese-oriented and English-oriented models in following complex instructions through extensive experiments. Resources of CELLO are publicly available at https://github.com/Abbey4799/CELLO.
Chain-of-thought (CoT) reasoning has exhibited impressive performance in language models for solving complex tasks and answering questions. However, many real-world questions require multi-modal information, such as text and images. Previous research on multi-modal CoT has primarily focused on extracting fixed image features from off-the-shelf vision models and then fusing them with text using attention mechanisms. This approach has limitations because these vision models were not designed for complex reasoning tasks and do not align well with language thoughts. To overcome this limitation, we introduce a novel approach for multi-modal CoT reasoning that utilizes latent space learning via diffusion processes to generate effective image features that align with language thoughts. Our method fuses image features and text representations at a deep level and improves the complex reasoning ability of multi-modal CoT. We demonstrate the efficacy of our proposed method on multi-modal ScienceQA and machine translation benchmarks, achieving state-of-the-art performance on ScienceQA. Overall, our approach offers a more robust and effective solution for multi-modal reasoning in language models, enhancing their ability to tackle complex real-world problems.
Few-shot intent classification and slot filling are important but challenging tasks due to the scarcity of finely labeled data. Therefore, current works first train a model on source domains with sufficiently labeled data, and then transfer the model to target domains where only rarely labeled data is available. However, experience transferring as a whole usually suffers from gaps that exist among source domains and target domains. For instance, transferring domain-specific-knowledge-related experience is difficult. To tackle this problem, we propose a new method that explicitly decouples the transferring of general-semantic-representation-related experience and the domain-specific-knowledge-related experience. Specifically, for domain-specific-knowledge-related experience, we design two modules to capture intent-slot relation and slot-slot relation respectively. Extensive experiments on Snips and FewJoint datasets show that our method achieves state-of-the-art performance. The method improves the joint accuracy metric from 27.72% to 42.20% in the 1-shot setting, and from 46.54% to 60.79% in the 5-shot setting.