Model editing is a technique that edits the large language models (LLMs) with updated knowledge to alleviate hallucinations without resource-intensive retraining. While current model editing methods can effectively modify a model’s behavior within a specific area of interest, they often overlook the potential unintended side effects on the general abilities of LLMs such as reasoning, natural language inference, and question answering. In this paper, we raise concerns that model editing’s improvements on factuality may come at the cost of a significant degradation of the model’s general abilities. We systematically analyze the side effects by evaluating four popular editing methods on three LLMs across eight representative tasks. Our extensive empirical experiments show that it is challenging for current editing methods to simultaneously improve factuality of LLMs and maintain their general abilities. Our analysis reveals that the side effects are caused by model editing altering the original model weights excessively, leading to overfitting to the edited facts. To mitigate this, a method named RECT is proposed to regularize the edit update weights by imposing constraints on their complexity based on the RElative Change in weighT. Evaluation results show that RECT can significantly mitigate the side effects of editing while still maintaining over 94% editing performance.

Automated audio captioning (AAC) aims to generate descriptions based on audio input, attracting exploration of emerging audio language models (ALMs). However, current evaluation metrics only provide a single score to assess the overall quality of captions without characterizing the nuanced difference by systematically going through an evaluation checklist. To this end, we propose the explainable and multi-factor audio captioning evaluation (X-ACE) paradigm. X-ACE identifies four main factors that constitute the majority of audio features, specifically sound event, source, attribute and relation. To assess a given caption from an ALM, it is firstly transformed into an audio graph, where each node denotes an entity in the caption and corresponds to a factor. On the one hand, graph matching is conducted from part to whole for a holistic assessment. On the other hand, the nodes contained within each factor are aggregated to measure the factor-level performance. The pros and cons of an ALM can be explicitly and clearly demonstrated through X-ACE, pointing out the direction for further improvements. Experiments show that X-ACE exhibits better correlation with human perception and can detect mismatches sensitively.

Recent Retrieval Augmented Generation (RAG) aims to enhance Large Language Models (LLMs) by incorporating extensive knowledge retrieved from external sources. However, such approach encounters some challenges: Firstly, the original queries may not be suitable for precise retrieval, resulting in erroneous contextual knowledge; Secondly, the language model can easily generate inconsistent answer with external references due to their knowledge boundary limitation. To address these issues, we propose the chain-of-verification (CoV-RAG) to enhance the external retrieval correctness and internal generation consistency. Specifically, we integrate the verification module into the RAG, engaging in scoring, judgment, and rewriting. To correct external retrieval errors, CoV-RAG retrieves new knowledge using a revised query. To correct internal generation errors, we unify QA and verification tasks with a Chain-of-Thought (CoT) reasoning during training. Our comprehensive experiments across various LLMs demonstrate the effectiveness and adaptability compared with other strong baselines. Especially, our CoV-RAG can significantly surpass the state-of-the-art baselines using different LLM backbones.

Speaker identification in novel dialogues can be widely applied to various downstream tasks, such as producing multi-speaker audiobooks and converting novels into scripts. However, existing state-of-the-art methods are limited to handling explicit narrative patterns like “Tom said, '...'", unable to thoroughly understand long-range contexts and to deal with complex cases. To this end, we propose a framework named SPC, which identifies implicit speakers in novels via symbolization, prompt, and classification. First, SPC symbolizes the mentions of candidate speakers to construct a unified label set. Then, by inserting a prompt we re-formulate speaker identification as a classification task to minimize the gap between the training objectives of speaker identification and the pre-training task. Two auxiliary tasks are also introduced in SPC to enhance long-range context understanding. Experimental results show that SPC outperforms previous methods by a large margin of 4.8% accuracy on the web novel collection, which reduces 47% of speaker identification errors, and also outperforms the emerging ChatGPT. In addition, SPC is more accurate in implicit speaker identification cases that require long-range context semantic understanding.

Large Language Models (LLMs) have emerged as influential instruments within the realm of natural language processing; nevertheless, their capacity to handle multi-party conversations (MPCs) – a scenario marked by the presence of multiple interlocutors involved in intricate information exchanges – remains uncharted. In this paper, we delve into the potential of generative LLMs such as ChatGPT and GPT-4 within the context of MPCs. An empirical analysis is conducted to assess the zero-shot learning capabilities of ChatGPT and GPT-4 by subjecting them to evaluation across three MPC datasets that encompass five representative tasks. The findings reveal that ChatGPT’s performance on a number of evaluated MPC tasks leaves much to be desired, whilst GPT-4’s results portend a promising future. Additionally, we endeavor to bolster performance through the incorporation of MPC structures, encompassing both speaker and addressee architecture. This study provides an exhaustive evaluation and analysis of applying generative LLMs to MPCs, casting a light upon the conception and creation of increasingly effective and robust MPC agents. Concurrently, this work underscores the challenges implicit in the utilization of LLMs for MPCs, such as deciphering graphical information flows and generating stylistically consistent responses.

Modeling multi-party conversations (MPCs) with graph neural networks has been proven effective at capturing complicated and graphical information flows. However, existing methods rely heavily on the necessary addressee labels and can only be applied to an ideal setting where each utterance must be tagged with an “@” or other equivalent addressee label. To study the scarcity of addressee labels which is a common issue in MPCs, we propose MADNet that maximizes addressee deduction expectation in heterogeneous graph neural networks for MPC generation. Given an MPC with a few addressee labels missing, existing methods fail to build a consecutively connected conversation graph, but only a few separate conversation fragments instead. To ensure message passing between these conversation fragments, four additional types of latent edges are designed to complete a fully-connected graph. Besides, to optimize the edge-type-dependent message passing for those utterances without addressee labels, an Expectation-Maximization-based method that iteratively generates silver addressee labels (E step), and optimizes the quality of generated responses (M step), is designed. Experimental results on two Ubuntu IRC channel benchmarks show that MADNet outperforms various baseline models on the task of MPC generation, especially under the more common and challenging setting where part of addressee labels are missing.

When multiple conversations occur simultaneously, a listener must decide which conversation each utterance is part of in order to interpret and respond to it appropriately. This task is referred as dialogue disentanglement. A significant drawback of previous studies on disentanglement lies in that they only focus on pair-wise relationships between utterances while neglecting the conversation structure which is important for conversation structure modeling. In this paper, we propose a hierarchical model, named Dialogue BERT (DIALBERT), which integrates the local and global semantics in the context range by using BERT to encode each message-pair and using BiLSTM to aggregate the chronological context information into the output of BERT. In order to integrate the conversation structure information into the model, two types of loss of conversation-structure loss and tree-structure loss are designed. In this way, our model can implicitly learn and leverage the conversation structures without being restricted to the lack of explicit access to such structures during the inference stage. Experimental results on two large datasets show that our method outperforms previous methods by substantial margins, achieving great performance on dialogue disentanglement.

Recently, various response generation models for two-party conversations have achieved impressive improvements, but less effort has been paid to multi-party conversations (MPCs) which are more practical and complicated. Compared with a two-party conversation where a dialogue context is a sequence of utterances, building a response generation model for MPCs is more challenging, since there exist complicated context structures and the generated responses heavily rely on both interlocutors (i.e., speaker and addressee) and history utterances. To address these challenges, we present HeterMPC, a heterogeneous graph-based neural network for response generation in MPCs which models the semantics of utterances and interlocutors simultaneously with two types of nodes in a graph. Besides, we also design six types of meta relations with node-edge-type-dependent parameters to characterize the heterogeneous interactions within the graph. Through multi-hop updating, HeterMPC can adequately utilize the structural knowledge of conversations for response generation. Experimental results on the Ubuntu Internet Relay Chat (IRC) channel benchmark show that HeterMPC outperforms various baseline models for response generation in MPCs.

Generating natural and informative texts has been a long-standing problem in NLP. Much effort has been dedicated into incorporating pre-trained language models (PLMs) with various open-world knowledge, such as knowledge graphs or wiki pages. However, their ability to access and manipulate the task-specific knowledge is still limited on downstream tasks, as this type of knowledge is usually not well covered in PLMs and is hard to acquire. To address the problem, we propose augmenting TExt Generation via Task-specific and Open-world Knowledge (TegTok) in a unified framework. Our model selects knowledge entries from two types of knowledge sources through dense retrieval and then injects them into the input encoding and output decoding stages respectively on the basis of PLMs. With the help of these two types of knowledge, our model can learn what and how to generate. Experiments on two text generation tasks of dialogue generation and question generation, and on two datasets show that our method achieves better performance than various baseline models.

This paper describes the system developed by the USTC-NELSLIP team for SemEval-2022 Task 11 Multilingual Complex Named Entity Recognition (MultiCoNER). We propose a gazetteer-adapted integration network (GAIN) to improve the performance of language models for recognizing complex named entities. The method first adapts the representations of gazetteer networks to those of language models by minimizing the KL divergence between them. After adaptation, these two networks are then integrated for backend supervised named entity recognition (NER) training. The proposed method is applied to several state-of-the-art Transformer-based NER models with a gazetteer built from Wikidata, and shows great generalization ability across them. The final predictions are derived from an ensemble of these trained models. Experimental results and detailed analysis verify the effectiveness of the proposed method. The official results show that our system ranked 1st on three tracks (Chinese, Code-mixed and Bangla) and 2nd on the other ten tracks in this task.

This paper presents a multi-level matching and aggregation network (MLMAN) for few-shot relation classification. Previous studies on this topic adopt prototypical networks, which calculate the embedding vector of a query instance and the prototype vector of the support set for each relation candidate independently. On the contrary, our proposed MLMAN model encodes the query instance and each support set in an interactive way by considering their matching information at both local and instance levels. The final class prototype for each support set is obtained by attentive aggregation over the representations of support instances, where the weights are calculated using the query instance. Experimental results demonstrate the effectiveness of our proposed methods, which achieve a new state-of-the-art performance on the FewRel dataset.

This paper presents a neural relation extraction method to deal with the noisy training data generated by distant supervision. Previous studies mainly focus on sentence-level de-noising by designing neural networks with intra-bag attentions. In this paper, both intra-bag and inter-bag attentions are considered in order to deal with the noise at sentence-level and bag-level respectively. First, relation-aware bag representations are calculated by weighting sentence embeddings using intra-bag attentions. Here, each possible relation is utilized as the query for attention calculation instead of only using the target relation in conventional methods. Furthermore, the representation of a group of bags in the training set which share the same relation label is calculated by weighting bag representations using a similarity-based inter-bag attention module. Finally, a bag group is utilized as a training sample when building our relation extractor. Experimental results on the New York Times dataset demonstrate the effectiveness of our proposed intra-bag and inter-bag attention modules. Our method also achieves better relation extraction accuracy than state-of-the-art methods on this dataset.

This paper proposes a dually interactive matching network (DIM) for presenting the personalities of dialogue agents in retrieval-based chatbots. This model develops from the interactive matching network (IMN) which models the matching degree between a context composed of multiple utterances and a response candidate. Compared with previous persona fusion approach which enhances the representation of a context by calculating its similarity with a given persona, the DIM model adopts a dual matching architecture, which performs interactive matching between responses and contexts and between responses and personas respectively for ranking response candidates. Experimental results on PERSONA-CHAT dataset show that the DIM model outperforms its baseline model, i.e., IMN with persona fusion, by a margin of 14.5% and outperforms the present state-of-the-art model by a margin of 27.7% in terms of top-1 accuracy hits@1.

Modeling natural language inference is a very challenging task. With the availability of large annotated data, it has recently become feasible to train complex models such as neural-network-based inference models, which have shown to achieve the state-of-the-art performance. Although there exist relatively large annotated data, can machines learn all knowledge needed to perform natural language inference (NLI) from these data? If not, how can neural-network-based NLI models benefit from external knowledge and how to build NLI models to leverage it? In this paper, we enrich the state-of-the-art neural natural language inference models with external knowledge. We demonstrate that the proposed models improve neural NLI models to achieve the state-of-the-art performance on the SNLI and MultiNLI datasets.

This paper proposes hybrid semi-Markov conditional random fields (SCRFs) for neural sequence labeling in natural language processing. Based on conventional conditional random fields (CRFs), SCRFs have been designed for the tasks of assigning labels to segments by extracting features from and describing transitions between segments instead of words. In this paper, we improve the existing SCRF methods by employing word-level and segment-level information simultaneously. First, word-level labels are utilized to derive the segment scores in SCRFs. Second, a CRF output layer and an SCRF output layer are integrated into a unified neural network and trained jointly. Experimental results on CoNLL 2003 named entity recognition (NER) shared task show that our model achieves state-of-the-art performance when no external knowledge is used.

Pooling is an essential component of a wide variety of sentence representation and embedding models. This paper explores generalized pooling methods to enhance sentence embedding. We propose vector-based multi-head attention that includes the widely used max pooling, mean pooling, and scalar self-attention as special cases. The model benefits from properly designed penalization terms to reduce redundancy in multi-head attention. We evaluate the proposed model on three different tasks: natural language inference (NLI), author profiling, and sentiment classification. The experiments show that the proposed model achieves significant improvement over strong sentence-encoding-based methods, resulting in state-of-the-art performances on four datasets. The proposed approach can be easily implemented for more problems than we discuss in this paper.

The RepEval 2017 Shared Task aims to evaluate natural language understanding models for sentence representation, in which a sentence is represented as a fixed-length vector with neural networks and the quality of the representation is tested with a natural language inference task. This paper describes our system (alpha) that is ranked among the top in the Shared Task, on both the in-domain test set (obtaining a 74.9% accuracy) and on the cross-domain test set (also attaining a 74.9% accuracy), demonstrating that the model generalizes well to the cross-domain data. Our model is equipped with intra-sentence gated-attention composition which helps achieve a better performance. In addition to submitting our model to the Shared Task, we have also tested it on the Stanford Natural Language Inference (SNLI) dataset. We obtain an accuracy of 85.5%, which is the best reported result on SNLI when cross-sentence attention is not allowed, the same condition enforced in RepEval 2017.

Reasoning and inference are central to human and artificial intelligence. Modeling inference in human language is very challenging. With the availability of large annotated data (Bowman et al., 2015), it has recently become feasible to train neural network based inference models, which have shown to be very effective. In this paper, we present a new state-of-the-art result, achieving the accuracy of 88.6% on the Stanford Natural Language Inference Dataset. Unlike the previous top models that use very complicated network architectures, we first demonstrate that carefully designing sequential inference models based on chain LSTMs can outperform all previous models. Based on this, we further show that by explicitly considering recursive architectures in both local inference modeling and inference composition, we achieve additional improvement. Particularly, incorporating syntactic parsing information contributes to our best result—it further improves the performance even when added to the already very strong model.