Zezhong Wang

Yiming Du, Hongru Wang, Zhengyi Zhao et al.

Long-term memory plays a critical role in personal interaction, considering long-term memory can better leverage world knowledge, historical information, and preferences in dialogues. Our research introduces PerLTQA, an innovative QA dataset that combines semantic and episodic memories, including world knowledge, profiles, social relationships, events, and dialogues. This dataset is collected to investigate the use of personalized memories, focusing on social interactions and events in the QA task. PerLTQA features two types of memory and a comprehensive benchmark of 8,593 questions for 30 characters, facilitating the exploration and application of personalized memories in Large Language Models (LLMs). Based on PerLTQA, we propose a novel framework for memory integration and generation, consisting of three main components: Memory Classification, Memory Retrieval, and Memory Synthesis. We evaluate this framework using five LLMs and three retrievers. Experimental results demonstrate that BERT-based classification models significantly outperform LLMs such as ChatGLM3 and ChatGPT in the memory classification task. Furthermore, our study highlights the importance of effective memory integration in the QA task.

Boyang Xue, Hongru Wang, Rui Wang et al.

The tendency of Large Language Models (LLMs) to generate hallucinations raises concerns regarding their reliability. Therefore, confidence estimations indicating the extent of trustworthiness of the generations become essential. However, current LLM confidence estimations in languages other than English remain underexplored. This paper addresses this gap by introducing a comprehensive investigation of Multilingual Confidence estimation (MlingConf) on LLMs, focusing on both language-agnostic (LA) and language-specific (LS) tasks to explore the performance and language dominance effects of multilingual confidence estimations on different tasks. The benchmark comprises four meticulously checked and human-evaluate high-quality multilingual datasets for LA tasks and one for the LS task tailored to specific social, cultural, and geographical contexts of a language. Our experiments reveal that on LA tasks English exhibits notable linguistic dominance in confidence estimations than other languages, while on LS tasks, using question-related language to prompt LLMs demonstrates better linguistic dominance in multilingual confidence estimations. The phenomena inspire a simple yet effective native-tone prompting strategy by employing language-specific prompts for LS tasks, effectively improving LLMs' reliability and accuracy on LS tasks.

Xingyu Lan, Leni Yang, Zezhong Wang et al.

Storytelling is an ancient and precious human ability that has been rejuvenated in the digital age. Over the last decade, there has been a notable surge in the recognition and application of data storytelling, both in academia and industry. Recently, the rapid development of generative AI has brought new opportunities and challenges to this field, sparking numerous new questions. These questions may not necessarily be quickly transformed into papers, but we believe it is necessary to promptly discuss them to help the community better clarify important issues and research agendas for the future. We thus invite you to join our workshop (Gen4DS) to discuss questions such as: How can generative AI facilitate the creation of data stories? How might generative AI alter the workflow of data storytellers? What are the pitfalls and risks of incorporating AI in storytelling? We have designed both paper presentations and interactive activities (including hands-on creation, group discussion pods, and debates on controversial issues) for the workshop. We hope that participants will learn about the latest advances and pioneering work in data storytelling, engage in critical conversations with each other, and have an enjoyable, unforgettable, and meaningful experience at the event.

Zhengyi Zhao, Shubo Zhang, Zezhong Wang et al.

Long-context question-answering (LCQA) systems have greatly benefited from the powerful reasoning capabilities of large language models (LLMs), which can be categorized into slow and quick reasoning modes. However, both modes have their limitations. Slow thinking generally leans to explore every possible reasoning path, which leads to heavy overthinking and wastes time. Quick thinking usually relies on pattern matching rather than truly understanding the query logic, which misses proper understanding. To address these issues, we propose FReM: Flexible Reasoning Mechanism, a method that adjusts reasoning depth according to the complexity of each question. Specifically, FReM leverages synthetic reference QA examples to provide an explicit chain of thought, enabling efficient handling of simple queries while allowing deeper reasoning for more complex ones. By doing so, FReM helps quick-thinking models move beyond superficial pattern matching and narrows the reasoning space for slow-thinking models to avoid unnecessary exploration. Experiments on seven QA datasets show that FReM improves reasoning accuracy and scalability, particularly for complex multihop questions, indicating its potential to advance LCQA methodologies.

Wai-Chung Kwan, Huimin Wang, Hongru Wang et al.

Dialogue policy learning (DPL) is a crucial component of dialogue modelling. Its primary role is to determine the appropriate abstract response, commonly referred to as the "dialogue action". Traditional DPL methodologies have treated this as a sequential decision problem, using pre-defined action candidates extracted from a corpus. However, these incomplete candidates can significantly limit the diversity of responses and pose challenges when dealing with edge cases, which are scenarios that occur only at extreme operating parameters. To address these limitations, we introduce a novel framework, JoTR. This framework is unique as it leverages a text-to-text Transformer-based model to generate flexible dialogue actions. Unlike traditional methods, JoTR formulates a word-level policy that allows for a more dynamic and adaptable dialogue action generation, without the need for any action templates. This setting enhances the diversity of responses and improves the system's ability to handle edge cases effectively. In addition, JoTR employs reinforcement learning with a reward-shaping mechanism to efficiently finetune the word-level dialogue policy, which allows the model to learn from its interactions, improving its performance over time. We conducted an extensive evaluation of JoTR to assess its effectiveness. Our extensive evaluation shows that JoTR achieves state-of-the-art performance on two benchmark dialogue modelling tasks, as assessed by both user simulators and human evaluators.

Liang Chen, Hongru Wang, Yang Deng et al.

Generating persona consistent dialogue response is important for developing an intelligent conversational agent. Recent works typically fine-tune large-scale pre-trained models on this task by concatenating persona texts and dialogue history as a single input sequence to generate the target response. While simple and effective, our analysis shows that this popular practice is seriously affected by order sensitivity where different input orders of persona sentences significantly impact the quality and consistency of generated response, resulting in severe performance fluctuations (i.e., 29.4% on GPT2 and 83.2% on BART). To mitigate the order sensitivity problem, we propose a model-agnostic framework, ORder Insensitive Generation (ORIG), which enables dialogue models to learn robust representation under different persona orders and improve the consistency of response generation. Experiments on the Persona-Chat dataset justify the effectiveness and superiority of our method with two dominant pre-trained models (GPT2 and BART).

Hongru Wang, Rui Wang, Fei Mi et al.

Large Language Models (LLMs), such as \texttt{ChatGPT}, greatly empower dialogue systems with strong language understanding and generation capabilities. However, most of the previous works prompt the LLMs to directly generate a response based on the dialogue context, overlooking the underlying linguistic cues about the user status exhibited in the context. Such in-depth dialogue scenarios are challenging for existing LLMs to figure out the user's hidden needs and respond satisfactorily through a single-step inference. To this end, we propose a novel linguistic cue-based chain-of-thoughts (\textit{Cue}-CoT), which enhances the LLMs inference with an intermediate reasoning step to find cues exhibited in the dialogue, aiming to provide a more personalized and engaging response. To evaluate the approach, we build a benchmark with in-depth dialogue questions, consisting of 6 datasets in both Chinese and English, targeting 3 major linguistic cues during the conversation: \textit{personality}, \textit{emotion}, and \textit{psychology}. We conduct extensive experiments on the proposed benchmark with 5 LLMs under both zero-shot and one-shot settings. Empirical results demonstrate our proposed \textit{Cue}-CoT method outperforms standard prompting methods in terms of both \textit{helpfulness} and \textit{acceptability} on all datasets.

Min Li, Zhengyuan Shi, Zezhong Wang et al.

XORNet-based low power controller is a popular technique to reduce circuit transitions in scan-based testing. However, existing solutions construct the XORNet evenly for scan chain control, and it may result in sub-optimal solutions without any design guidance. In this paper, we propose a novel testability-aware low power controller with evolutionary learning. The XORNet generated from the proposed genetic algorithm (GA) enables adaptive control for scan chains according to their usages, thereby significantly improving XORNet encoding capacity, reducing the number of failure cases with ATPG and decreasing test data volume. Experimental results indicate that under the same control bits, our GA-guided XORNet design can improve the fault coverage by up to 2.11%. The proposed GA-guided XORNets also allows reducing the number of control bits, and the total testing time decreases by 20.78% on average and up to 47.09% compared to the existing design without sacrificing test coverage.

Hongru Wang, Huimin Wang, Zezhong Wang et al.

Reinforcement Learning (RL) has been witnessed its potential for training a dialogue policy agent towards maximizing the accumulated rewards given from users. However, the reward can be very sparse for it is usually only provided at the end of a dialog session, which causes unaffordable interaction requirements for an acceptable dialog agent. Distinguished from many efforts dedicated to optimizing the policy and recovering the reward alternatively which suffers from easily getting stuck in local optima and model collapse, we decompose the adversarial training into two steps: 1) we integrate a pre-trained language model as a discriminator to judge whether the current system action is good enough for the last user action (i.e., \textit{next action prediction}); 2) the discriminator gives and extra local dense reward to guide the agent's exploration. The experimental result demonstrates that our method significantly improves the complete rate (~4.4\%) and success rate (~8.0\%) of the dialogue system.

Zezhong Wang, Hongru Wang, Kwan Wai Chung et al.

Few-shot slot tagging is an emerging research topic in the field of Natural Language Understanding (NLU). With sufficient annotated data from source domains, the key challenge is how to train and adapt the model to another target domain which only has few labels. Conventional few-shot approaches use all the data from the source domains without considering inter-domain relations and implicitly assume each sample in the domain contributes equally. However, our experiments show that the data distribution bias among different domains will significantly affect the adaption performance. Moreover, transferring knowledge from dissimilar domains will even introduce some extra noises so that affect the performance of models. To tackle this problem, we propose an effective similarity-based method to select data from the source domains. In addition, we propose a Shared-Private Network (SP-Net) for the few-shot slot tagging task. The words from the same class would have some shared features. We extract those shared features from the limited annotated data on the target domain and merge them together as the label embedding to help us predict other unlabelled data on the target domain. The experiment shows that our method outperforms the state-of-the-art approaches with fewer source data. The result also proves that some training data from dissimilar sources are redundant and even negative for the adaption.

Hongru Wang, Zezhong Wang, Wai Chung Kwan et al.

Meta-learning is widely used for few-shot slot tagging in task of few-shot learning. The performance of existing methods is, however, seriously affected by \textit{sample forgetting issue}, where the model forgets the historically learned meta-training tasks while solely relying on support sets when adapting to new tasks. To overcome this predicament, we propose the \textbf{M}emory-based \textbf{C}ontrastive \textbf{M}eta-\textbf{L}earning (aka, MCML) method, including \textit{learn-from-the-memory} and \textit{adaption-from-the-memory} modules, which bridge the distribution gap between training episodes and between training and testing respectively. Specifically, the former uses an explicit memory bank to keep track of the label representations of previously trained episodes, with a contrastive constraint between the label representations in the current episode with the historical ones stored in the memory. In addition, the \emph{adaption-from-memory} mechanism is introduced to learn more accurate and robust representations based on the shift between the same labels embedded in the testing episodes and memory. Experimental results show that the MCML outperforms several state-of-the-art methods on both SNIPS and NER datasets and demonstrates strong scalability with consistent improvement when the number of shots gets greater.