Temporal Density-aware Sequential Recommendation Networks with Contrastive Learning

Wang, Jihu; Shi, Yuliang; Yu, Han; Zhang, Kun; Wang, Xinjun; Zhou, Yan; Li, Hui

doi:10.1016/j.eswa.2022.118563

Cited by 14 publications

(10 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, the emergence of large language models (LLMs), such as GPT-3.5, have garnered significant attention from both the research community and industry. These LLMs have shown remarkable capabilities, achieving human-level performance in benchmark datasets, including SDS [17] and cross-language summarization (CLS) [18]. The introduction of LLMs has sparked a growing interest in harnessing the potential of LLMs for various applications.…”

Section: Contradictionsmentioning

confidence: 99%

“…This is because the summaries generated by LLMs-based models can sometimes surpass the quality of the standard reference summaries within the dataset. Although some meticulously curated summary datasets are available to address the aforementioned issues, most of them consist of single documents [17]. Consequently, there remains a shortage of curated datasets specifically designed for MDS.…”

Section: Contradictionsmentioning

confidence: 99%

“…For summarization tasks, LLMs also show strong performance [17,18]. For example, in single-document summarization, Goyal et al [19] compared the summaries generated by GPT-3 with traditional fine-tuning methods.…”

Section: Large Langauge Modelsmentioning

confidence: 99%

“…Recently, COT has significantly improved the inferential performance and interpretability of LLMs by decomposing multi-step problems into intermediate steps [20,36,37]. Inspired by this, Wang et al [17] proposed an element-aware summarization method to elicit LLMs to generate summaries step by step. In cross-language summarization (CLS), Wang et al [18] pioneered the exploration of LLMs' zero-shot CLS performance.…”

Section: Large Langauge Modelsmentioning

confidence: 99%

“…To meet the first criterion, we refer to the method proposed by Wang et al [17] and the "Lasswell Communication Model" theory to design a prompt to extract key elements of the news, including entities, data, events, and the result of events from each document. To satisfy the second criterion, we design a prompt to generate key summary sentences from each document based on the above key elements.…”

Section: Stage 1: Attend-prompt For Each Documentmentioning

confidence: 99%

See 4 more Smart Citations

3A-COT: An Attend-Arrange-Abstract Chain-of-Thought for Multi-Document Summarization

Zhang,

Gao,

Huang

et al. 2024

Preprint

View full text Add to dashboard Cite

Large language models (LLMs) have shown significant promise in single-document summarization (SDS). However, there has been limited exploration in multi-document summarization (MDS). The primary challenge in MDS lies in effectively navigating the intricate and complex relationships between multiple documents, including issues such as contradictions, redundancies, and complementarities. Taking inspiration from chain-of-thought (CoT) and how human approaches to MDS, this paper proposes a novel MDS method called Attend-Arrange-Abstract CoT (3A-COT). The method comprises three critical stages. Attend: To alleviate redundancies between multiple documents, key information should be focused first. Therefore, we design an Attend-prompt to extract key elements from each document and generate key information based on these key elements. Arrange: Considering the contradictions and complementarities between these key information, we design an Arrange-prompt to arrange the relationships and connections between these key information across multiple documents to form reasonable logic relationships. Abstract: Finally, an Abstract-prompt is designed to guide the generation of a summary based on the arranged key information. To evaluate 3A-COT, we conduct experiments on four MDS test sets: Multi-News, Human-edited Multi-News, WCEP-10, and Multi-XScience. Experimental results show that 3A-COT outperforms other LLM-based MDS methods. Furthermore, in-depth analysis experiments are conducted to evaluate 3A-COT's efficacy in handling contradictions, redundancies, and complementarities.

show abstract

Section: Contradictionsmentioning

confidence: 99%

Section: Contradictionsmentioning

confidence: 99%

Section: Large Langauge Modelsmentioning

confidence: 99%

Section: Large Langauge Modelsmentioning

confidence: 99%

Section: Stage 1: Attend-prompt For Each Documentmentioning

confidence: 99%

See 3 more Smart Citations