THUIR at the NTCIR-14 WWW-2 Task

Zheng, Yukun; Chu, Zhumin; Li, Xiangsheng; Mao, Jiaxin; Liu, Yiqun; Zhang, Min; Ma, Shaoping

doi:10.1007/978-3-030-36805-0_13

Cited by 4 publications

(3 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Learning-to-rank, a popular machine learning method for ranking, has been widely used in information retrieval and data mining [12,15]. In this section, we implement several learning-to-rank methods with diverse features.…”

Section: Learning To Rankmentioning

confidence: 99%

Towards Better Web Search Performance: Pre-training, Fine-tuning and Learning to Rank

Li¹,

Chen²,

Su³

et al. 2023

Preprint

View full text Add to dashboard Cite

This paper describes the approach of the THUIR team at the WSDM Cup 2023 Pre-training for Web Search task. This task requires the participant to rank the relevant documents for each query. We propose a new data pre-processing method and conduct pre-training and fine-tuning with the processed data. Moreover, we extract statistical, axiomatic, and semantic features to enhance the ranking performance. After the feature extraction, diverse learning-to-rank models are employed to merge those features. The experimental results show the superiority of our proposal. We finally achieve second place in this competition. CCS CONCEPTS• Information systems → Retrieval models and ranking; Information retrieval.

show abstract

Section: Learning To Rankmentioning

confidence: 99%

Towards Better Web Search Performance: Pre-training, Fine-tuning and Learning to Rank

Li¹,

Chen²,

Su³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…MAP nDCG@10 2019 TREC BM25 0.237 0.517 ucas_runid1 [13] 0.264 0.644 TUW19-d3-re [35] 0.271 0.644 idst_bert_r1 [100] 0 Model nDCG@10 Q@10 nERR@10 BM25 0.5748. 0.5850 0.6757 Technion-E-CO-NEW-1 [81] 0.6581 0.6815 0.7791 KASYS-E-CO-NEW-1 [89] 0.6935 0.7123 0.7959 PARADE-Max 0.6337 0.6556 0.7395 PARADE-Transformer 0.6897 0.7016 0.8090 structBERT [95], which strengthens on the task of sentence order prediction. All PARADE variants outperform ucas_runid1 and TUW19-d3-re in terms of nDCG@10, but cannot outperform idst_bert_r1.…”

Section: Year Group Runidmentioning

confidence: 99%

PARADE: Passage Representation Aggregation forDocument Reranking

Yates

MacAvaney

et al. 2023

ACM Trans. Inf. Syst.

View full text Add to dashboard Cite

Pre-trained transformer models, such as BERT and T5, have shown to be highly effective at ad-hoc passage and document ranking. Due to the inherent sequence length limits of these models, they need to process document passages one at a time rather than processing the entire document sequence at once. Although several approaches for aggregating passage-level signals into a document-level relevance score have been proposed, there has yet to be an extensive comparison of these techniques. In this work, we explore strategies for aggregating relevance signals from a document’s passages into a final ranking score. We find that passage representation aggregation techniques can significantly improve over score aggregation techniques proposed in prior work, such as taking the maximum passage score. We call this new approach PARADE. In particular, PARADE can significantly improve results on collections with broad information needs where relevance signals can be spread throughout the document (such as TREC Robust04 and GOV2). Meanwhile, less complex aggregation techniques may work better on collections with an information need that can often be pinpointed to a single passage (such as TREC DL and TREC Genomics). We also conduct efficiency analyses and highlight several strategies for improving transformer-based aggregation.

show abstract

“…However, state-of-the-art retrieval systems are usually constructed based on neural models trained with large-scale annotated data. Hence, IR researchers propose to utilize pre-trained language models (PLM), i.e., large-scale neural models trained without supervised data for language understanding, to conduct effective retrieval [13,17,36,43]. Previous studies [7,10,16,42] have shown that PLM such as BERT [9] and RoBERTa [18] significantly outperform existing neural retrieval models on passage and document retrieval datasets like MS MARCO and TREC DL in both zero-shot and few-shot settings [8,25,35].…”

Section: Introductionmentioning

confidence: 99%

Building Chinese interdisciplinary research centers : the case of Tsinghua University

Li¹,

李蓉暉²

View full text Add to dashboard Cite

Graph machine learning has been extensively studied in both academia and industry. However, in the literature, most existing graph machine learning models are designed to conduct training with data samples in a random order, which may suffer from suboptimal performance due to ignoring the importance of different graph data samples and their training orders for the model optimization status. To tackle this critical problem, curriculum graph machine learning (Graph CL), which integrates the strength of graph machine learning and curriculum learning, arises and attracts an increasing amount of attention from the research community. Therefore, in this paper, we comprehensively overview approaches on Graph CL and present a detailed survey of recent advances in this direction. Specifically, we first discuss the key challenges of Graph CL and provide its formal problem definition. Then, we categorize and summarize existing methods into three classes based on three kinds of graph machine learning tasks, i.e., node-level, linklevel, and graph-level tasks. Finally, we share our thoughts on future research directions. To the best of our knowledge, this paper is the first survey for curriculum graph machine learning.

show abstract

THUIR at the NTCIR-14 WWW-2 Task

Cited by 4 publications

References 31 publications

Towards Better Web Search Performance: Pre-training, Fine-tuning and Learning to Rank

Towards Better Web Search Performance: Pre-training, Fine-tuning and Learning to Rank

PARADE: Passage Representation Aggregation forDocument Reranking

Building Chinese interdisciplinary research centers : the case of Tsinghua University

Contact Info

Product

Resources

About