Abstract:This paper describes our participation in STC-2 Chinese subtask of NTCIR-13. All runs are submitted for both two tasks, namely generation-based task and retrieval-based task. Various methods based on Seq2Seq framework were used to generate the responses. Interaction-focused method based on deep learning models is used to deal with relevance between queries and comments. As for generation-based task, we propose to add a constraint term to the original objective function of Seq2Seq, which further incorporates the coherent between post and reply into consideration. Specifically, three different approaches have been adopted to measure the similarity between post and reply, i.e. unlearned constraint such as cosine function and pretrained matching function such as bilinear function and MatchPyramid. As for information retrieval task, we apply a method to obtain relevance that is crucial to retrieval. As called interactionfocused method, we apply method based on MatchPyramid, which first builds interactions between a query and a comment and then uses a deep model to obtain the representation for the interactions and the relevance score.

ISCAS at NTCIR-3: Monolingual, Bilingual and MultiLingual IR Tasks

Cross-Lingual Text Image Recognition Via Multi-Task Sequence to Sequence Learning.

ISCAS: Text Retrieval in NTCIR Workshop II.

Research On English-Chinese Bi-Directional Cross-Language Information Retrieval

Simple Yet Effective Neural Ranking and Reranking Baselines for Cross-Lingual Information Retrieval

Research on Lucene-based English-Chinese Cross-Language Information Retrieval

HLTCOE at TREC 2023 NeuCLIR Track

Research on Chinese-English Cross-Language Information Retrieval

SG01 at the NTCIR-13 STC-2 Task

THUIR at NTCIR-10 INTENT-2 Task

THUIR at the NTCIR-14 WWW-2 Task.

Microsoft Research Asia at NTCIR-12 STC Task.

Overview of the TREC 2023 NeuCLIR Track

THUIR at NTCIR-9 INTENT Task.

The ICT , CAS MT Systems for the IWSLT 09 Evaluation

Dictionary-based Method for English-Chinese Bidirectional Cross-Language Information Retrieval

THUIR at NTCIR-12 IMine Task.

Gbot at the NTCIR-13 STC-2 Task Hainan

English-Chinese Cross-language Information Retrieval Using Lucene System

Distillation for Multilingual Information Retrieval

Bilingual query translation and expansion for supporting more effective cross-language image retrieval.