Search Results for author: Xinyu Ma

Found 26 papers, 17 papers with code

MAIR: A Massive Benchmark for Evaluating Instructed Retrieval

1 code implementation14 Oct 2024 Weiwei Sun, Zhengliang Shi, Jiulong Wu, Lingyong Yan, Xinyu Ma, Yiding Liu, Min Cao, Dawei Yin, Zhaochun Ren

Recent information retrieval (IR) models are pre-trained and instruction-tuned on massive datasets and tasks, enabling them to perform well on a wide range of tasks and potentially generalize to unseen tasks with instructions.

Information Retrieval Re-Ranking +1

Parenting: Optimizing Knowledge Selection of Retrieval-Augmented Language Models with Parameter Decoupling and Tailored Tuning

no code implementations14 Oct 2024 Yongxin Xu, Ruizhe Zhang, Xinke Jiang, Yujie Feng, Yuzhen Xiao, Xinyu Ma, Runchuan Zhu, Xu Chu, Junfeng Zhao, Yasha Wang

Retrieval-Augmented Generation (RAG) offers an effective solution to the issues faced by Large Language Models (LLMs) in hallucination generation and knowledge obsolescence by incorporating externally retrieved knowledge.

Hallucination RAG +1

JAILJUDGE: A Comprehensive Jailbreak Judge Benchmark with Multi-Agent Enhanced Explanation Evaluation Framework

1 code implementation11 Oct 2024 Fan Liu, Yue Feng, Zhao Xu, Lixin Su, Xinyu Ma, Dawei Yin, Hao liu

Despite advancements in enhancing LLM safety against jailbreak attacks, evaluating LLM defenses remains a challenge, with current methods often lacking explainability and generalization to complex scenarios, leading to incomplete assessments (e. g., direct judgment without reasoning, low F1 score of GPT-4 in complex cases, bias in multilingual scenarios).

Understanding the Collapse of LLMs in Model Editing

1 code implementation17 Jun 2024 Wanli Yang, Fei Sun, Jiajun Tan, Xinyu Ma, Du Su, Dawei Yin, HuaWei Shen

Despite significant progress in model editing methods, their application in real-world scenarios remains challenging as they often cause large language models (LLMs) to collapse.

Model Editing

3AM: An Ambiguity-Aware Multi-Modal Machine Translation Dataset

1 code implementation29 Apr 2024 Xinyu Ma, Xuebo Liu, Derek F. Wong, Jun Rao, Bei Li, Liang Ding, Lidia S. Chao, DaCheng Tao, Min Zhang

Experimental results show that MMT models trained on our dataset exhibit a greater ability to exploit visual information than those trained on other MMT datasets.

Multimodal Machine Translation Sentence +2

LoRA Dropout as a Sparsity Regularizer for Overfitting Control

no code implementations15 Apr 2024 Yang Lin, Xinyu Ma, Xu Chu, Yujie Jin, Zhibang Yang, Yasha Wang, Hong Mei

We then demonstrate the theoretical mechanism of our LoRA Dropout mechanism from the perspective of sparsity regularization by providing a generalization error bound under this framework.

parameter-efficient fine-tuning

Parameter Efficient Quasi-Orthogonal Fine-Tuning via Givens Rotation

2 code implementations5 Apr 2024 Xinyu Ma, Xu Chu, Zhibang Yang, Yang Lin, Xin Gao, Junfeng Zhao

With the increasingly powerful performances and enormous scales of pretrained models, promoting parameter efficiency in fine-tuning has become a crucial need for effective and efficient adaptation to various downstream tasks.

The Butterfly Effect of Model Editing: Few Edits Can Trigger Large Language Models Collapse

1 code implementation15 Feb 2024 Wanli Yang, Fei Sun, Xinyu Ma, Xun Liu, Dawei Yin, Xueqi Cheng

In this work, we reveal a critical phenomenon: even a single edit can trigger model collapse, manifesting as significant performance degradation in various benchmark tasks.

Benchmarking Model Editing

Clustering Pseudo Language Family in Multilingual Translation Models with Fisher Information Matrix

1 code implementation5 Dec 2023 Xinyu Ma, Xuebo Liu, Min Zhang

In multilingual translation research, the comprehension and utilization of language families are of paramount importance.

Clustering Translation

Instruction Distillation Makes Large Language Models Efficient Zero-shot Rankers

1 code implementation2 Nov 2023 Weiwei Sun, Zheng Chen, Xinyu Ma, Lingyong Yan, Shuaiqiang Wang, Pengjie Ren, Zhumin Chen, Dawei Yin, Zhaochun Ren

Furthermore, our approach surpasses the performance of existing supervised methods like monoT5 and is on par with the state-of-the-art zero-shot methods.

Prompt Engineering

Domain-invariant Clinical Representation Learning by Bridging Data Distribution Shift across EMR Datasets

no code implementations11 Oct 2023 Zhongji Zhang, Yuhang Wang, Yinghao Zhu, Xinyu Ma, Tianlong Wang, Chaohe Zhang, Yasha Wang, Liantao Ma

Due to the limited information about emerging diseases, symptoms are hard to be noticed and recognized, so that the window for clinical intervention could be ignored.

Ethics Representation Learning +1

Pre-training with Aspect-Content Text Mutual Prediction for Multi-Aspect Dense Retrieval

1 code implementation22 Aug 2023 Xiaojie Sun, Keping Bi, Jiafeng Guo, Xinyu Ma, Fan Yixing, Hongyu Shan, Qishen Zhang, Zhongyi Liu

Extensive experiments on two real-world datasets (product and mini-program search) show that our approach can outperform competitive baselines both treating aspect values as classes and conducting the same MLM for aspect and content strings.

Language Modelling Masked Language Modeling +1

Fused Gromov-Wasserstein Graph Mixup for Graph-level Classifications

1 code implementation NeurIPS 2023 Xinyu Ma, Xu Chu, Yasha Wang, Yang Lin, Junfeng Zhao, Liantao Ma, Wenwu Zhu

To solve this problem, we propose a novel graph mixup algorithm called FGWMixup, which seeks a midpoint of source graphs in the Fused Gromov-Wasserstein (FGW) metric space.

Data Augmentation

A Contrastive Pre-training Approach to Learn Discriminative Autoencoder for Dense Retrieval

no code implementations21 Aug 2022 Xinyu Ma, Ruqing Zhang, Jiafeng Guo, Yixing Fan, Xueqi Cheng

Empirical results show that our method can significantly outperform the state-of-the-art autoencoder-based language models and other pre-trained models for dense retrieval.

Decoder Information Retrieval +1

Scattered or Connected? An Optimized Parameter-efficient Tuning Approach for Information Retrieval

no code implementations21 Aug 2022 Xinyu Ma, Jiafeng Guo, Ruqing Zhang, Yixing Fan, Xueqi Cheng

Unlike the promising results in NLP, we find that these methods cannot achieve comparable performance to full fine-tuning at both stages when updating less than 1\% of the original model parameters.

Information Retrieval Re-Ranking +1

Pre-train a Discriminative Text Encoder for Dense Retrieval via Contrastive Span Prediction

1 code implementation22 Apr 2022 Xinyu Ma, Jiafeng Guo, Ruqing Zhang, Yixing Fan, Xueqi Cheng

% Therefore, in this work, we propose to drop out the decoder and introduce a novel contrastive span prediction task to pre-train the encoder alone.

Contrastive Learning Decoder +3

MedFACT: Modeling Medical Feature Correlations in Patient Health Representation Learning via Feature Clustering

no code implementations21 Apr 2022 Xinyu Ma, Xu Chu, Yasha Wang, Hailong Yu, Liantao Ma, Wen Tang, Junfeng Zhao

Thus, to address the issues, we expect to group up strongly correlated features and learn feature correlations in a group-wise manner to reduce the learning complexity without losing generality.

Clustering Representation Learning

Pre-training Methods in Information Retrieval

no code implementations27 Nov 2021 Yixing Fan, Xiaohui Xie, Yinqiong Cai, Jia Chen, Xinyu Ma, Xiangsheng Li, Ruqing Zhang, Jiafeng Guo

The core of information retrieval (IR) is to identify relevant information from large-scale resources and return it as a ranked list to respond to the user's information need.

Information Retrieval Re-Ranking +1

B-PROP: Bootstrapped Pre-training with Representative Words Prediction for Ad-hoc Retrieval

1 code implementation20 Apr 2021 Xinyu Ma, Jiafeng Guo, Ruqing Zhang, Yixing Fan, Yingyan Li, Xueqi Cheng

The basic idea of PROP is to construct the \textit{representative words prediction} (ROP) task for pre-training inspired by the query likelihood model.

Information Retrieval Language Modelling +1

A Linguistic Study on Relevance Modeling in Information Retrieval

no code implementations1 Mar 2021 Yixing Fan, Jiafeng Guo, Xinyu Ma, Ruqing Zhang, Yanyan Lan, Xueqi Cheng

We employ 16 linguistic tasks to probe a unified retrieval model over these three retrieval tasks to answer this question.

Information Retrieval Natural Language Understanding +2

PROP: Pre-training with Representative Words Prediction for Ad-hoc Retrieval

1 code implementation20 Oct 2020 Xinyu Ma, Jiafeng Guo, Ruqing Zhang, Yixing Fan, Xiang Ji, Xueqi Cheng

Recently pre-trained language representation models such as BERT have shown great success when fine-tuned on downstream tasks including information retrieval (IR).

Information Retrieval Language Modelling +1

AdaCare: Explainable Clinical Health Status Representation Learning via Scale-Adaptive Feature Extraction and Recalibration

1 code implementation27 Nov 2019 Liantao Ma, Junyi Gao, Yasha Wang, Chaohe Zhang, Jiangtao Wang, Wenjie Ruan, Wen Tang, Xin Gao, Xinyu Ma

It also models the correlation between clinical features to enhance the ones which strongly indicate the health status and thus can maintain a state-of-the-art performance in terms of prediction accuracy while providing qualitative interpretability.

Representation Learning

ConCare: Personalized Clinical Feature Embedding via Capturing the Healthcare Context

1 code implementation27 Nov 2019 Liantao Ma, Chaohe Zhang, Yasha Wang, Wenjie Ruan, Jiantao Wang, Wen Tang, Xinyu Ma, Xin Gao, Junyi Gao

Predicting the patient's clinical outcome from the historical electronic medical records (EMR) is a fundamental research problem in medical informatics.

Cannot find the paper you are looking for? You can Submit a new open access paper.