Zobrazeno 1 - 10
of 61
pro vyhledávání: '"Qi, Fanchao"'
Autor:
Si, Shuzheng, Zhao, Haozhe, Chen, Gang, Li, Yunshui, Luo, Kangyang, Lv, Chuancheng, An, Kaikai, Qi, Fanchao, Chang, Baobao, Sun, Maosong
The expansion of large language models to effectively handle instructions with extremely long contexts has yet to be fully investigated. The primary obstacle lies in constructing a high-quality long instruction-following dataset devised for long cont
Externí odkaz:
http://arxiv.org/abs/2410.15633
Autor:
Qin, Yujia, Cai, Zihan, Jin, Dian, Yan, Lan, Liang, Shihao, Zhu, Kunlun, Lin, Yankai, Han, Xu, Ding, Ning, Wang, Huadong, Xie, Ruobing, Qi, Fanchao, Liu, Zhiyuan, Sun, Maosong, Zhou, Jie
Long-form question answering (LFQA) aims at answering complex, open-ended questions with detailed, paragraph-length responses. The de facto paradigm of LFQA necessitates two procedures: information retrieval, which searches for relevant supporting fa
Externí odkaz:
http://arxiv.org/abs/2305.06849
Autor:
Chen, Yangyi, Gao, Hongcheng, Cui, Ganqu, Qi, Fanchao, Huang, Longtao, Liu, Zhiyuan, Sun, Maosong
Textual adversarial samples play important roles in multiple subfields of NLP research, including security, evaluation, explainability, and data augmentation. However, most work mixes all these roles, obscuring the problem definitions and research go
Externí odkaz:
http://arxiv.org/abs/2210.10683
In linguistics, a sememe is defined as the minimum semantic unit of languages. Sememe knowledge bases (KBs), which are built by manually annotating words with sememes, have been successfully applied to various NLP tasks. However, existing sememe KBs
Externí odkaz:
http://arxiv.org/abs/2203.07426
It is very common to use quotations (quotes) to make our writings more elegant or convincing. To help people find appropriate quotes efficiently, the task of quote recommendation is presented, aiming to recommend quotes that fit the current context o
Externí odkaz:
http://arxiv.org/abs/2202.13145
Autor:
Yao, Yuan, Dong, Qingxiu, Guan, Jian, Cao, Boxi, Zhang, Zhengyan, Xiao, Chaojun, Wang, Xiaozhi, Qi, Fanchao, Bao, Junwei, Nie, Jinran, Zeng, Zheni, Gu, Yuxian, Zhou, Kun, Huang, Xuancheng, Li, Wenhao, Ren, Shuhuai, Lu, Jinliang, Xu, Chengqiang, Wang, Huadong, Zeng, Guoyang, Zhou, Zile, Zhang, Jiajun, Li, Juanzi, Huang, Minlie, Yan, Rui, He, Xiaodong, Wan, Xiaojun, Zhao, Xin, Sun, Xu, Liu, Yang, Liu, Zhiyuan, Han, Xianpei, Yang, Erhong, Sui, Zhifang, Sun, Maosong
Realizing general-purpose language intelligence has been a longstanding goal for natural language processing, where standard evaluation benchmarks play a fundamental and guiding role. We argue that for general-purpose language intelligence evaluation
Externí odkaz:
http://arxiv.org/abs/2112.13610
Backdoor attacks are a kind of emergent security threat in deep learning. After being injected with a backdoor, a deep neural model will behave normally on standard inputs but give adversary-specified predictions once the input contains specific back
Externí odkaz:
http://arxiv.org/abs/2110.08247
Adversarial attacks and backdoor attacks are two common security threats that hang over deep learning. Both of them harness task-irrelevant features of data in their implementation. Text style is a feature that is naturally irrelevant to most NLP tas
Externí odkaz:
http://arxiv.org/abs/2110.07139
Autor:
Zhang, Zhengyan, Gu, Yuxian, Han, Xu, Chen, Shengqi, Xiao, Chaojun, Sun, Zhenbo, Yao, Yuan, Qi, Fanchao, Guan, Jian, Ke, Pei, Cai, Yanzheng, Zeng, Guoyang, Tan, Zhixing, Liu, Zhiyuan, Huang, Minlie, Han, Wentao, Liu, Yang, Zhu, Xiaoyan, Sun, Maosong
In recent years, the size of pre-trained language models (PLMs) has grown by leaps and bounds. However, efficiency issues of these large-scale PLMs limit their utilization in real-world scenarios. We present a suite of cost-effective techniques for t
Externí odkaz:
http://arxiv.org/abs/2106.10715
Recent studies show that neural natural language processing (NLP) models are vulnerable to backdoor attacks. Injected with backdoors, models perform normally on benign examples but produce attacker-specified predictions when the backdoor is activated
Externí odkaz:
http://arxiv.org/abs/2106.06361