40 citations · 76 across the 33 of their papers we have counts for
33 papers
EpiCoDe: Boosting Model Performance Beyond Training with Extrapolation and Contrastive Decoding
Mingxu Tao, Jie Hu, Mingchuan Yang +3
The remarkable performance of Large language models (LLMs) relies heavily on the availability of abundant high-quality training data. However, the high cost of acquiring annotated…
MiLiC-Eval: Benchmarking Multilingual LLMs for China's Minority Languages
Chen Zhang, Mingxu Tao, Zhiyuan Liao +1
Large language models (LLMs) excel in high-resource languages but struggle with low-resource languages (LRLs), particularly those spoken by minority communities in China, such as T…
Chain of Condition: Construct, Verify and Solve Conditions for Conditional Question Answering
Jiuheng Lin, Yuxuan Lai, Yansong Feng
Conditional question answering (CQA) is an important task that aims to find probable answers and identify missing conditions. Existing approaches struggle with CQA due to two chall…
Unlocking the Potential of Model Merging for Low-Resource Languages
Mingxu Tao, Chen Zhang, Quzhe Huang +4
Adapting large language models (LLMs) to new languages typically involves continual pre-training (CT) followed by supervised fine-tuning (SFT). However, this CT-then-SFT approach s…
What Kinds of Tokens Benefit from Distant Text? An Analysis on Long Context Language Modeling
Yutong Hu, Quzhe Huang, Kangcheng Luo +1
As the context length that large language models can handle continues to increase, these models demonstrate an enhanced ability to utilize distant information for tasks such as lan…
Can Perplexity Reflect Large Language Model's Ability in Long Text Understanding?
Yutong Hu, Quzhe Huang, Mingxu Tao +2
Recent studies have shown that Large Language Models (LLMs) have the potential to process extremely long text. Many works only evaluate LLMs' long-text processing ability on the la…
Harder Tasks Need More Experts: Dynamic Routing in MoE Models
Quzhe Huang, Zhenwei An, Nan Zhuang +7
In this paper, we introduce a novel dynamic expert selection framework for Mixture of Experts (MoE) models, aiming to enhance computational efficiency and model performance by adju…
Probing Multimodal Large Language Models for Global and Local Semantic Representations
Mingxu Tao, Quzhe Huang, Kun Xu +3
The advancement of Multimodal Large Language Models (MLLMs) has greatly accelerated the development of applications in understanding integrated texts and images. Recent works lever…
Chain-of-Discussion: A Multi-Model Framework for Complex Evidence-Based Question Answering
Mingxu Tao, Dongyan Zhao, Yansong Feng
Open-ended question answering requires models to find appropriate evidence to form wellreasoned, comprehensive and helpful answers. In practical applications, models also need to e…
MC: Towards Transparent and Culturally-Aware NLP for Minority Languages in China
Chen Zhang, Mingxu Tao, Quzhe Huang +3
Current large language models demonstrate deficiencies in understanding low-resource languages, particularly the minority languages in China. This limitation stems from the scarcit…
From the One, Judge of the Whole: Typed Entailment Graph Construction with Predicate Generation
Zhibin Chen, Yansong Feng, Dongyan Zhao
Entailment Graphs (EGs) have been constructed based on extracted corpora as a strong and explainable form to indicate context-independent entailment relations in natural languages.…
How Many Answers Should I Give? An Empirical Study of Multi-Answer Reading Comprehension
Chen Zhang, Jiuheng Lin, Xiao Liu +3
The multi-answer phenomenon, where a question may have multiple answers scattered in the document, can be well handled by humans but is challenging enough for machine reading compr…
The Magic of IF: Investigating Causal Reasoning Abilities in Large Language Models of Code
Xiao Liu, Da Yin, Chen Zhang +2
Causal reasoning, the ability to identify cause-and-effect relationship, is crucial in human thinking. Although large language models (LLMs) succeed in many NLP tasks, it is still…
More than Classification: A Unified Framework for Event Temporal Relation Extraction
Quzhe Huang, Yutong Hu, Shengqi Zhu +3
Event temporal relation extraction~(ETRE) is usually formulated as a multi-label classification task, where each type of relation is simply treated as a one-hot label. This formula…
A Frustratingly Easy Improvement for Position Embeddings via Random Padding
Mingxu Tao, Yansong Feng, Dongyan Zhao
Position embeddings, encoding the positional relationships among tokens in text sequences, make great contributions to modeling local context features in Transformer-based pre-trai…
Lawyer LLaMA Technical Report
Quzhe Huang, Mingxu Tao, Chen Zhang +5
Large Language Models (LLMs), like LLaMA, have exhibited remarkable performance across various tasks. Nevertheless, when deployed to specific domains such as law or medicine, the m…
Generalized Implicit Factorization Problem
Yansong Feng, Abderrahmane Nitaj, Yanbin Pan
The Implicit Factorization Problem was first introduced by May and Ritzenhofen at PKC'09. This problem aims to factorize two RSA moduli and when their pri…
Can BERT Refrain from Forgetting on Sequential Tasks? A Probing Study
Mingxu Tao, Yansong Feng, Dongyan Zhao
Large pre-trained language models help to achieve state of the art on a variety of natural language processing (NLP) tasks, nevertheless, they still suffer from forgetting when inc…
Cross-Lingual Question Answering over Knowledge Base as Reading Comprehension
Chen Zhang, Yuxuan Lai, Yansong Feng +3
Although many large-scale knowledge bases (KBs) claim to contain multilingual information, their support for many non-English languages is often incomplete. This incompleteness giv…
Do Charge Prediction Models Learn Legal Theory?
Zhenwei An, Quzhe Huang, Cong Jiang +2
The charge prediction task aims to predict the charge for a case given its fact description. Recent models have already achieved impressive accuracy in this task, however, little i…
Counterfactual Recipe Generation: Exploring Compositional Generalization in a Realistic Scenario
Xiao Liu, Yansong Feng, Jizhi Tang +2
People can acquire knowledge in an unsupervised manner by reading, and compose the knowledge to make novel combinations. In this paper, we investigate whether pretrained language m…
Improve Discourse Dependency Parsing with Contextualized Representations
Yifei Zhou, Yansong Feng
Recent works show that discourse analysis benefits from modeling intra- and inter-sentential levels separately, where proper representations for text units of different granulariti…
Things not Written in Text: Exploring Spatial Commonsense from Visual Signals
Xiao Liu, Da Yin, Yansong Feng +1
Spatial commonsense, the knowledge about spatial position and relationship between objects (like the relative size of a lion and a girl, and the position of a boy relative to a bic…
Does Recommend-Revise Produce Reliable Annotations? An Analysis on Missing Instances in DocRED
Quzhe Huang, Shibo Hao, Yuan Ye +3
DocRED is a widely used dataset for document-level relation extraction. In the large-scale annotation, a \textit{recommend-revise} scheme is adopted to reduce the workload. Within…