Li Lin

Tsinghua University | TH · School of Software

With the growth of the depth of neural networks and the scale of data, the difficulty of network training also increases. When the GPU memory is insufficient, it is challenging to train deeper models. Recent research uses tensor swapping and recomputation techniques in a combined manner to optimize memory usage. However, complex dependencies and en...

Natural Language Inference (NLI) is a growingly essential task in natural language understanding, which requires inferring the relationship between the sentence pairs ( premise and hypothesis ). Recently, low-resource natural language inference has gained increasing attention, due to significant savings in manual annotation costs and a better fi...

We propose the first character-level white-box adversarial attack method against transformer models. The intuition of our method comes from the observation that words are split into subtokens before being fed into the transformer models and the substitution between two close subtokens has a similar effect to the character modification. Our method m...

The generalizability to new databases is of vital importance to Text-to-SQL systems which aim to parse human utterances into SQL statements. Existing works achieve this goal by leveraging the exact matching method to identify the lexical matching between the question words and the schema items. However, these methods fail in other challenging scena...

Natural Language Inference (NLI) is a growingly essential task in natural language understanding, which requires inferring the relationship between the sentence pairs (premise and hypothesis). Recently, low-resource natural language inference has gained increasing attention, due to significant savings in manual annotation costs and a better fit wit...

Natural language inference (NLI) is an increasingly important task for natural language understanding, which requires one to infer the relationship between the sentence pair (premise and hypothesis). Many recent works have used contrastive learning by incorporating the relationship of the sentence pair from NLI datasets to learn sentence representa...

Future Event Generation aims to generate fluent and reasonable future event descriptions given preceding events. It requires not only fluent text generation but also commonsense reasoning to maintain the coherence of the entire event story. However, existing FEG methods are easily trapped into repeated or general events without imposing any logical...

Low-resource Relation Extraction (LRE) aims to extract relation facts from limited labeled corpora when human annotation is scarce. Existing works either utilize self-training scheme to generate pseudo labels that will cause the gradual drift problem, or leverage meta-learning scheme which does not solicit feedback explicitly. To alleviate selectio...

Sequence prediction is a well-defined problem with a proliferation of applications, such as recommendation systems, social media monitor, economic analysis, etc. Recently, RNN-based methodologies have shown their superiority in time-series data analysis and sequence modeling. The question of which event would happen next is not difficult to answer...

Distributed in-memory computing frameworks usually have lots of parameters (e.g., the buffer size of shuffle) to form a configuration for each execution. A well-tuned configuration can bring large improvements of performance. However, to improve resource utilization, jobs are often share the same cluster, which causes dynamic cluster load condition...

Flexibility and evolution have been a hot topic in the context of business process management. Business process drift detection is a family of methods to detect changes by analyzing the event log, but existing methods have some disadvantages in dealing with concept drifts. First, most of these methods detect changes depending on an exploration of a...

Text classification in low-resource languages (e.g., Thai) is of great practical value for some information retrieval applications (e.g., sentiment-analysis-based restaurant recommendation). Due to lacking large-scale corpus for learning comprehensive text representation , bilingual text classification which borrows the linguistics knowledge from a...

Sequential Text Classification (STC) aims to classify a sequence of text fragments (e.g., words in a sentence or sentences in a document) into a sequence of labels. In addition to the intra-fragment text contents, considering the inter-fragment context dependencies is also important for STC. Previous sequence labeling approaches largely generate a...

Process model extraction (PME) is a recently emerged in-terdiscipline between natural language processing (NLP) and business process management (BPM), which aims to extract process models from textual descriptions. Previous process extractors heavily depend on manual features and ignore the potential relations between clues of different text granul...

Process model extraction (PME) is a recently emerged interdiscipline between natural language processing (NLP) and business process management (BPM), which aims to extract process models from textual descriptions. Previous process extractors heavily depend on manual features and ignore the potential relations between clues of different text granula...

Sequential Text Classification (STC) aims to classify a sequence of text fragments (e.g., words in a sentence or sentences in a document) into a sequence of labels. In addition to the intra-fragment text contents, considering the inter-fragment context dependencies is also important for STC. Previous sequence labeling approaches largely generate a...

Event sequence prediction has wide applications on economics , electronic health and social media monitoring. Accurate prediction of event sequences can help provide better service to customers and prevent risks. Recent works try to address the problem aiming at learning the impact of past events on the future events using deep learning methods. Su...

Most traditional business process recommendation methods cannot deal with complex structures such as interacting loops, and they cannot handle large complex datasets with a great quantity of processes and activities. To address these issues, RLRecommender, a method based on representation learning, is proposed. RLRecommender extracts three kinds of...