Wai Lam
YOU?
Author Swipe
View article: LNE-Blocking: An Efficient Framework for Contamination Mitigation Evaluation on Large Language Models
LNE-Blocking: An Efficient Framework for Contamination Mitigation Evaluation on Large Language Models Open
The problem of data contamination is now almost inevitable during the development of large language models (LLMs), with the training data commonly integrating those evaluation benchmarks even unintentionally. This problem subsequently make…
View article: Proactive Conversational AI: A Comprehensive Survey of Advancements and Opportunities
Proactive Conversational AI: A Comprehensive Survey of Advancements and Opportunities Open
Dialogue systems are designed to offer human users social support or functional services through natural language interactions. Traditional conversation research has put significant emphasis on a system’s response-ability, including its ca…
View article: LLM2: Let Large Language Models Harness System 2 Reasoning
LLM2: Let Large Language Models Harness System 2 Reasoning Open
Large language models (LLMs) have exhibited impressive capabilities across a myriad of tasks, yet they occasionally yield undesirable outputs. We posit that these limitations are rooted in the foundational autoregressive architecture of LL…
View article: Stephanie: Step-by-Step Dialogues for Mimicking Human Interactions in Social Conversations
Stephanie: Step-by-Step Dialogues for Mimicking Human Interactions in Social Conversations Open
In the rapidly evolving field of natural language processing, dialogue systems primarily employ a single-step dialogue paradigm. Although this paradigm is efficient, it lacks the depth and fluidity of human interactions and does not appear…
View article: Not All Preference Pairs Are Created Equal: A Recipe for Annotation-Efficient Iterative Preference Learning
Not All Preference Pairs Are Created Equal: A Recipe for Annotation-Efficient Iterative Preference Learning Open
Iterative preference learning, though yielding superior performances, requires online annotated preference labels. In this work, we study strategies to select worth-annotating response pairs for cost-efficient annotation while achieving co…
View article: A Comprehensive Survey on Relation Extraction: Recent Advances and New Frontiers
A Comprehensive Survey on Relation Extraction: Recent Advances and New Frontiers Open
Relation extraction (RE) involves identifying the relations between entities from underlying content. RE serves as the foundation for many natural language processing (NLP) and information retrieval applications, such as knowledge graph co…
View article: Unveiling the Generalization Power of Fine-Tuned Large Language Models
Unveiling the Generalization Power of Fine-Tuned Large Language Models Open
While Large Language Models (LLMs) have demonstrated exceptional multitasking abilities, fine-tuning these models on downstream, domain-specific datasets is often necessary to yield superior performance on test sets compared to their count…
View article: CodeAttack: Revealing Safety Generalization Challenges of Large Language Models via Code Completion
CodeAttack: Revealing Safety Generalization Challenges of Large Language Models via Code Completion Open
The rapid advancement of Large Language Models (LLMs) has brought about remarkable generative capabilities but also raised concerns about their potential misuse. While strategies like supervised fine-tuning and reinforcement learning from …
View article: Plug-and-Play Policy Planner for Large Language Model Powered Dialogue Agents
Plug-and-Play Policy Planner for Large Language Model Powered Dialogue Agents Open
Proactive dialogues serve as a practical yet challenging dialogue problem in the era of large language models (LLMs), where the dialogue policy planning is the key to improving the proactivity of LLMs. Most existing studies enable the dial…
View article: Once Upon a $\textit{Time}$ in $\textit{Graph}$: Relative-Time Pretraining for Complex Temporal Reasoning
Once Upon a $\textit{Time}$ in $\textit{Graph}$: Relative-Time Pretraining for Complex Temporal Reasoning Open
Our physical world is constantly evolving over time, rendering challenges for pre-trained language models to understand and reason over the temporal contexts of texts. Existing work focuses on strengthening the direct association between a…
View article: JsonTuning: Towards Generalizable, Robust, and Controllable Instruction Tuning
JsonTuning: Towards Generalizable, Robust, and Controllable Instruction Tuning Open
Instruction tuning is vital for enhancing the performance of large language models (LLMs), but existing text-to-text methods, referred to as TextTuning, struggle with issues such as generalization, robustness, and controllability due to th…
View article: Social Media Fashion Knowledge Extraction as Captioning
Social Media Fashion Knowledge Extraction as Captioning Open
Social media plays a significant role in boosting the fashion industry, where a massive amount of fashion-related posts are generated every day. In order to obtain the rich fashion information from the posts, we study the task of social me…
View article: EPA: Easy Prompt Augmentation on Large Language Models via Multiple Sources and Multiple Targets
EPA: Easy Prompt Augmentation on Large Language Models via Multiple Sources and Multiple Targets Open
Large language models (LLMs) have shown promising performance on various NLP tasks via task prompting. And their performance can be further improved by appending task demonstrations to the head of the prompt. And usually, a better performa…
View article: On the Effectiveness of Parameter-Efficient Fine-Tuning
On the Effectiveness of Parameter-Efficient Fine-Tuning Open
Fine-tuning pre-trained models has been ubiquitously proven to be effective in a wide range of NLP tasks. However, fine-tuning the whole model is parameter inefficient as it always yields an entirely new model for each task. Currently, man…
View article: A Comprehensive Survey on Relation Extraction: Recent Advances and New Frontiers
A Comprehensive Survey on Relation Extraction: Recent Advances and New Frontiers Open
Relation extraction (RE) involves identifying the relations between entities from underlying content. RE serves as the foundation for many natural language processing (NLP) and information retrieval applications, such as knowledge graph co…
View article: Enhancing Grammatical Error Correction Systems with Explanations
Enhancing Grammatical Error Correction Systems with Explanations Open
Grammatical error correction systems improve written communication by detecting and correcting language mistakes. To help language learners better understand why the GEC system makes a certain correction, the causes of errors (evidence wor…
View article: mPMR: A Multilingual Pre-trained Machine Reader at Scale
mPMR: A Multilingual Pre-trained Machine Reader at Scale Open
We present multilingual Pre-trained Machine Reader (mPMR), a novel method for multilingual machine reading comprehension (MRC)-style pre-training. mPMR aims to guide multilingual pre-trained language models (mPLMs) to perform natural langu…
View article: A Frustratingly Simple Decoding Method for Neural Text Generation
A Frustratingly Simple Decoding Method for Neural Text Generation Open
We introduce a frustratingly simple, super efficient and surprisingly effective decoding method, which we call Frustratingly Simple Decoding (FSD), for neural text generation. The idea behind FSD is straightforward: we build an anti-LM bas…
View article: Chain-of-Symbol Prompting Elicits Planning in Large Langauge Models
Chain-of-Symbol Prompting Elicits Planning in Large Langauge Models Open
In this paper, we take the initiative to investigate the performance of LLMs on complex planning tasks that require LLMs to understand a virtual spatial environment simulated via natural language and act correspondingly in text. We propose…
View article: Knowledge-enhanced Mixed-initiative Dialogue System for Emotional Support Conversations
Knowledge-enhanced Mixed-initiative Dialogue System for Emotional Support Conversations Open
Unlike empathetic dialogues, the system in emotional support conversations (ESC) is expected to not only convey empathy for comforting the help-seeker, but also proactively assist in exploring and addressing their problems during the conve…
View article: Easy-to-Hard Learning for Information Extraction
Easy-to-Hard Learning for Information Extraction Open
Information extraction (IE) systems aim to automatically extract structured information, such as named entities, relations between entities, and events, from unstructured texts. While most existing work addresses a particular IE task, univ…