Papers by Kyusong Lee
Talk to Papers: Bringing Neural Question Answering to Academic Search (2020.acl-demos)
Copied to clipboard
| Challenge: | Talk to Papers aims to improve the current experience of academic search by using open-domain question answering (QA) techniques. |
| Approach: | They propose to use open-domain question answering techniques to improve the current experience of academic search by combining natural language queries with machine reading at scale. |
| Outcome: | The proposed tool improves on existing search engines and provides a collaborative data collection tool to curate the first natural language processing research QA dataset. |
SPARTA: Efficient Open-Domain Question Answering via Sparse Transformer Matching Retrieval (2021.naacl-main)
Copied to clipboard
| Challenge: | SPARTA is a novel neural retrieval method for open-domain question answering . it learns a sparse representation that can be efficiently implemented as an Inverted Index . |
| Approach: | They propose a method that learns a sparse representation that can be implemented as an Inverted Index. |
| Outcome: | The proposed method achieves state-of-the-art results on 4 open-domain question answering tasks and 11 retrieval question answering (ReQA) tasks. |
SF-QA: Simple and Fair Evaluation Library for Open-domain Question Answering (2021.eacl-demos)
Copied to clipboard
| Challenge: | Open-domain question answering (QA) requires large amounts of resources and is difficult to reproduce results due to complex configurations. |
| Approach: | They propose a simple and fair evaluation framework for open-domain question answering (QA) it modularizes the pipeline open- domain QA system, making it easily accessible . |
| Outcome: | The proposed evaluation framework is publicly available and anyone can contribute to the code and evaluations. |
Unsupervised Discrete Sentence Representation Learning for Interpretable Neural Dialog Generation (P18-1)
Copied to clipboard
| Challenge: | Existing encoder-decoder dialog models cannot output interpretable actions as in traditional systems. |
| Approach: | They propose an unsupervised discrete sentence representation learning method that integrates with existing encoder-decoder dialog models for interpretable response generation. |
| Outcome: | The proposed model can be integrated with existing encoder-decoder dialog models and discover interpretable semantics via either auto encoding or context predicting. |
Unifying Language Agent Algorithms with Graph-based Orchestration Engine for Reproducible Agent Research (2025.acl-demo)
Copied to clipboard
Qianqian Zhang, Jiajia Liao, Heting Ying, Yibo Ma, Haozhan Shen, Jingcheng Li, Peng Liu, Lu Zhang, Chunxin Fang, Kyusong Lee, Ruochen Xu, Tiancheng Zhao
| Challenge: | Language agents powered by large language models (LLMs) have demonstrated remarkable capabilities in understanding, reasoning, and executing complex tasks. |
| Approach: | They propose a flexible framework that addresses engineering overhead and insufficient evaluation frameworks for fair comparison. |
| Outcome: | The proposed framework simplifies language agent development and establishes a foundation for reproducible agent research. |
OmAgent: A Multi-modal Agent Framework for Complex Video Understanding with Task Divide-and-Conquer (2024.emnlp-main)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have expanded their capabilities to multimodal contexts, including comprehensive video understanding. |
| Approach: | They propose to store and retrieve relevant video frames for specific queries and a Divide-and-Conquer loop capable of autonomous reasoning. |
| Outcome: | The proposed model efficiently stores and retrieves relevant video frames for specific queries, preserving the detailed content of videos. |
An Explainable Toolbox for Evaluating Pre-trained Vision-Language Models (2022.emnlp-demos)
Copied to clipboard
| Challenge: | Existing studies evaluate VLP models by comparing the fine-tuned downstream task performance with the average downstream task accuracy. |
| Approach: | They propose a toolbox for evaluating Vision-Language Pretraining (VLP) models. |
| Outcome: | The proposed toolbox provides the preliminary datasets that deepen the image-texting ability of a VLP model. |
VisualSparta: An Embarrassingly Simple Approach to Large-scale Text-to-Image Search with Weighted Bag-of-words (2021.acl-long)
Copied to clipboard
| Challenge: | Existing text-to-image retrieval models lack good representations for textual and visual modalities. |
| Approach: | They propose a novel text-to-image retrieval model that uses a transformer to match images with textual queries. |
| Outcome: | The proposed model outperforms state-of-the-art models in MSCOCO and Flickr30K . it achieves substantial speed advantages for a 1 million image index . |