Export Citations
Save this search
Please login to be able to save your searches and receive alerts for new content matching your search criteria.
- short-paperOctober 2024
Reliable Knowledge Graph Reasoning with Uncertainty Quantification
CIKM '24: Proceedings of the 33rd ACM International Conference on Information and Knowledge ManagementPages 5463–5466https://doi.org/10.1145/3627673.3680266Recently, Knowledge Graphs (KGs) have been successfully coupled with Large Language Models (LLMs) to mitigate their hallucinations and enhance their reasoning capability, e.g., KG-based retrieval-augmented framework for question-answering. However, ...
- short-paperOctober 2024
In Situ Answer Sentence Selection at Web-scale
CIKM '24: Proceedings of the 33rd ACM International Conference on Information and Knowledge ManagementPages 4298–4302https://doi.org/10.1145/3627673.3679946Current answer sentence selection (AS2) applied in open-domain question answering (ODQA) selects answers by ranking a large set of candidates, i.e., sentences, extracted from the retrieved text. In this paper, we present Passage-based Extracting Answer ...
- research-articleOctober 2024
A GAIL Fine-Tuned LLM Enhanced Framework for Low-Resource Knowledge Graph Question Answering
CIKM '24: Proceedings of the 33rd ACM International Conference on Information and Knowledge ManagementPages 3300–3309https://doi.org/10.1145/3627673.3679753Recent studies on knowledge graph question answering (KGQA) have focused on tackling complex inquiries to enhance the applicability of models in real-life settings. Unfortunately, KGQA models encounter significant challenges due to the lack of high-...
- research-articleOctober 2024
Retrieval-enhanced Knowledge Editing in Language Models for Multi-Hop Question Answering
CIKM '24: Proceedings of the 33rd ACM International Conference on Information and Knowledge ManagementPages 2056–2066https://doi.org/10.1145/3627673.3679722Large Language Models (LLMs) have shown proficiency in question-answering tasks but often struggle to integrate real-time knowledge, leading to potentially outdated or inaccurate responses. This problem becomes even more challenging when dealing with ...
- research-articleAugust 2024
On Early Detection of Hallucinations in Factual Question Answering
KDD '24: Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data MiningPages 2721–2732https://doi.org/10.1145/3637528.3671796While large language models (LLMs) have taken great strides towards helping humans with a plethora of tasks, hallucinations remain a major impediment towards gaining user trust. The fluency and coherence of model generations even when hallucinating makes ...
-
- research-articleAugust 2024
Sponsored Question Answering
ICTIR '24: Proceedings of the 2024 ACM SIGIR International Conference on Theory of Information RetrievalPages 167–173https://doi.org/10.1145/3664190.3672517The potential move from search to question answering (QA) ignited the question of how should the move from sponsored search to sponsored QA look like. We present the first formal analysis of a sponsored QA platform. The platform fuses an organic answer ...
- short-paperJuly 2024
Retrieval-Augmented Generation with Knowledge Graphs for Customer Service Question Answering
SIGIR '24: Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information RetrievalPages 2905–2909https://doi.org/10.1145/3626772.3661370In customer service technical support, swiftly and accurately retrieving relevant past issues is critical for efficiently resolving customer inquiries. The conventional retrieval methods in retrieval-augmented generation (RAG) for large language models (...
- research-articleJuly 2024
ChroniclingAmericaQA: A Large-scale Question Answering Dataset based on Historical American Newspaper Pages
SIGIR '24: Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information RetrievalPages 2038–2048https://doi.org/10.1145/3626772.3657891Question answering (QA) and Machine Reading Comprehension (MRC) tasks have significantly advanced in recent years due to the rapid development of deep learning techniques and, more recently, large language models. At the same time, many benchmark ...
- research-articleJuly 2024
TriviaHG: A Dataset for Automatic Hint Generation from Factoid Questions
SIGIR '24: Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information RetrievalPages 2060–2070https://doi.org/10.1145/3626772.3657855Nowadays, individuals tend to engage in dialogues with Large Language Models, seeking answers to their questions. In times when such answers are readily accessible to anyone, the stimulation and preservation of human's cognitive abilities, as well as the ...
- research-articleJuly 2024
CIQA: A Coding Inspired Question Answering Model
SIGIR '24: Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information RetrievalPages 1973–1983https://doi.org/10.1145/3626772.3657830Methods in question-answering (QA) that transform texts detailing processes into an intermediate code representation, subsequently executed to generate a response to the presented question, have demonstrated promising results in analyzing scientific ...
- research-articleJuly 2024
IM-RAG: Multi-Round Retrieval-Augmented Generation Through Learning Inner Monologues
SIGIR '24: Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information RetrievalPages 730–740https://doi.org/10.1145/3626772.3657760Although the Retrieval-Augmented Generation (RAG) paradigms can use external knowledge to enhance and ground the outputs of Large Language Models (LLMs) to mitigate generative hallucinations and static knowledge base problems, they still suffer from ...
- short-paperJuly 2024
TextData: Save What You Know and Find What You Don't
SIGIR '24: Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information RetrievalPages 2806–2810https://doi.org/10.1145/3626772.3657681In this demonstration, we present TextData, a novel online system that enables users to both "save what they know" and "find what they don't". TextData was developed based on the Community Digital Library (CDL) system. Although the CDL allowed users to ...
- short-paperJuly 2024
Towards Robust QA Evaluation via Open LLMs
SIGIR '24: Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information RetrievalPages 2811–2816https://doi.org/10.1145/3626772.3657675Instruction-tuned large language models (LLMs) have been shown to be viable surrogates for the widely used, albeit overly rigid, lexical matching metrics in evaluating question answering (QA) models. However, these LLM-based evaluation methods are ...
- short-paperJuly 2024
A Question-Answering Assistant over Personal Knowledge Graph
SIGIR '24: Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information RetrievalPages 2708–2712https://doi.org/10.1145/3626772.3657665We develop a Personal Knowledge Graph Question-Answering (PKGQA) assistant, seamlessly integrating information from multiple mobile applications into a unified and user-friendly query interface to offer users convenient information retrieval and ...
- ArticleSeptember 2024
Konstruktor: A Strong Baseline for Simple Knowledge Graph Question Answering
Natural Language Processing and Information SystemsPages 107–118https://doi.org/10.1007/978-3-031-70242-6_11AbstractWhile being one of the most popular question types, simple questions such as “Who is the author of Cinderella?”, are still not completely solved. Surprisingly, even most powerful modern Large Language Models (LLMs) are prone to errors when dealing ...
- research-articleJune 2024
MyEachtraX: Lifelog Question Answering on Mobile
LSC '24: Proceedings of the 7th Annual ACM Workshop on the Lifelog Search ChallengePages 93–98https://doi.org/10.1145/3643489.3661128Your whole life in your pocket. That is the premise of lifelogging, a technology that captures and stores every moment of your life in digital form. Built on top of MyEachtra and the lifelog question-answering pipeline, MyEachtraX is a mobile-based ...
- short-paperJune 2024
Rethinking Table Retrieval from Data Lakes
aiDM '24: Proceedings of the Seventh International Workshop on Exploiting Artificial Intelligence Techniques for Data ManagementArticle No.: 2, Pages 1–5https://doi.org/10.1145/3663742.3663972Table retrieval from data lakes has recently become important for many downstream tasks, including data discovery and table question answering. Existing table retrieval approaches estimate each table's relevance to a particular information need and ...
- keynoteJune 2024
The Journey to a Knowledgeable Assistant with Retrieval-Augmented Generation (RAG)
SIGMOD/PODS '24: Companion of the 2024 International Conference on Management of DataPage 3https://doi.org/10.1145/3626246.3655999For decades, multiple communities (Database, Information Retrieval, Natural Language Processing, Data Mining, AI) have pursued the mission of providing the right information at the right time. Efforts span web search, data integration, knowledge graphs, ...
- research-articleJune 2024
QAVidCap: Enhancing Video Captioning through Question Answering Techniques
ICMR '24: Proceedings of the 2024 International Conference on Multimedia RetrievalPages 155–164https://doi.org/10.1145/3652583.3658061Video captioning is the task of describing video content using natural sentences. While recent models have shown significant improvements in metrics, there are still some unresolved issues. Model-generated captions often contain factual errors and omit ...
- research-articleJune 2024
Dynamic Segmentation for Efficient Retrieval of Podcasts: The Repping Algorithm
ICMR '24: Proceedings of the 2024 International Conference on Multimedia RetrievalPages 29–36https://doi.org/10.1145/3652583.3658047In the following article, we present a method that makes it possible to find specific segments in a podcast from a large collection using a query (keywords or question). What differentiates our method is that there is no segmentation process at the ...