arXiv subjects

Yannis Katsis

Publications and source records attributed to Yannis Katsis.

16 recordsLinked to original sources

arXiv2026

MTRAG-UN: A Benchmark for Open Challenges in Multi-Turn RAG Conversations

We present MTRAG-UN, a benchmark for exploring open challenges in multi-turn retrieval augmented generation, a popular use of large language models. We release a benchmark of 666 tasks containing over 2,800 conversation turns across 6 domains with accompanying corpora. Our experiments show that retrieval and generation models continue to struggle on conversations with UNanswerable, UNderspecified, and NONstandalone questions and UNclear responses. Our benchmark is available at https://github.com/IBM/mt-rag-benchmark

cs.CL↗

arXiv2025

A Longitudinal Study on Different Annotator Feedback Loops in Complex RAG Tasks

Grounding conversations in existing passages, known as Retrieval-Augmented Generation (RAG), is an important aspect of Chat-Based Assistants powered by Large Language Models (LLMs) to ensure they are faithful and don't provide misinformation. Several benchmarks have been created to measure the performance of LLMs on this task. We present a longitudinal study comparing the feedback loop of an internal and external human annotator group for the complex annotation task of creating multi-turn RAG conversations for evaluating LLMs. We analyze the conversations produced by both groups and provide results of a survey comparing their experiences. Our study highlights the advantages of each annotator population and the impact of the different feedback loops; a closer loop creates higher quality conversations with a decrease in quantity and diversity. Further, we present guidance for how to best utilize two different population groups when performing annotation tasks, particularly when the task is complex.

cs.HC↗

arXiv2025

RAGAPHENE: A RAG Annotation Platform with Human Enhancements and Edits

Retrieval Augmented Generation (RAG) is an important aspect of conversing with Large Language Models (LLMs) when factually correct information is important. LLMs may provide answers that appear correct, but could contain hallucinated information. Thus, building benchmarks that can evaluate LLMs on multi-turn RAG conversations has become an increasingly important task. Simulating real-world conversations is vital for producing high quality evaluation benchmarks. We present RAGAPHENE, a chat-based annotation platform that enables annotators to simulate real-world conversations for benchmarking and evaluating LLMs. RAGAPHENE has been successfully used by approximately 40 annotators to build thousands of real-world conversations.

cs.CL↗

arXiv2025

A Library of LLM Intrinsics for Retrieval-Augmented Generation

In the developer community for large language models (LLMs), there is not yet a clean pattern analogous to a software library, to support very large scale collaboration. Even for the commonplace use case of Retrieval-Augmented Generation (RAG), it is not currently possible to write a RAG application against a well-defined set of APIs that are agreed upon by different LLM providers. Inspired by the idea of compiler intrinsics, we propose some elements of such a concept through introducing a library of LLM Intrinsics for RAG. An LLM intrinsic is defined as a capability that can be invoked through a well-defined API that is reasonably stable and independent of how the LLM intrinsic itself is implemented. The intrinsics in our library are released as LoRA adapters on HuggingFace, and through a software interface with clear structured input/output characteristics on top of vLLM as an inference platform, accompanied in both places with documentation and code. This article describes the intended usage, training details, and evaluations for each intrinsic, as well as compositions of multiple intrinsics.

cs.AI↗

arXiv2025

InspectorRAGet: An Introspection Platform for RAG Evaluation

Large Language Models (LLM) have become a popular approach for implementing Retrieval Augmented Generation (RAG) systems, and a significant amount of effort has been spent on building good models and metrics. In spite of increased recognition of the need for rigorous evaluation of RAG systems, few tools exist that go beyond the creation of model output and automatic calculation. We present InspectorRAGet, an introspection platform for performing a comprehensive analysis of the quality of RAG system output. InspectorRAGet allows the user to analyze aggregate and instance-level performance of RAG systems, using both human and algorithmic metrics as well as annotator quality. InspectorRAGet is suitable for multiple use cases and is available publicly to the community. A live instance of the platform is available at https://ibm.biz/InspectorRAGet.

cs.SE↗

arXiv2025

MTRAG: A Multi-Turn Conversational Benchmark for Evaluating Retrieval-Augmented Generation Systems

Retrieval-augmented generation (RAG) has recently become a very popular task for Large Language Models (LLMs). Evaluating them on multi-turn RAG conversations, where the system is asked to generate a response to a question in the context of a preceding conversation is an important and often overlooked task with several additional challenges. We present MTRAG: an end-to-end human-generated multi-turn RAG benchmark that reflects several real-world properties across diverse dimensions for evaluating the full RAG pipeline. MTRAG contains 110 conversations averaging 7.7 turns each across four domains for a total of 842 tasks. We also explore automation paths via synthetic data and LLM-as-a-Judge evaluation. Our human and automatic evaluations show that even state-of-the-art LLM RAG systems struggle on MTRAG. We demonstrate the need for strong retrieval and generation systems that can handle later turns, unanswerable questions, non-standalone questions, and multiple domains. MTRAG is available at https://github.com/ibm/mt-rag-benchmark.

cs.CL↗

arXiv2024

Leveraging Large Language Models to Enhance Domain Expert Inclusion in Data Science Workflows

Domain experts can play a crucial role in guiding data scientists to optimize machine learning models while ensuring contextual relevance for downstream use. However, in current workflows, such collaboration is challenging due to differing expertise, abstract documentation practices, and lack of access and visibility into low-level implementation artifacts. To address these challenges and enable domain expert participation, we introduce CellSync, a collaboration framework comprising (1) a Jupyter Notebook extension that continuously tracks changes to dataframes and model metrics and (2) a Large Language Model powered visualization dashboard that makes those changes interpretable to domain experts. Through CellSync's cell-level dataset visualization with code summaries, domain experts can interactively examine how individual data and modeling operations impact different data segments. The chat features enable data-centric conversations and targeted feedback to data scientists. Our preliminary evaluation shows that CellSync provides transparency and promotes critical discussions about the intents and implications of data operations.

cs.HC↗

arXiv2023

Beyond Labels: Empowering Human Annotators with Natural Language Explanations through a Novel Active-Learning Architecture

Real-world domain experts (e.g., doctors) rarely annotate only a decision label in their day-to-day workflow without providing explanations. Yet, existing low-resource learning techniques, such as Active Learning (AL), that aim to support human annotators mostly focus on the label while neglecting the natural language explanation of a data point. This work proposes a novel AL architecture to support experts' real-world need for label and explanation annotations in low-resource scenarios. Our AL architecture leverages an explanation-generation model to produce explanations guided by human explanations, a prediction model that utilizes generated explanations toward prediction faithfully, and a novel data diversity-based AL sampling strategy that benefits from the explanation annotations. Automated and human evaluations demonstrate the effectiveness of incorporating explanations into AL sampling and the improved human annotation efficiency and trustworthiness with our AL architecture. Additional ablation studies illustrate the potential of our AL architecture for transfer learning, generalizability, and integration with large language models (LLMs). While LLMs exhibit exceptional explanation-generation capabilities for relatively simple tasks, their effectiveness in complex real-world tasks warrants further in-depth study.

cs.CL↗

arXiv2022

Label Sleuth: From Unlabeled Text to a Classifier in a Few Hours

Text classification can be useful in many real-world scenarios, saving a lot of time for end users. However, building a custom classifier typically requires coding skills and ML knowledge, which poses a significant barrier for many potential users. To lift this barrier, we introduce Label Sleuth, a free open source system for labeling and creating text classifiers. This system is unique for (a) being a no-code system, making NLP accessible to non-experts, (b) guiding users through the entire labeling process until they obtain a custom classifier, making the process efficient -- from cold start to classifier in a few hours, and (c) being open for configuration and extension by developers. By open sourcing Label Sleuth we hope to build a community of users and developers that will broaden the utilization of NLP models.

cs.CL↗

arXiv2022

SPOT: Knowledge-Enhanced Language Representations for Information Extraction

Knowledge-enhanced pre-trained models for language representation have been shown to be more effective in knowledge base construction tasks (i.e.,~relation extraction) than language models such as BERT. These knowledge-enhanced language models incorporate knowledge into pre-training to generate representations of entities or relationships. However, existing methods typically represent each entity with a separate embedding. As a result, these methods struggle to represent out-of-vocabulary entities and a large amount of parameters, on top of their underlying token models (i.e.,~the transformer), must be used and the number of entities that can be handled is limited in practice due to memory constraints. Moreover, existing models still struggle to represent entities and relationships simultaneously. To address these problems, we propose a new pre-trained model that learns representations of both entities and relationships from token spans and span pairs in the text respectively. By encoding spans efficiently with span modules, our model can represent both entities and their relationships but requires fewer parameters than existing models. We pre-trained our model with the knowledge graph extracted from Wikipedia and test it on a broad range of supervised and unsupervised information extraction tasks. Results show that our model learns better representations for both entities and relationships than baselines, while in supervised settings, fine-tuning our model outperforms RoBERTa consistently and achieves competitive results on information extraction tasks.

cs.CL↗

arXiv2022

Theoretical Rule-based Knowledge Graph Reasoning by Connectivity Dependency Discovery

Discovering precise and interpretable rules from knowledge graphs is regarded as an essential challenge, which can improve the performances of many downstream tasks and even provide new ways to approach some Natural Language Processing research topics. In this paper, we present a fundamental theory for rule-based knowledge graph reasoning, based on which the connectivity dependencies in the graph are captured via multiple rule types. It is the first time for some of these rule types in a knowledge graph to be considered. Based on these rule types, our theory can provide precise interpretations to unknown triples. Then, we implement our theory by what we call the RuleDict model. Results show that our RuleDict model not only provides precise rules to interpret new triples, but also achieves state-of-the-art performances on one benchmark knowledge graph completion task, and is competitive on other tasks.

cs.CL↗

arXiv2021

Abstractified Multi-instance Learning (AMIL) for Biomedical Relation Extraction

Relation extraction in the biomedical domain is a challenging task due to a lack of labeled data and a long-tail distribution of fact triples. Many works leverage distant supervision which automatically generates labeled data by pairing a knowledge graph with raw textual data. Distant supervision produces noisy labels and requires additional techniques, such as multi-instance learning (MIL), to denoise the training signal. However, MIL requires multiple instances of data and struggles with very long-tail datasets such as those found in the biomedical domain. In this work, we propose a novel reformulation of MIL for biomedical relation extraction that abstractifies biomedical entities into their corresponding semantic types. By grouping entities by types, we are better able to take advantage of the benefits of MIL and further denoise the training signal. We show this reformulation, which we refer to as abstractified multi-instance learning (AMIL), improves performance in biomedical relationship extraction. We also propose a novel relationship embedding architecture that further improves model performance.

cs.CL↗

arXiv2021

AIT-QA: Question Answering Dataset over Complex Tables in the Airline Industry

Recent advances in transformers have enabled Table Question Answering (Table QA) systems to achieve high accuracy and SOTA results on open domain datasets like WikiTableQuestions and WikiSQL. Such transformers are frequently pre-trained on open-domain content such as Wikipedia, where they effectively encode questions and corresponding tables from Wikipedia as seen in Table QA dataset. However, web tables in Wikipedia are notably flat in their layout, with the first row as the sole column header. The layout lends to a relational view of tables where each row is a tuple. Whereas, tables in domain-specific business or scientific documents often have a much more complex layout, including hierarchical row and column headers, in addition to having specialized vocabulary terms from that domain. To address this problem, we introduce the domain-specific Table QA dataset AIT-QA (Airline Industry Table QA). The dataset consists of 515 questions authored by human annotators on 116 tables extracted from public U.S. SEC filings (publicly available at: https://www.sec.gov/edgar.shtml) of major airline companies for the fiscal years 2017-2019. We also provide annotations pertaining to the nature of questions, marking those that require hierarchical headers, domain-specific terminology, and paraphrased forms. Our zero-shot baseline evaluation of three transformer-based SOTA Table QA methods - TaPAS (end-to-end), TaBERT (semantic parsing-based), and RCI (row-column encoding-based) - clearly exposes the limitation of these methods in this practical setting, with the best accuracy at just 51.8\% (RCI). We also present pragmatic table preprocessing steps used to pivot and project these complex tables into a layout suitable for the SOTA Table QA models.

cs.CL↗

arXiv2020

A Survey of the State of Explainable AI for Natural Language Processing

Recent years have seen important advances in the quality of state-of-the-art models, but this has come at the expense of models becoming less interpretable. This survey presents an overview of the current state of Explainable AI (XAI), considered within the domain of Natural Language Processing (NLP). We discuss the main categorization of explanations, as well as the various ways explanations can be arrived at and visualized. We detail the operations and explainability techniques currently available for generating explanations for NLP model predictions, to serve as a resource for model developers in the community. Finally, we point out the current gaps and encourage directions for future work in this important research area.

cs.CL↗

arXiv2020

CORD-19: The COVID-19 Open Research Dataset

The COVID-19 Open Research Dataset (CORD-19) is a growing resource of scientific papers on COVID-19 and related historical coronavirus research. CORD-19 is designed to facilitate the development of text mining and information retrieval systems over its rich collection of metadata and structured full text papers. Since its release, CORD-19 has been downloaded over 200K times and has served as the basis of many COVID-19 text mining and discovery systems. In this article, we describe the mechanics of dataset construction, highlighting challenges and key design decisions, provide an overview of how CORD-19 has been used, and describe several shared tasks built around the dataset. We hope this resource will continue to bring together the computing community, biomedical experts, and policy makers in the search for effective treatments and management policies for COVID-19.

cs.DL↗

arXiv2017

Efficient Approximate Query Answering over Sensor Data with Deterministic Error Guarantees

With the recent proliferation of sensor data, there is an increasing need for the efficient evaluation of analytical queries over multiple sensor datasets. The magnitude of such datasets makes exact query answering infeasible, leading researchers into the development of approximate query answering approaches. However, existing approximate query answering algorithms are not suited for the efficient processing of queries over sensor data, as they exhibit at least one of the following shortcomings: (a) They do not provide deterministic error guarantees, resorting to weaker probabilistic error guarantees that are in many cases not acceptable, (b) they allow queries only over a single dataset, thus not supporting the multitude of queries over multiple datasets that appear in practice, such as correlation or cross-correlation and (c) they support relational data in general and thus miss speedup opportunities created by the special nature of sensor data, which are not random but follow a typically smooth underlying phenomenon. To address these problems, we propose PlatoDB; a system that exploits the nature of sensor data to compress them and provide efficient processing of queries over multiple sensor datasets, while providing deterministic error guarantees. PlatoDB achieves the above through a novel architecture that (a) at data import time pre-processes each dataset, creating for it an intermediate hierarchical data structure that provides a hierarchy of summarizations of the dataset together with appropriate error measures and (b) at query processing time leverages the pre-computed data structures to compute an approximate answer and deterministic error guarantees for ad hoc queries even when these combine multiple datasets.

cs.DB↗