arXiv subjects

Evangelos Kanoulas

Publications and source records attributed to Evangelos Kanoulas.

At least 37 records · Page 2Linked to original sources

arXiv2025

PSCon: Product Search Through Conversations

Conversational Product Search ( CPS ) systems interact with users via natural language to offer personalized and context-aware product lists. However, most existing research on CPS is limited to simulated conversations, due to the lack of a real CPS dataset driven by human-like language. Moreover, existing conversational datasets for e-commerce are constructed for a particular market or a particular language and thus can not support cross-market and multi-lingual usage. In this paper, we propose a CPS data collection protocol and create a new CPS dataset, called PSCon, which assists product search through conversations with human-like language. The dataset is collected by a coached human-human data collection protocol and is available for dual markets and two languages. By formulating the task of CPS, the dataset allows for comprehensive and in-depth research on six subtasks: user intent detection, keyword extraction, system action prediction, question selection, item ranking, and response generation. Moreover, we present a concise analysis of the dataset and propose a benchmark model on the proposed CPS dataset. Our proposed dataset and model will be helpful for facilitating future research on CPS.

cs.CL↗

arXiv2025

Leveraging Decoder Architectures for Learned Sparse Retrieval

Learned Sparse Retrieval (LSR) has traditionally focused on small-scale encoder-only transformer architectures. With the advent of large-scale pre-trained language models, their capability to generate sparse representations for retrieval tasks across different transformer-based architectures, including encoder-only, decoder-only, and encoder-decoder models, remains largely unexplored. This study investigates the effectiveness of LSR across these architectures, exploring various sparse representation heads and model scales. Our results highlight the limitations of using large language models to create effective sparse representations in zero-shot settings, identifying challenges such as inappropriate term expansions and reduced performance due to the lack of expansion. We find that the encoder-decoder architecture with multi-tokens decoding approach achieves the best performance among the three backbones. While the decoder-only model performs worse than the encoder-only model, it demonstrates the potential to outperform when scaled to a high number of parameters.

cs.IR↗

arXiv2025

Reproducing HotFlip for Corpus Poisoning Attacks in Dense Retrieval

HotFlip is a topical gradient-based word substitution method for attacking language models. Recently, this method has been further applied to attack retrieval systems by generating malicious passages that are injected into a corpus, i.e., corpus poisoning. However, HotFlip is known to be computationally inefficient, with the majority of time being spent on gradient accumulation for each query-passage pair during the adversarial token generation phase, making it impossible to generate an adequate number of adversarial passages in a reasonable amount of time. Moreover, the attack method itself assumes access to a set of user queries, a strong assumption that does not correspond to how real-world adversarial attacks are usually performed. In this paper, we first significantly boost the efficiency of HotFlip, reducing the adversarial generation process from 4 hours per document to only 15 minutes, using the same hardware. We further contribute experiments and analysis on two additional tasks: (1) transfer-based black-box attacks, and (2) query-agnostic attacks. Whenever possible, we provide comparisons between the original method and our improved version. Our experiments demonstrate that HotFlip can effectively attack a variety of dense retrievers, with an observed trend that its attack performance diminishes against more advanced and recent methods. Interestingly, we observe that while HotFlip performs poorly in a black-box setting, indicating limited capacity for generalization, in query-agnostic scenarios its performance is correlated to the volume of injected adversarial passages.

cs.IR↗

arXiv2025

Agent-centric Information Access

As large language models (LLMs) become more specialized, we envision a future where millions of expert LLMs exist, each trained on proprietary data and excelling in specific domains. In such a system, answering a query requires selecting a small subset of relevant models, querying them efficiently, and synthesizing their responses. This paper introduces a framework for agent-centric information access, where LLMs function as knowledge agents that are dynamically ranked and queried based on their demonstrated expertise. Unlike traditional document retrieval, this approach requires inferring expertise on the fly, rather than relying on static metadata or predefined model descriptions. This shift introduces several challenges, including efficient expert selection, cost-effective querying, response aggregation across multiple models, and robustness against adversarial manipulation. To address these issues, we propose a scalable evaluation framework that leverages retrieval-augmented generation and clustering techniques to construct and assess thousands of specialized models, with the potential to scale toward millions.

cs.IR↗

arXiv2025

Self-seeding and Multi-intent Self-instructing LLMs for Generating Intent-aware Information-Seeking dialogs

Identifying user intents in information-seeking dialogs is crucial for a system to meet user's information needs. Intent prediction (IP) is challenging and demands sufficient dialogs with human-labeled intents for training. However, manually annotating intents is resource-intensive. While large language models (LLMs) have been shown to be effective in generating synthetic data, there is no study on using LLMs to generate intent-aware information-seeking dialogs. In this paper, we focus on leveraging LLMs for zero-shot generation of large-scale, open-domain, and intent-aware information-seeking dialogs. We propose SOLID, which has novel self-seeding and multi-intent self-instructing schemes. The former improves the generation quality by using the LLM's own knowledge scope to initiate dialog generation; the latter prompts the LLM to generate utterances sequentially, and mitigates the need for manual prompt design by asking the LLM to autonomously adapt its prompt instruction when generating complex multi-intent utterances. Furthermore, we propose SOLID-RL, which is further trained to generate a dialog in one step on the data generated by SOLID. We propose a length-based quality estimation mechanism to assign varying weights to SOLID-generated dialogs based on their quality during the training process of SOLID-RL. We use SOLID and SOLID-RL to generate more than 300k intent-aware dialogs, surpassing the size of existing datasets. Experiments show that IP methods trained on dialogs generated by SOLID and SOLID-RL achieve better IP quality than ones trained on human-generated dialogs.

cs.CL↗

arXiv2025

Constructing Set-Compositional and Negated Representations for First-Stage Ranking

Set compositional and negated queries are crucial for expressing complex information needs and enable the discovery of niche items like Books about non-European monarchs. Despite the recent advances in LLMs, first-stage ranking remains challenging due to the requirement of encoding documents and queries independently from each other. This limitation calls for constructing compositional query representations that encapsulate logical operations or negations, and can be used to match relevant documents effectively. In the first part of this work, we explore constructing such representations in a zero-shot setting using vector operations between lexically grounded Learned Sparse Retrieval (LSR) representations. Specifically, we introduce Disentangled Negation that penalizes only the negated parts of a query, and a Combined Pseudo-Term approach that enhances LSRs ability to handle intersections. We find that our zero-shot approach is competitive and often outperforms retrievers fine-tuned on compositional data, highlighting certain limitations of LSR and Dense Retrievers. Finally, we address some of these limitations and improve LSRs representation power for negation, by allowing them to attribute negative term scores and effectively penalize documents containing the negated terms.

cs.IR↗

arXiv2025

Gradient Weight-normalized Low-rank Projection for Efficient LLM Training

Large Language Models (LLMs) have shown remarkable performance across various tasks, but the escalating demands on computational resources pose significant challenges, particularly in the extensive utilization of full fine-tuning for downstream tasks. To address this, parameter-efficient fine-tuning (PEFT) methods have been developed, but they often underperform compared to full fine-tuning and struggle with memory efficiency. In this work, we introduce Gradient Weight-Normalized Low-Rank Projection (GradNormLoRP), a novel approach that enhances both parameter and memory efficiency while maintaining comparable performance to full fine-tuning. GradNormLoRP normalizes the weight matrix to improve gradient conditioning, facilitating better convergence during optimization. Additionally, it applies low-rank approximations to the weight and gradient matrices, significantly reducing memory usage during training. Extensive experiments demonstrate that our 8-bit GradNormLoRP reduces optimizer memory usage by up to 89.5% and enables the pre-training of large LLMs, such as LLaMA 7B, on consumer-level GPUs like the NVIDIA RTX 4090, without additional inference costs. Moreover, GradNormLoRP outperforms existing low-rank methods in fine-tuning tasks. For instance, when fine-tuning the RoBERTa model on all GLUE tasks with a rank of 8, GradNormLoRP achieves an average score of 80.65, surpassing LoRA's score of 79.23. These results underscore GradNormLoRP as a promising alternative for efficient LLM pre-training and fine-tuning. Source code: https://github.com/Jhhuangkay/Gradient-Weight-normalized-Low-rank-Projection-for-Efficient-LLM-Training

cs.LG↗

arXiv2024

Knowledge-Enhanced Conversational Recommendation via Transformer-based Sequential Modelling

In conversational recommender systems (CRSs), conversations usually involve a set of items and item-related entities or attributes, e.g., director is a related entity of a movie. These items and item-related entities are often mentioned along the development of a dialog, leading to potential sequential dependencies among them. However, most of existing CRSs neglect these potential sequential dependencies. In this article, we first propose a Transformer-based sequential conversational recommendation method, named TSCR, to model the sequential dependencies in the conversations to improve CRS. In TSCR, we represent conversations by items and the item-related entities, and construct user sequences to discover user preferences by considering both the mentioned items and item-related entities. Based on the constructed sequences, we deploy a Cloze task to predict the recommended items along a sequence. Meanwhile, in certain domains, knowledge graphs formed by the items and their related entities are readily available, which provide various different kinds of associations among them. Given that TSCR does not benefit from such knowledge graphs, we then propose a knowledge graph enhanced version of TSCR, called TSCRKG. In specific, we leverage the knowledge graph to offline initialize our model TSCRKG, and augment the user sequence of conversations (i.e., sequence of the mentioned items and item-related entities in the conversation) with multi-hop paths in the knowledge graph. Experimental results demonstrate that our TSCR model significantly outperforms state-of-the-art baselines, and the enhanced version TSCRKG further improves recommendation performance on top of TSCR.

cs.IR↗

arXiv2024

Fine Tuning vs. Retrieval Augmented Generation for Less Popular Knowledge

Language Models (LMs) memorize a vast amount of factual knowledge, exhibiting strong performance across diverse tasks and domains. However, it has been observed that the performance diminishes when dealing with less-popular or low-frequency concepts and entities, for example in domain specific applications. The two prominent approaches to enhance the performance of LMs on low-frequent topics are: Retrieval Augmented Generation (RAG) and fine-tuning (FT) over synthetic data. This paper explores and evaluates the impact of RAG and FT on customizing LMs in handling low-frequency entities on question answering tasks. We conduct extensive experiments on twelve LMs of varying size and type and different fine tuning, data augmentation, and retrieval models. Our findings indicate that while FT boosts the performance across entities of varying popularity, RAG surpasses FT by a large margin particularly for least popular factual knowledge. Additionally, the success of both RAG and FT approaches is amplified by improving retrieval and data augmentation techniques. Fine tuning, while beneficial for small LMs, requires extensive resources. To address this issue, we propose the new Stimulus RAG approach that surpasses the effectiveness of fine tuning based approaches, thereby eliminating the need for the costly data augmentation and fine tuning step for enriching LMs with less popular factual knowledge. The code is available at \url{https://github.com/informagi/RAGvsFT}.

cs.CL↗

arXiv2024

Learning to Ask: Conversational Product Search via Representation Learning

Online shopping platforms, such as Amazon and AliExpress, are increasingly prevalent in society, helping customers purchase products conveniently. With recent progress in natural language processing, researchers and practitioners shift their focus from traditional product search to conversational product search. Conversational product search enables user-machine conversations and through them collects explicit user feedback that allows to actively clarify the users' product preferences. Therefore, prospective research on an intelligent shopping assistant via conversations is indispensable. Existing publications on conversational product search either model conversations independently from users, queries, and products or lead to a vocabulary mismatch. In this work, we propose a new conversational product search model, ConvPS, to assist users in locating desirable items. The model is first trained to jointly learn the semantic representations of user, query, item, and conversation via a unified generative framework. After learning these representations, they are integrated to retrieve the target items in the latent semantic space. Meanwhile, we propose a set of greedy and explore-exploit strategies to learn to ask the user a sequence of high-performance questions for conversations. Our proposed ConvPS model can naturally integrate the representation learning of the user, query, item, and conversation into a unified generative framework, which provides a promising avenue for constructing accurate and robust conversational product search systems that are flexible and adaptive. Experimental results demonstrate that our ConvPS model significantly outperforms state-of-the-art baselines.

cs.CL↗

arXiv2024

Image2Text2Image: A Novel Framework for Label-Free Evaluation of Image-to-Text Generation with Text-to-Image Diffusion Models

Evaluating the quality of automatically generated image descriptions is a complex task that requires metrics capturing various dimensions, such as grammaticality, coverage, accuracy, and truthfulness. Although human evaluation provides valuable insights, its cost and time-consuming nature pose limitations. Existing automated metrics like BLEU, ROUGE, METEOR, and CIDEr attempt to fill this gap, but they often exhibit weak correlations with human judgment. To address this challenge, we propose a novel evaluation framework called Image2Text2Image, which leverages diffusion models, such as Stable Diffusion or DALL-E, for text-to-image generation. In the Image2Text2Image framework, an input image is first processed by a selected image captioning model, chosen for evaluation, to generate a textual description. Using this generated description, a diffusion model then creates a new image. By comparing features extracted from the original and generated images, we measure their similarity using a designated similarity metric. A high similarity score suggests that the model has produced a faithful textual description, while a low score highlights discrepancies, revealing potential weaknesses in the model's performance. Notably, our framework does not rely on human-annotated reference captions, making it a valuable tool for assessing image captioning models. Extensive experiments and human evaluations validate the efficacy of our proposed Image2Text2Image evaluation framework. The code and dataset will be published to support further research in the community.

cs.CV↗

arXiv2024

The art of connections: constructing a social network from the correspondence archive of Sybren Valkema

Social network analysis allows researchers to discover insights from connections between people. While the process of building a social network is relatively straightforward for contemporary social media, deriving connections from historical archives remains a challenging task, with every data collection presenting its unique challenges. Our contribution focuses on building and analysing a social network from the correspondence archive of Sybren Valkema (1916-1996), a Dutch glass artist and educator. The archive contains both typewritten and handwritten documents in multiple languages, and includes letters from glass artists, art students, art collectors and other agents. We develop an automatic pipeline approach which includes separating handwritten and typed documents, performing text recognition specific to the document modality, extracting names of people from text using named entity recognition, de-duplicating the resulting names to create actor nodes, classifying the actors using entity linking, and, finally, connecting them together and analysing the resulting network. Every part of the pipeline is evaluated against a manual analysis performed by an art historian on a subset of the data collection in order to find out which pitfalls of the automatic approach need to be resolved in future work and, on the contrary, whether using the automatic approach allows to discover any additional insights. The results show strong performance in discovering sender-receiver connections as well as additional meaningful connections in text, with the main challenge being text recognition on scanned pages.

cs.DL↗

arXiv2024

Table Question Answering for Low-resourced Indic Languages

TableQA is the task of answering questions over tables of structured information, returning individual cells or tables as output. TableQA research has focused primarily on high-resource languages, leaving medium- and low-resource languages with little progress due to scarcity of annotated data and neural models. We address this gap by introducing a fully automatic large-scale tableQA data generation process for low-resource languages with limited budget. We incorporate our data generation method on two Indic languages, Bengali and Hindi, which have no tableQA datasets or models. TableQA models trained on our large-scale datasets outperform state-of-the-art LLMs. We further study the trained models on different aspects, including mathematical reasoning capabilities and zero-shot cross-lingual transfer. Our work is the first on low-resource tableQA focusing on scalable data generation and evaluation procedures. Our proposed data generation method can be applied to any low-resource language with a web presence. We release datasets, models, and code (https://github.com/kolk/Low-Resource-TableQA-Indic-languages).

cs.CL↗

arXiv2024

Corpus-informed Retrieval Augmented Generation of Clarifying Questions

This study aims to develop models that generate corpus informed clarifying questions for web search, in a way that ensures the questions align with the available information in the retrieval corpus. We demonstrate the effectiveness of Retrieval Augmented Language Models (RAG) in this process, emphasising their ability to (i) jointly model the user query and retrieval corpus to pinpoint the uncertainty and ask for clarifications end-to-end and (ii) model more evidence documents, which can be used towards increasing the breadth of the questions asked. However, we observe that in current datasets search intents are largely unsupported by the corpus, which is problematic both for training and evaluation. This causes question generation models to ``hallucinate'', ie. suggest intents that are not in the corpus, which can have detrimental effects in performance. To address this, we propose dataset augmentation methods that align the ground truth clarifications with the retrieval corpus. Additionally, we explore techniques to enhance the relevance of the evidence pool during inference, but find that identifying ground truth intents within the corpus remains challenging. Our analysis suggests that this challenge is partly due to the bias of current datasets towards clarification taxonomies and calls for data that can support generating corpus-informed clarifications.

cs.IR↗

arXiv2024

A Multimodal Dense Retrieval Approach for Speech-Based Open-Domain Question Answering

Speech-based open-domain question answering (QA over a large corpus of text passages with spoken questions) has emerged as an important task due to the increasing number of users interacting with QA systems via speech interfaces. Passage retrieval is a key task in speech-based open-domain QA. So far, previous works adopted pipelines consisting of an automatic speech recognition (ASR) model that transcribes the spoken question before feeding it to a dense text retriever. Such pipelines have several limitations. The need for an ASR model limits the applicability to low-resource languages and specialized domains with no annotated speech data. Furthermore, the ASR model propagates its errors to the retriever. In this work, we try to alleviate these limitations by proposing an ASR-free, end-to-end trained multimodal dense retriever that can work directly on spoken questions. Our experimental results showed that, on shorter questions, our retriever is a promising alternative to the \textit{ASR and Retriever} pipeline, achieving better retrieval performance in cases where ASR would have mistranscribed important words in the question or have produced a transcription with a high word error rate.

cs.CL↗

arXiv2024

QFMTS: Generating Query-Focused Summaries over Multi-Table Inputs

Table summarization is a crucial task aimed at condensing information from tabular data into concise and comprehensible textual summaries. However, existing approaches often fall short of adequately meeting users' information and quality requirements and tend to overlook the complexities of real-world queries. In this paper, we propose a novel method to address these limitations by introducing query-focused multi-table summarization. Our approach, which comprises a table serialization module, a summarization controller, and a large language model (LLM), utilizes textual queries and multiple tables to generate query-dependent table summaries tailored to users' information needs. To facilitate research in this area, we present a comprehensive dataset specifically tailored for this task, consisting of 4909 query-summary pairs, each associated with multiple tables. Through extensive experiments using our curated dataset, we demonstrate the effectiveness of our proposed method compared to baseline approaches. Our findings offer insights into the challenges of complex table reasoning for precise summarization, contributing to the advancement of research in query-focused multi-table summarization.

cs.CL↗

arXiv2024

Towards Fine-Grained Citation Evaluation in Generated Text: A Comparative Analysis of Faithfulness Metrics

Large language models (LLMs) often produce unsupported or unverifiable content, known as "hallucinations." To mitigate this, retrieval-augmented LLMs incorporate citations, grounding the content in verifiable sources. Despite such developments, manually assessing how well a citation supports the associated statement remains a major challenge. Previous studies use faithfulness metrics to estimate citation support automatically but are limited to binary classification, overlooking fine-grained citation support in practical scenarios. To investigate the effectiveness of faithfulness metrics in fine-grained scenarios, we propose a comparative evaluation framework that assesses the metric effectiveness in distinguishing citations between three-category support levels: full, partial, and no support. Our framework employs correlation analysis, classification evaluation, and retrieval evaluation to measure the alignment between metric scores and human judgments comprehensively. Our results show no single metric consistently excels across all evaluations, revealing the complexity of assessing fine-grained support. Based on the findings, we provide practical recommendations for developing more effective metrics.

cs.IR↗

arXiv2024

A Comparative Analysis of Faithfulness Metrics and Humans in Citation Evaluation

Large language models (LLMs) often generate content with unsupported or unverifiable content, known as "hallucinations." To address this, retrieval-augmented LLMs are employed to include citations in their content, grounding the content in verifiable sources. Despite such developments, manually assessing how well a citation supports the associated statement remains a major challenge. Previous studies tackle this challenge by leveraging faithfulness metrics to estimate citation support automatically. However, they limit this citation support estimation to a binary classification scenario, neglecting fine-grained citation support in practical scenarios. To investigate the effectiveness of faithfulness metrics in fine-grained scenarios, we propose a comparative evaluation framework that assesses the metric effectiveness in distinguishing citations between three-category support levels: full, partial, and no support. Our framework employs correlation analysis, classification evaluation, and retrieval evaluation to measure the alignment between metric scores and human judgments comprehensively. Our results indicate no single metric consistently excels across all evaluations, highlighting the complexity of accurately evaluating fine-grained support levels. Particularly, we find that the best-performing metrics struggle to distinguish partial support from full or no support. Based on these findings, we provide practical recommendations for developing more effective metrics.

cs.IR↗