Advancements in large language models (LLMs) have driven the emergence of complex new systems to provide access to information, that we will collectively refer to as modular generative information access (GenIA) systems. They integrate a broad and evolving range of specialized components, including LLMs, retrieval models, and a heterogeneous set of sources and tools. While modularity offers flexibility, it also raises critical challenges: How can we systematically characterize the space of possible modules and their interactions? How can we automate and optimize interactions among these heterogeneous components? And, how do we enable this modular system to dynamically adapt to varying user query requirements and evolving module capabilities? In this perspective paper, we argue that the architecture of future modular generative information access systems will not just assemble powerful components, but enable a self-organizing system through real-time adaptive orchestration – where components’ interactions are dynamically configured for each user input, maximizing information relevance while minimizing computational overhead. We give provisional answers to the questions raised above with a roadmap that depicts the key principles and methods for designing such an adaptive modular system. We identify pressing challenges, and propose avenues for addressing them in the years ahead. This perspective urges the IR community to rethink modular system designs for developing adaptive, self-optimizing, and future-ready architectures that evolve alongside their rapidly advancing underlying technologies.

Entity linking (EL) in conversations faces notable challenges in practical applications, primarily due to the scarcity of entity-annotated conversational datasets and sparse knowledge bases (KB) containing domain-specific, long-tail entities. Our findings reveal that previous evaluation approaches fall short of capturing real-world complexities for zero-shot EL, highlighting the necessity for new approaches to design and assess conversational EL models to adapt to limited resources. The evaluation setup and the dataset proposed in this research are made publicly available.

Developed a statistical framework with configuration models and hypothesis testing for size-invariant comparison and clustering of non-isomorphic networks, and applied graph-theoretic metrics to analyze semantic representations across machine-learned (Word2Vec, BERT, VisWord2Vec), human-curated (WordNet, ConceptNet), and hybrid (ConceptNet Numberbatch) spaces.