works on

From the 2 of 6 linked papers with an AI index.

activity
20242026
collaborators

6 papers

cs.AI2026

Enhancing Small Language Models Reasoning through Knowledge Graph Grounding

Dimitrios Kelesis, Konstantinos Bougiatiotis, Georgios Paliouras

The paper proposes a neuro‑symbolic framework that equips small language models with knowledge‑graph grounding via fact extraction and RGCN‑based hints to improve multi‑hop reasoni…

cs.AI2026

Improving Molecular Property Prediction in Small Language Models Using Graph-based Tools

Konstantinos Bougiatiotis, Dimitrios Kelesis, Georgios Paliouras

The paper proposes a Context‑Augmented Prompting framework that lets small language models query a graph neural network expert for structural hints and explanatory subgraphs, impro…

cs.LG2025

Analyzing the Effect of Embedding Norms and Singular Values to Oversmoothing in Graph Neural Networks

Dimitrios Kelesis, Dimitris Fotakis, Georgios Paliouras

In this paper, we study the factors that contribute to the effect of oversmoothing in deep Graph Neural Networks (GNNs). Specifically, our analysis is based on a new metric (Mean A…

cs.IR2025

Tree-based Focused Web Crawling with Reinforcement Learning

Andreas Kontogiannis, Dimitrios Kelesis, Vasilis Pollatos +2

A focused crawler aims at discovering as many web pages and web sites relevant to a target topic as possible, while avoiding irrelevant ones. Reinforcement Learning (RL) has been a…

cs.LG2024

Reducing Oversmoothing through Informed Weight Initialization in Graph Neural Networks

Dimitrios Kelesis, Dimitris Fotakis, Georgios Paliouras

In this work, we generalize the ideas of Kaiming initialization to Graph Neural Networks (GNNs) and propose a new scheme (G-Init) that reduces oversmoothing, leading to very good r…

cs.LG2024

Partially Trained Graph Convolutional Networks Resist Oversmoothing

Dimitrios Kelesis, Dimitris Fotakis, Georgios Paliouras

In this work we investigate an observation made by Kipf \& Welling, who suggested that untrained GCNs can generate meaningful node embeddings. In particular, we investigate the eff…