9 papers
Big Claims, Small Margins: Do Reported GraphLLM Gains Survive Simple Baselines?
Michal Podstawski
LLM-based methods for node classification on text-attributed graphs (TAGs) routinely report state-of-the-art accuracy. We ask how these results compare with inexpensive baselines r…
SLM-Conditioned Hierarchical Relation Routing for Labeled Property Graph Learning
Michal Podstawski
Labeled property graphs combine relational structure with heterogeneous textual and categorical properties attached to both nodes and relationships. Conventional graph neural netwo…
Closed-Loop Graph Algorithm Execution with Small Language Models: Step Accuracy and Rollout Reliability
Michal Podstawski
Small language models offer an efficient alternative to large-scale systems, but their ability to execute structured algorithms over multiple dependent decisions remains poorly und…
Graph Property Inference in Small Language Models: Effects of Representation and Reasoning Strategy
Michal Podstawski
Recent progress in language modeling has expanded the range of tasks that can be approached through natural language interfaces, including problems that require structured reasonin…
Generalization Boundaries of Fine-Tuned Small Language Models for Graph Structural Inference
Michal Podstawski
Small language models fine-tuned for graph property estimation have demonstrated strong in-distribution performance, yet their generalization capabilities beyond training condition…
TinyGraphEstimator: Adapting Lightweight Language Models for Graph Structure Inference
Michal Podstawski
Graphs provide a universal framework for representing complex relational systems, and inferring their structural properties is a core challenge in graph analysis and reasoning. Whi…
Applying Text Embedding Models for Efficient Analysis in Labeled Property Graphs
Michal Podstawski
Labeled property graphs often contain rich textual attributes that can enhance analytical tasks when properly leveraged. This work explores the use of pretrained text embedding mod…
Face Consistency Benchmark for GenAI Video
Michal Podstawski, Malgorzata Kudelska, Haohong Wang
Video generation driven by artificial intelligence has advanced significantly, enabling the creation of dynamic and realistic content. However, maintaining character consistency ac…
EdgeWisePersona: A Dataset for On-Device User Profiling from Natural Language Interactions
Patryk Bartkowiak, Michal Podstawski
This paper introduces a novel dataset and evaluation benchmark designed to assess and improve small language models deployable on edge devices, with a focus on user profiling from…