Publications

Research publications

Reports and notes written alongside the research programmes. DOIs and external citations are recorded only once formally issued.

TECHNICAL REPORT2026toap-tr-2026-01

UACP-HMA: A Universal Communication Protocol for Heterogeneous Multi-Agent AI Systems

Kongkiat Hirankerd

We describe a transport-agnostic message contract enabling models from different vendors and runtimes to cooperate on a shared task graph, with capability negotiation and provenance preservation across hand-offs.

Multi-Agent AIProtocolInteroperability
Related project
TECHNICAL REPORT2026toap-tr-2026-02

Drone Dataset TH: Data Collection Methodology for Tropical Aerial Detection

Kongkiat Hirankerd

A description of the capture protocol, AI-assisted labeling workflow and mandatory human verification stage used to build a versioned drone detection dataset for Thai conditions.

DatasetComputer VisionDrone Detection
Related project
WHITE PAPER2026toap-wp-2026-03

Sovereign AI Infrastructure for Thai Institutions

Kongkiat Hirankerd

A position paper on operating capable language models on locally owned GPU infrastructure, covering routing policy, quantisation trade-offs and data-residency requirements.

Local AIData SovereigntyGPU
Related project
RESEARCH NOTE2026toap-rn-2026-04

Pose-Conditioned Strike Recognition under Clinch Occlusion

Kongkiat Hirankerd

Preliminary observations on failure modes of temporal action recognition when fighter skeletons overlap during clinch sequences in Muay Thai footage.

Action RecognitionPose EstimationSports Analytics
Related project
TECHNICAL REPORT2025toap-tr-2025-05

Grounded Retrieval for Field Reconnaissance Decision Support

Kongkiat Hirankerd

An architecture for citation-grounded visual identification and hazard assessment support, scoped strictly to search, verification, classification, assessment and documentation.

AI for SafetyRetrievalDecision Support
Related project
PREPRINT2025toap-pp-2025-06

Simulation-Grounded Tutoring in Virtual Science Laboratories

Kongkiat Hirankerd

We outline a tutoring architecture in which the language model reads live simulation state, and describe the pilot design intended to evaluate conceptual gain.

AI for EducationSimulationTutoring
Related project