AI-Powered Interactive Legal Role-Playing Tool for Courtroom Advocacy Training

This paper presents an AI-powered interactive legal role-playing tool designed to simulate courtroom advocacy while continuously evaluating the legal quality of learner arguments. The system implements a two-stage Fast Brain and Smart Brain audit pipeline, where a fine-tuned BERT-based model assesses legal density and Gemini generates plain-English explanatory feedback. The platform generates Sri Lankan courtroom scenarios, manages five AI personas, and maintains autonomous dialogue through a heartbeat-driven engine. Model B achieves 98.28% accuracy and AUC-ROC of 0.9951. A reward-based response selection loop produces a 29.5% score uplift by selecting the highest-scoring of three generated candidates.

Authors

Institutions

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-06-18
DOI
https://doi.org/10.5281/zenodo.20752813
Primary Topic
Artificial Intelligence in Law
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

AI-Powered Interactive Legal Role-Playing Tool for Courtroom Advocacy Training

Nuwan Kodagoda, Kavinga Yapa Abeywardena, Sumudu Deshan Dias, K.T. Himaya Wijesinghe et al.
Zenodo (CERN European Organization for Nuclear Research)
Artificial Intelligence in Law
article

AI-Powered Interactive Legal Role-Playing Tool for Courtroom Advocacy Training

Nuwan Kodagoda, Kavinga Yapa Abeywardena, Sumudu Deshan Dias, K.T. Himaya Wijesinghe, A. Kamesh Dilshan Muthunayaka, Janith Viranga Majuwana Gamage
article en

Abstract

This paper presents an AI-powered interactive legal role-playing tool designed to simulate courtroom advocacy while continuously evaluating the legal quality of learner arguments. The system implements a two-stage Fast Brain and Smart Brain audit pipeline, where a fine-tuned BERT-based model assesses legal density and Gemini generates plain-English explanatory feedback. The platform generates Sri Lankan courtroom scenarios, manages five AI personas, and maintains autonomous dialogue through a heartbeat-driven engine. Model B achieves 98.28% accuracy and AUC-ROC of 0.9951. A reward-based response selection loop produces a 29.5% score uplift by selecting the highest-scoring of three generated candidates.

Zenodo (CERN European Organization for Nuclear Research)
Sri Lanka Institute of Information Technology (LK)
Peace, Justice and strong institutions
Openalex Percentile: Top 2%
Artificial Intelligence in Law
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

AI-Powered Interactive Legal Role-Playing Tool for Courtroom Advocacy Training — Nuwan Kodagoda, Kavinga Yapa Abeywardena, et al. · Zenodo (CERN European Organization for Nuclear Research) (2026) | TGRS Research Map | TGRS