Program
Keynote
 Dominik Stammbach Postdoctoral Research Associate CITP, Princeton University USA |
Bio: Dr. Dominik Stammbach is a postdoctoral research associate at the Center for Information Technology Policy (CITP) and the Polaris Lab at Princeton University. Prior to that, he obtained his PhD. in Spring 2024 at ETH Zurich. His current research centers on legal NLP and AI to support public defenders and improve access to justice. In collaboration with a public defender’s office, he has developed an AI tool to provide computational assistance and to make their work more effective. In parallel, his team is conducting interviews with public defenders across the US to understand their perspectives on AI, with the goal of identifying opportunities and challenges around AI and its implications on the day-to-day work of public defenders, and especially what would be good use cases to develop AI tools for. His research has been published on top conferences such as ACL, EMNLP, and others.
Title: How Can AI Strengthen Indigent Defense? Evidence from Qualitative Interviews and a Collaboration with a Public Defender's Office
Abstract: AI is expected to reshape or automate work and offer solutions to pressing societal challenges. One such challenge is indigent defense, a constitutional guarantee in the United States. Yet public defenders are asked to do more with less: representing clients deserving of adequate counsel, while facing overwhelming caseloads and scarce resources. AI has been proposed as a remedy for this access-to-justice crisis, but its practical role and limits remain poorly understood. In this keynote, Dr. Stammbach will discuss opportunities and concerns around AI for indigent defense, based on insights derived from semi-structured interviews and an ongoing collaboration with the New Jersey Office of the Public Defender. In qualitative interviews, participants view AI as most useful for evidence investigation to analyze overwhelming amounts of digital records, with narrower roles in legal research & writing, and client communication. In collaboration with NJ OPD, Dr. Stammbach's team developed a retrieval tool searching through internal briefs, designed to streamline legal writing and surface reusable arguments. Their findings reveal a mismatch between academic benchmarks and the queries public defenders submit: Models trained on existing benchmarks often transfer poorly, while domain-specific data, legal domain adaptation, and curated synthetic datasets improve retrieval. Their findings can inform future work on legal search, and outline a research agenda for practical AI applications for public defenders more broadly.
In addtion, Dr. Stammbach will address barriers limiting AI adoption for public defenders, and ethical considerations around using AI for indigent defense and deploying AI tools in public agencies.
Pay careful attention to the time zone difference!
Session Schedule
Date: Friday 14 November 2025 (Eastern Standard Time (EST), USA and Canada)
Location: Pan American, 2nd Floor Coatcheck
Time Slot |
Activity |
Speaker | Moderator |
|---|---|---|---|
| 1:30pm ~ 1:40pm | Opening | TBD | |
| 1:40pm ~ 2:30pm | Keynote Speech - How Can AI Strengthen Indigent Defense? Evidence from Qualitative Interviews and a Collaboration with a Public Defender's Office | Dr. Dominik Stammbach | Prof. Jiangping Chen |
| 2:40pm ~ 3:30pm | Paper Session 1 | Prof. Jiangping Chen | |
| The Challenges of Data Mining and A.I. for State Laws and Legislative Research: Recommendations for the Development of Legislative A.I. (25 minutes) | Sarah Ryan and Lingzi Hong | ||
| PatentGRPO: Group Relative Policy Optimization for Patent Text Generation (25 minutes) | Jieh-Sheng Lee | ||
| 3:30pm ~ 4:00pm | Coffee Break | ||
| 4:00pm ~ 5:15pm | Paper Session 2 | Prof. Jiangping Chen | |
| GRAPH-GRPO-LEX: Legal Contracts Graph Modeling via Reinforcement Learning with Group Relative Policy Optimization (25 minutes) | Moriya Dechtiar, et al. | ||
| Automatic Coherence-driven Inference on Arguments (25 mintues) | Steve Huntsman | ||
| Legal Argument Mining the TCA’s Technology Undertaking Exception with Turkish BERT (25 minutes) | M. Mikail Demir and M. Abdullah Canbaz | ||
| 5:15pm ~ 5:25pm | Closing | TBD |
Accepted Papers
- Sarah Ryan and Lingzi Hong, "The Challenges of Data Mining and A.I. for State Laws and Legislative Research: Recommendations for the Development of Legislative A.I."
- Jieh-Sheng Lee, "PatentGRPO: Group Relative Policy Optimization for Patent Text Generation"
- Moriya Dechtiar, Daniel Martin Katz, Mari Sundaresan, Sylvain Jaume, and Hongming Wang, "GRAPH-GRPO-LEX: Legal Contracts Graph Modeling via Reinforcement Learning with Group Relative Policy Optimization"
- Steve Huntsman, "Automatic Coherence-driven Inference on Arguments"
- M. Mikail Demir and M. Abdullah Canbaz, "Legal Argument Mining the TCA’s Technology Undertaking Exception with Turkish BERT"