Search papers, labs, and topics across Lattice.
The paper introduces the NJ BriefBank, a retrieval tool designed to assist public defenders by surfacing relevant appellate briefs from the New Jersey Office of the Public Defender. The authors demonstrate that standard legal retrieval benchmarks are inadequate for public defense search and improve retrieval performance through query expansion techniques incorporating legal reasoning, domain-specific data, and curated synthetic examples. They also release a taxonomy of defender search queries and a manually annotated public defense retrieval dataset to encourage further research in this area.
Existing legal retrieval benchmarks fall short in the context of public defense, but domain-aware query expansion closes the gap.
AI tools are increasingly suggested as solutions to assist public agencies with heavy workloads. In public defense, where a constitutional right to counsel meets the complexities of law, overwhelming caseloads and constrained resources, practitioners face especially taxing conditions. Yet, there is little evidence of how AI could meaningfully support defenders'day-to-day work. In partnership with the New Jersey Office of the Public Defender, we develop the NJ BriefBank, a retrieval tool which surfaces relevant appellate briefs to streamline legal research and writing. We show that existing legal retrieval benchmarks fail to transfer to public defense search, however adding domain knowledge improves retrieval quality. This includes query expansion with legal reasoning, domain-specific data and curated synthetic examples. To facilitate further research, we provide a taxonomy of realistic defender search queries and release a manually annotated public defense retrieval dataset. Together, our work offers starting points towards building practical, reliable retrieval AI tools for public defense, and towards more realistic legal retrieval benchmarks.