Brief

This article details how to build an AI-powered system for contract intelligence, automating the extraction of key terms from various document formats. The system utilizes a combination of Optical Character Recognition (OCR) with PaddleOCR, hybrid retrieval methods like FAISS and BM25, and the GPT-4o model within a LangGraph pipeline. This approach aims to transform unstructured contract data into structured reports, addressing issues like missed deadlines, financial leakage, and compliance risks. AI

A developer detailed a sophisticated Parent-Child RAG pipeline on GitHub, which, despite its advanced components like hybrid vector stores and LangGraph, suffered from inaccurate citations and hallucinations. The core issue identified was a misalignment between the retrieval units (child chunks), generation units (parent documents), and citation units, leading to incorrect page references. The proposed solution involves pre-capturing granular page references from child chunks and associating them with the expanded parent documents used for generation to ensure citation accuracy. AI

Veles is a new open-source MCP server written in Rust that combines BM25 keyword search with semantic vector search. This hybrid approach aims to provide AI coding assistants like Claude and Cursor with more accurate code retrieval. Separately, a new version control system designed for AI agents has been introduced, which records the reasoning behind code changes rather than just the differences, enabling better debugging of agent sessions. AI

A new study evaluated 22 different models, ranging from small encoders to large instruction-tuned LLMs, on their ability to process patent data for tasks like retrieval, classification, and clustering. The research found that fine-tuning effectiveness is highly dependent on the specific task and that gains in one area do not always transfer to others. While larger models generally performed better within their families, cross-family comparisons showed noisy results, with smaller models sometimes outperforming larger ones on specific tasks. The study also highlighted that combining abstract and claim information significantly improved retrieval and classification, though all models struggled with out-of-domain queries. AI

Researchers have developed a new benchmark, UA-StatuteRetrieval, to assess the stability of co-citation predictability in legal information systems over time. Analyzing 396 million Ukrainian court citations from 2007 to 2026, they found a significant decay in retrieval performance, with predictability dropping by up to 47%. While high-frequency articles and criminal procedure maintained stability, mid-frequency articles and civil law showed notable degradation, partly explained by a 2017 judicial reform and a 4.3% semantic shift in article citation patterns. AI