We present a first known prototype of a dynamic, systematic benchmark of medical guidelines for 400+ questions, with 3.3+ trillion possible combinations, covering 100\% of guideline relationships. We transformed the WHO IMCI handbook into a directed graph with 200+ nodes (conditions, symptoms, treatments, follow-ups, severities) and 300+ edges, then used graph traversal to generate questions that incorporated age-specific scenarios and contextual distractors to ensure clinical relevance. Our graph-based approach enables systematic evaluation across clinical tasks (45-67\% accuracy), and we find models excel at symptom recognition but struggle with triaging severity, treatment protocols and follow-up care, demonstrating how customized benchmarks can identify specific capability gaps that general-domain evaluations miss. Beyond evaluation, this dynamic MCQA methodology enhances LLM post-training (supervised finetuning, GRPO, DPO), where correct answers provide high-reward samples without expensive human annotation. The graph-based approach successfully addresses the coverage limitations of manually curated benchmarks. This methodology is a step toward scalable, contamination-resistant solution for creating comprehensive benchmarks that can be dynamically generated, including when the guidelines are updated. Code and datasets are available at https://github.com/jessicalundin/graph_testing_harness

Predicting how old someone is based only on how they look is incredibly hard to get right, especially in those awkward early teen years. And yet bouncers, liquor store owners, and other age-restricted goods gatekeepers make that quick estimation all the time. This story originally appeared on WIRED UK. Their predictions are often wrong. Now London-based digital identity company Yoti believes its AI-powered age estimation can predict how old someone is if they're aged anywhere from 6 to 60. For the first time, it claims, it can accurately determine whether children are under or over 13, the minimum age many social media firms require their users to be.

Depth-first proof-number search (df-pn) is powerful AND/OR tree search to solve positions in games. However, df-pn has a notorious problem of infinite loops when applied to domains with repetitions. Df-pn(r) cures it by ignoring proof and disproof numbers that may lead to infinite loops. This paper points out that df-pn(r) has a serious issue of underestimating proof and disproof numbers, while it also suffers from the overestimation problem occurring in directed acyclic graph. It then presents two practical solutions to these problems. While bypassing infinite loops, the threshold controlling algorithm (TCA) solves the underestimation problem by increasing the thresholds of df-pn. The source node detection algorithm (SNDA) detects the cause of overestimation and modifies the computation of proof and disproof numbers. Both TCA and SNDA are implemented on top of df-pn to solve tsume-shogi (checkmating problem in Japanese chess). Results show that df-pn with TCA and SNDA is far superior to df-pn(r). Our tsume-shogi solver is able to solve several difficult positions previously unsolved by any other solvers.