AI systems increasingly answer scientific questions about health, safety, and the environment. But most retrieval-augmented generation systems are tuned to provide factually correct, on-topic answers rather than to help non-experts understand what those answers mean for their lives and decisions. We focus on consequential scientific questions whose results directly shape people’s lives and study them through a public water-quality communication system, where residents and community leaders interpret the findings and choose actions. Their experiences show that on-topic answers can still be insufficient without explanation and context and that the emotional weight of risk information cannot be ignored. Our system first classifies each question by reasoning type (for example, causal versus policy-based), then generates follow-up questions to identify missing evidence and retrieve it. One component clearly distinguishes between what is known and what is uncertain, while another rewrites scientific details into accessible language, using persona-based styles, such as a caring neighbor or an administrative official, to adapt tone and readability. Ablations on over 160 questions show that the system uses an order of magnitude less context and, in several configurations, improves human-rated completeness. A completeness metric co-designed with community members and a fine-tuned learned judge reveal that standard relevance scores explain about 1% of variation in human completeness ratings, and even the tuned judge only moderately aligns with humans, indicating that completeness is a distinct human-centered objective that current metrics do not reliably capture.

This paper describes the submission of team POINTERS to the UZH ArgMining 2026 Shared Task, which aims to recover the argumentation structure of UN and UNESCO resolutions by labeling paragraph types, assigning specific tags, and predicting relations between paragraphs. We take a generative approach, treating each resolution as a sequence of claim-evidence pairs connected by explicit reasoning strategies. First, each paragraph is classified as preambular or operative and assigned tags from a 126-code vocabulary, with the model required to quote specific phrases to justify every decision. Second, for each paragraph, we first retrieve semantically related candidates using sentence transformers, then use reasoning strategies as a diagnostic scaffold to label the relation—supporting, complemental, contradictive, or modifying—along with a quoted, strategy-grounded rationale. Both steps run locally on Qwen3-8B-GGUF (Team, 2025) (NVIDIA RTX 4080, 16 GB VRAM) without any cloud API calls. In the absence of labeled data, we use Claude Sonnet 4.6 only for an internal diagnostic evaluation of the generated reasoning traces. The results show that a sub-8B open-source model can produce evidence-grounded explanations for formal diplomatic text, while relation labeling remains sensitive to the distinction between retrieval and reasoning strategy-based diagnosis.