{
  "abstract": "Mental health conversational systems must balance empathy, risk-aware safety, and actionable structure, yet optimising hand-designed proxies for these qualities can expose reward design failures. We present EvoRiri, a template-based Bangla–Banglish–English mental health assistant that uses no language model. Its response policy is an eight-gene mixed continuous–discrete genome (empathy, safety and structure weights, a distress-gating gene, two reserved escalation thresholds, a template selector, and a response-part budget). A genetic algorithm (population 100, 20 generations, tournament selection, and uniform crossover, with 2000 evaluations per run) maximises a scalar fitness that weights rule-based empathy, safety and structure scores (0.40/0.40/0.15) with per-example safety penalties. On 2872 de-identified records rendered as synthetic prompts, the evolved genome raises structural completeness from 0.65 to 1.00 and fitness from 0.556 to 0.703 over a fixed reference genome, with comparable safety. A one-factor ablation attributes this gain to a single structure-weight threshold, and over 30 paired seeds, genetic and equal-budget random search are equivalent on validation prompts. Trace analysis exposed failures that aggregate fitness hid, including question-only collapse, empathy-marker stuffing, and response truncation. The contribution is a diagnostic case study of reward design on a synthetic single-turn corpus, not an optimisation advantage or clinical validation.",
  "abstract_shown": true,
  "affiliations": [
    "Department of Computer Science & Engineering, Rajshahi University of Engineering & Technology, Rajshahi 6204, Bangladesh",
    "Transdisciplinary School, University of Technology Sydney, 15 Broadway, Ultimo, NSW 2007, Australia"
  ],
  "article_type": null,
  "arxiv": null,
  "author_total": 2,
  "authors": [
    {
      "affs": [
        0
      ],
      "name": "Jahnnobi Rahman",
      "orcid": "0009-0002-7744-9853"
    },
    {
      "affs": [
        1
      ],
      "name": "Mir Md. Jahangir Kabir",
      "orcid": "0000-0003-4963-8905"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/90bhfphdg0q18.bib",
  "cited_by": 0,
  "doi": "10.3390/math14213982",
  "fulltext_url": null,
  "id": "90bhfphdg0q18",
  "issns": [
    "2227-7390"
  ],
  "landing_url": "https://www.mdpi.com/2227-7390/14/21/3982",
  "language": null,
  "licence": "CC-BY-4.0",
  "licence_label": "CC BY 4.0",
  "licence_source": "article",
  "licence_url": "https://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": null,
  "native_subjects": [],
  "openalex": null,
  "pdf_url": null,
  "pmcid": null,
  "pmid": null,
  "published": "2026-11-01",
  "publisher": "MDPI AG",
  "ris": "https://research.draconitic.com/paper/90bhfphdg0q18.ris",
  "sources": [
    {
      "first_seen": "2026-11-04T00:23:28Z",
      "native_id": "10.3390/math14213982",
      "source": "crossref",
      "url": "https://www.mdpi.com/2227-7390/14/21/3982"
    }
  ],
  "subjects": [
    {
      "field": "mathematics",
      "origin": "issn:2227-7390",
      "subfield": "general"
    }
  ],
  "title": "EvoRiri: Interpretable Evolutionary Search for Diagnosing Reward Design Failures in a Bangla Mental Health Assistant",
  "title_text": "EvoRiri: Interpretable Evolutionary Search for Diagnosing Reward Design Failures in a Bangla Mental Health Assistant",
  "type": "peer-reviewed",
  "updated_at": "2026-11-05T00:20:49Z",
  "url": "https://research.draconitic.com/paper/90bhfphdg0q18",
  "venue": "Mathematics",
  "versions": []
}