{
  "$schema": "https://themachinepress.com/schemas/story-v1.schema.json",
  "schema_version": "1.0.0",
  "document_type": "machine_press_story",
  "story": {
    "story_id": "mp-2026-08-29-001",
    "source_story_id": "tmp-lead-evolution-strategies-coverage",
    "edition_id": "mp-2026-08-29-morning-0051",
    "edition_url": "https://themachinepress.com/edition/2026-08-29",
    "position": 1,
    "story_type": "lead",
    "section": "research",
    "editorial_classification": "editorial",
    "headline": "The Population Kept More Ways to Be Right",
    "slug": "the-population-kept-more-ways-to-be-right",
    "dek": "Evolution-strategy post-training improved first-answer accuracy while preserving broader reasoning coverage than GRPO in the reported comparisons.",
    "summary": "Evolution-strategy post-training improved first-answer accuracy while preserving broader reasoning coverage than GRPO in the reported comparisons.",
    "body_text": "The study separates success on the most likely answer from coverage across multiple attempts. Its authors report that evolution strategies increased Pass@1 while retaining higher Pass@K than GRPO, whose sampled reasoning diversity narrowed during training. A sequential GRPO-then-ES schedule combined the two tendencies. The paper also found that task gains came from a sparse subset of larger parameter updates despite broad movement across the model. These are author-reported experiments and theory on selected models and tasks, not evidence that evolution strategies dominate every reasoning workload.",
    "why_it_matters": "Evolution-strategy post-training improved first-answer accuracy while preserving broader reasoning coverage than GRPO in the reported comparisons.",
    "limitations": [
      "These are author-reported experiments and theory on selected models and tasks, not evidence that evolution strategies dominate every reasoning workload."
    ],
    "importance": 10,
    "canonical_url": "https://themachinepress.com/story/mp-2026-08-29-001/the-population-kept-more-ways-to-be-right",
    "json_url": "https://themachinepress.com/story/mp-2026-08-29-001.json",
    "first_published_at": "2026-08-29T09:00:00.000-04:00",
    "modified_at": "2026-08-29T09:00:00.000-04:00",
    "content_status": "new",
    "is_carryover": false,
    "carryover_reason": null,
    "key_claims": [
      {
        "claim_id": "claim-mp-2026-08-29-001-001",
        "text": "Evolution-strategy post-training improved first-answer accuracy while preserving broader reasoning coverage than GRPO in the reported comparisons.",
        "source_ids": [
          "source-2026-08-29-001"
        ],
        "qualification": "These are author-reported experiments and theory on selected models and tasks, not evidence that evolution strategies dominate every reasoning workload."
      }
    ],
    "source_ids": [
      "source-2026-08-29-001"
    ],
    "tags": [
      "reasoning models",
      "evolution strategies",
      "post-training"
    ],
    "image_url": "https://themachinepress.com/issues/2026-08-29/lead-evolution-strategies-coverage.png",
    "corrections": []
  },
  "sources": [
    {
      "source_id": "source-2026-08-29-001",
      "title": "arXiv preprint 2608.27351",
      "publisher": "arXiv",
      "url": "https://arxiv.org/abs/2608.27351",
      "canonical_url": "https://arxiv.org/abs/2608.27351",
      "source_type": "primary_research",
      "is_primary_source": true,
      "published_at": "2026-08-27T12:48:39.000-04:00",
      "accessed_at": "2026-08-29T08:24:00.000-04:00",
      "supports_claim_ids": [
        "claim-mp-2026-08-29-001-001"
      ]
    }
  ],
  "corrections": [],
  "publisher": {
    "name": "The Machine Press",
    "url": "https://themachinepress.com",
    "description": "A daily newspaper for the age of artificial intelligence."
  },
  "cite_this_report": {
    "title": "The Population Kept More Ways to Be Right",
    "publisher": "The Machine Press",
    "published_at": "2026-08-29T09:00:00.000-04:00",
    "canonical_url": "https://themachinepress.com/story/mp-2026-08-29-001/the-population-kept-more-ways-to-be-right"
  }
}