{"ID":2863872,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.25184","arxiv_id":"2509.25184","title":"Incentive-Aligned Multi-Source LLM Summaries","abstract":"Large language models (LLMs) are increasingly used in modern search and answer systems to synthesize multiple, sometimes conflicting, texts into a single response, yet current pipelines offer weak incentives for sources to be accurate and are vulnerable to adversarial content. We introduce Truthful Text Summarization (TTS), an incentive-aligned framework that improves factual robustness without ground-truth labels. TTS (i) decomposes a draft synthesis into atomic claims, (ii) elicits each source's stance on every claim, (iii) scores sources with an adapted multi-task peer-prediction mechanism that rewards informative agreement, and (iv) filters unreliable sources before re-summarizing. We establish formal guarantees that align a source's incentives with informative honesty, making truthful reporting the utility-maximizing strategy. Experiments show that TTS improves factual accuracy and robustness while preserving fluency, aligning exposure with informative corroboration and disincentivizing manipulation.","short_abstract":"Large language models (LLMs) are increasingly used in modern search and answer systems to synthesize multiple, sometimes conflicting, texts into a single response, yet current pipelines offer weak incentives for sources to be accurate and are vulnerable to adversarial content. We introduce Truthful Text Summarization (...","url_abs":"https://arxiv.org/abs/2509.25184","url_pdf":"https://arxiv.org/pdf/2509.25184v2","authors":"[\"Yanchen Jiang\",\"Zhe Feng\",\"Aranyak Mehta\"]","published":"2025-09-29T17:59:42Z","proceeding":"cs.CL","tasks":"[\"cs.CL\",\"cs.AI\",\"cs.GT\"]","methods":"[\"Large Language Model\",\"Language Model\"]","has_code":false}