{"ID":2854681,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2510.14751","arxiv_id":"2510.14751","title":"Beyond Multi-Token Prediction: Pretraining LLMs with Future Summaries","abstract":"Next-token prediction (NTP) has driven the success of large language models (LLMs), but it struggles with long-horizon reasoning, planning, and creative writing, with these limitations largely attributed to teacher-forced training. Multi-token prediction (MTP) partially mitigates these issues by predicting several future tokens at once, but it mostly captures short-range dependencies and offers limited improvement. We propose future summary prediction (FSP), which trains an auxiliary head to predict a compact representation of the long-term future, preserving information relevant for long-form generations. We explore two variants of FSP: handcrafted summaries, for example, a bag of words summary of the future sequence, and learned summaries, which use embeddings produced by a reverse language model trained from right-to-left order. Large-scale pretraining experiments (3B and 8B-parameter models) demonstrate that FSP provides improvements over both NTP and MTP across math, reasoning, and coding benchmarks.","short_abstract":"Next-token prediction (NTP) has driven the success of large language models (LLMs), but it struggles with long-horizon reasoning, planning, and creative writing, with these limitations largely attributed to teacher-forced training. Multi-token prediction (MTP) partially mitigates these issues by predicting several futu...","url_abs":"https://arxiv.org/abs/2510.14751","url_pdf":"https://arxiv.org/pdf/2510.14751v2","authors":"[\"Divyat Mahajan\",\"Sachin Goyal\",\"Badr Youbi Idrissi\",\"Mohammad Pezeshki\",\"Ioannis Mitliagkas\",\"David Lopez-Paz\",\"Kartik Ahuja\"]","published":"2025-10-16T14:52:52Z","proceeding":"cs.LG","tasks":"[\"cs.LG\",\"cs.AI\"]","methods":"[\"Large Language Model\",\"Language Model\"]","has_code":false}