LongFaith: Enhancing Long-Context Reasoning in LLMs with Faithful Synthetic Data

Despite the growing development of longcontext large language models (LLMs), datacentric approaches relying on synthetic data have been hindered by issues related to faithfulness, which limit their effectiveness in enhancing model performance on tasks such as long-context reasoning and question answering (QA).These challenges are often exacerbated by misinformation caused by lack of verification, reasoning without attribution, and potential knowledge conflicts.We propose LONG-FAITH, a novel pipeline for synthesizing faithful long-context reasoning instruction datasets.By integrating ground truth and citation-based reasoning prompts, we eliminate distractions and improve the accuracy of reasoning chains, thus mitigating the need for costly verification processes.We open-source two synthesized datasets-LONGFAITH-SFT and LONGFAITH-PO-which systematically address multiple dimensions of faithfulness, including verified reasoning, attribution, and contextual grounding.Extensive experiments on multi-hop reasoning datasets and LongBench demonstrate that models fine-tuned on these datasets significantly improve performance.Our ablation studies highlight the scalability and adaptability of the LONGFAITH pipeline, showcasing its broad applicability in developing long-context LLMs.