发现大模型家族中真理感知能力可继承，通过激活关键注意力头提升事实准确性。
The Truth Stays in the Family: Enhancing Contextual Grounding via Inherited Truthful Heads in Model Lineages
- 通过分析注意力头权重，发现模型家族间事实性表现高度继承。
- 提出TruthProbe方法，使模型在问答中更依赖真实证据，减少幻觉。
- 适用于需要高可信度输出的场景，如医疗、金融等专业领域应用。
近期大型语言模型（LLMs）发展催生了多个共享基础模型的多模态语言模型（MLLMs），形成不同模型谱系。我们通过量化头级别上下文事实性得分，研究基础模型与下游变体间是否存在行为关联。在涵盖Vicuna、Qwen2.5、LLaMA2和Mistral系列的多种模型谱系中，发现事实性得分在家族内部高度保留，即使经过指令微调或多模态适配也依然如此。进一步表明该继承性与注意力头权重保持一致，且事实性头会关注与查询相关的证据。基于此，我们提出TruthProbe——一种软门控策略，增强事实性头的作用同时保留其他头的贡献。TruthProbe在HaluEval上提升上下文事实性，在POPE和CHAIR上降低多模态幻觉，且基础模型的事实性得分能有效传递至其微调后的语言模型和多模态模型后代。代码已公开于https://github.com/miso-choi/TruthProbe。
原文摘要 · Abstract (English)
Recent advances in large language models (LLMs) have produced many specialized multimodal LLMs (MLLMs) that share common foundational LLMs, forming distinct model lineages. It remains unclear whether a fundamental behavioral link exists between the foundational LLMs and downstream variants. We investigate this question by quantifying head-level context-truthfulness scores. Across diverse LLM and MLLM lineages, including Vicuna-, Qwen2.5-, LLaMA2-, and Mistral-based models, we find that Truth Scores are strongly preserved within model families, even after instruction tuning or multimodal adaptation. We further show that this inheritance is consistent with attention-head weight preservation, and that context-truthful heads attend to query-relevant evidence. Building on this finding, we propose TruthProbe, a soft-gating strategy that amplifies context-truthful heads while preserving other head contributions. TruthProbe improves contextual truthfulness on HaluEval and reduces multimodal hallucination on POPE and CHAIR, with base-LLM Truth Scores transferring effectively to their fine-tuned LLM and MLLM descendants. Code is available at https://github.com/miso-choi/TruthProbe.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。