How Transparent is DiffusionGemma?
作者: Joshua Engels, Callum McDougall, Bilal Chughtai, Janos Kramar, Senthoran Rajamanoharan, Cindy Wu, Arthur Conmy, Asic Q Chen, Jean Tarbouriech, Min Ma, Brendan O'Donoghue, João Gabriel Lopes de Oliveira, Rohin Shah, Neel Nanda
分类: cs.LG, cs.AI
发布日期: 2026-06-18
备注: 20 main text pages and 6 pages of references and appendices
💡 一句话要点
研究DiffusionGemma的透明性以提升模型可解释性
🎯 匹配领域: 支柱三：空间感知与语义 (Perception & Semantics)
关键词: 推理透明性 大规模语言模型 可解释性研究 扩散模型 中间状态映射 监控性 算法透明性
📋 核心要点
- 现有的DiffusionGemma模型在透明性方面存在挑战，尤其是在理解中间计算状态时表现不佳。
- 论文提出通过可解释的中间状态映射来提高DiffusionGemma的透明性，从而降低不透明的串行深度。
- 实验结果表明，DiffusionGemma在监控性方面与Gemma 4相当，且揭示了扩散模型特有的推理现象。
📝 摘要（中文）
大规模语言模型（LLM）的推理透明性对于理解模型决策、减轻误用和调试意外行为至关重要。本文探讨DiffusionGemma在连续潜在空间中进行计算的透明性问题，分解为变量透明性和算法透明性两个方面。尽管DiffusionGemma在变量透明性上表现不佳，但通过可解释的中间状态映射，减少了不透明的串行深度。我们还进行了多项可解释性案例研究，发现了扩散特有现象，并测试了模型输出的监控能力，结果显示DiffusionGemma在监控性上与Gemma 4相似。
🔬 方法详解
问题定义：本文旨在解决DiffusionGemma模型在推理透明性方面的不足，尤其是在变量透明性和算法透明性上的挑战。现有方法在理解模型的中间计算状态时存在较高的串行深度，导致透明性不足。
核心思路：通过引入可解释的中间状态映射，论文提出了一种新的方法来降低DiffusionGemma的串行深度，从而提升模型的透明性。这种设计使得我们能够更好地理解模型的推理过程。
技术框架：整体架构包括两个主要模块：第一，利用可解释的中间状态来映射去噪步骤中的信息流；第二，进行一系列可解释性案例研究，以揭示扩散模型的特有现象。
关键创新：最重要的创新在于通过可解释的中间状态显著降低了DiffusionGemma的不透明串行深度，从28.6倍降至1.1倍Gemma 4。这一创新使得模型的推理过程更加透明。
关键设计：在设计中，论文关注了中间状态的可解释性，采用了特定的损失函数和网络结构，以确保在降低串行深度的同时不影响下游任务的性能。
🖼️ 关键图片
📊 实验亮点
实验结果显示，DiffusionGemma的变量透明性显著提高，串行深度从28.6倍降至1.1倍Gemma 4，且在监控性方面与Gemma 4相当。这表明DiffusionGemma在保持性能的同时，能够提供更好的可解释性。
🎯 应用场景
该研究的潜在应用领域包括自然语言处理、机器翻译和对话系统等。通过提升模型的透明性，研究者和开发者能够更好地理解和调试模型，从而提高模型在实际应用中的可靠性和安全性。
📄 摘要（原文）
LLM reasoning transparency is a critical affordance for understanding model decisions, mitigating misuse and misalignment, and debugging surprising model behaviors. However, DiffusionGemma performs a larger fraction of its computation in a continuous latent space; does this make its reasoning less transparent? We study this question by decomposing transparency into two components: variable transparency, whether we understand intermediate snapshots of a model's computational state; and algorithmic transparency, whether we can use these snapshots to reconstruct the process by which the model arrived at its outputs. Naively, DiffusionGemma has poor variable transparency: its opaque serial depth, the amount of serial computation that occurs in between interpretable model states, seems at first 28.6X higher than the corresponding autoregressive Gemma 4 model. However, we show that we can map the information flowing between denoising steps through an interpretable token bottleneck with no decrease in downstream performance. Treating these intermediate states as interpretable reduces the opaque serial depth to just 1.1X that of Gemma 4. Algorithmic transparency is harder for diffusion models than for autoregressive models because all token predictions in the canvas can change at every denoising step, giving the model the power to implement complicated distributed algorithms during the denoising process. To begin bridging this gap, we conduct a suite of interpretability case studies, uncovering initial evidence of novel diffusion-specific phenomena such as non-chronological reasoning, token and sequence smearing, and intermediate-context reasoning. Finally, we test monitorability, a key application of transparency that measures whether model outputs are useful for downstream tasks. We find that DiffusionGemma is similarly monitorable to Gemma 4.