通过隐空间迭代推理，实现测试时可扩展的计算能力。
Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach
- 在潜在空间中循环迭代，动态扩展推理深度。
- 35亿参数模型在推理任务上表现提升，等效计算量达500亿参数。
- 无需额外训练数据，适合小上下文场景下的复杂推理。
我们研究了一种新型语言模型架构，能够通过在潜在空间中隐式推理来扩展测试时的计算能力。该模型通过迭代一个循环模块，在测试时可无限制展开至任意深度。与主流依赖生成更多标记的推理模型不同，本方法无需专门训练数据，可在小上下文窗口下运行，并能捕捉难以用文字表达的推理类型。我们将一个概念验证模型扩展至35亿参数和8000亿标记。结果表明，该模型在推理基准测试中性能显著提升，部分任务提升幅度巨大，其等效计算负载可达500亿参数水平。
原文摘要 · Abstract (English)
We study a novel language model architecture that is capable of scaling test-time computation by implicitly reasoning in latent space. Our model works by iterating a recurrent block, thereby unrolling to arbitrary depth at test-time. This stands in contrast to mainstream reasoning models that scale up compute by producing more tokens. Unlike approaches based on chain-of-thought, our approach does not require any specialized training data, can work with small context windows, and can capture types of reasoning that are not easily represented in words. We scale a proof-of-concept model to 3.5 billion parameters and 800 billion tokens. We show that the resulting model can improve its performance on reasoning benchmarks, sometimes dramatically, up to a computation load equivalent to 50 billion parameters.
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。