语言生成的重排序定律:一种通信理论视角
Reranking Laws for Language Generation: A Communication-Theoretic Perspective
AI总结:
本文从通信理论视角将LLM多假设生成后重排序的安全策略类比为噪声信道冗余传输,推导其渐近无错条件,提出重排序定律并在文本到代码生成和医疗数据翻译任务中验证了有效性。
摘要:
为确保大型语言模型(LLMs)的安全使用,必须降低其幻觉或生成不可接受答案的倾向。一种简单且常用的策略是先让LLM生成多个假设,再使用重排序器选择最佳假设。本文将该策略与利用冗余降低噪声通信信道错误率的方法进行类比,把生成器概念化为通过并行噪声信道传输消息多个描述的发送方,接收方通过对(可能受损的)描述进行排序并选择最可靠的描述来解码消息。我们提供了该协议渐近无错(即几乎必然产生可接受答案)的条件,即使在重排序器不完善(遵循Mallows或Zipf-Mandelbrot模型)且信道分布统计相关的情况下。我们利用该框架获得重排序定律,并在两个真实世界任务中通过LLMs进行实证验证:使用DeepSeek-Coder 7B的文本到代码生成,以及使用TowerInstruct 13B的医疗数据机器翻译。
英文摘要:
To ensure large language models (LLMs) are used safely, one must reduce their propensity to hallucinate or to generate unacceptable answers. A simple and often used strategy is to first let the LLM generate multiple hypotheses and then employ a reranker to choose the best one. In this paper, we draw a parallel between this strategy and the use of redundancy to decrease the error rate in noisy communication channels. We conceptualize the generator as a sender transmitting multiple descriptions of a message through parallel noisy channels. The receiver decodes the message by ranking the (potentially corrupted) descriptions and selecting the one found to be most reliable. We provide conditions under which this protocol is asymptotically error-free (i.e., yields an acceptable answer almost surely) even in scenarios where the reranker is imperfect (governed by Mallows or Zipf-Mandelbrot models) and the channel distributions are statistically dependent. We use our framework to obtain reranking laws which we validate empirically on two real-world tasks using LLMs: text-to-code generation with DeepSeek-Coder 7B and machine translation of medical data with TowerInstruct 13B.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。