大语言模型中的生成与感知：一种令牌概率方法

中文摘要 AI 辅助

心理语言学中语言生成与感知之间的不对称已有充分记录。大语言模型（LLMs）是否表现出功能上类似的区别仍是一个悬而未决的问题，尤其是它们在输入和输出处理上依赖相同的基础机制（下一个令牌预测）。在这项探索性研究中，我们通过直接令牌概率测量而非元语言提示来操作生成 - 感知区别。使用基础的Llama - 3.1 - 8B模型，我们在生成提示下生成诗歌，并在重新表述的生成提示和感知导向提示下对相同令牌重新评分。在一个包含四个生成提示和三个感知提示的扩展实验中，生成 - 感知距离始终且大幅超过生成 - 生成距离，不同条件下范围不重叠，总体平均比率约为1.8。生成 - 生成控制中的近上限相关性证实该效应特定于交际框架而非提示表面变化，且该效应在五个开放权重模型中都能复制。时间分析表明感知提示在序列开始时影响最强，随着生成上下文积累差异衰减，不过衰减的具体形状因提示对而异。这些发现表明仅提示框架就能在LLM概率分布中引发生成 - 感知区别，即使在仅解码器架构中也是如此。

英文摘要

The asymmetry between language production and perception has been well-documented in psycholinguistics. Whether large language models (LLMs) exhibit a functionally analogous distinction remains an open question, particularly given that LLMs rely on the same underlying mechanism (next-token prediction) for both input and output processing. In this exploratory study, we operationalize the production-perception distinction through direct token probability measurements rather than metalinguistic prompting. Using the base Llama-3.1-8B model, we generated poems under a production prompt and re-scored the same tokens under both rephrased production prompts and perception-oriented prompts. Across an extended experiment with four production and three perception prompts, production-perception distances consistently and substantially exceeded production-production distances, with non-overlapping ranges across conditions and an overall average ratio of approximately 1.8. Near-ceiling correlations in the production-production control confirm that the effect is specific to communicative framing rather than prompt surface variation, and we show the effect replicates across five open-weight models (Llama-3.1-8B, EuroLLM-9B, gemma-2-9b-it, Mistral-7B-Instruct-v0.3, and Qwen2.5-7B-Instruct), spanning both base and instruction-tuned variants. Temporal analysis revealed that the perception prompt exerts its strongest influence at the beginning of the sequence, with divergence decaying as generated context accumulates, though the specific shape of this decay varies across prompt pairs. These findings suggest that prompt framing alone induces a production-perception distinction in LLM probability distributions, even within a decoder-only architecture.

发表机构

• Faculty of Arts, Charles University, Prague（布拉格查理大学文学院）
• Faculty of Mathematics and Physics, Charles University, Prague（布拉格查理大学数学与物理系）

机构由 AI 辅助整理，请以论文原文为准。