arXiv cs.AI· Sujung Kim, Seung Hwan Cho, Sangjin Park, Young-Min Kim·· 10 天前AI 评分22
面向股票收益预测的可信 LLM 叙事：推理外化框架
Reasoning Externalization for Faithful Large Language Model Narratives of Stock Return Predictions
AI 导读
研究提出一种 LLM 叙事框架，将时序 SHAP 证据与历史市场状态类比结合，用于横截面股票收益预测的可解释叙事生成。在 Qwen3 上逐步外化数值与关系推理后，Qwen3-32B-Instruct 的证据忠实度从 0.696 提升至 0.996，时间与关系准确率同步改善。历史类比虽未提升结构化自动忠实度，但获得了更高的人工有用性评分。
正文 · 原文
Abstract:In finance, interpreting machine learning predictions is essential, yet the numerical outputs of explainable AI can be difficult for non-experts to understand. While large language models (LLMs) can translate these outputs into natural language, they may produce errors when inferring numerical changes and feature relations. We propose an LLM narrative framework for cross-sectional stock return prediction that combines temporal Shapley additive explanations (SHAP) evidence with historical regime analogs. Temporal evidence tracks changes in the normalized global SHAP importance of an XGBoost model over six months. Historical analogs are past periods with similar changes in SHAP importance, their model performance and subsequent market returns are provided as comparative context. Using this framework, we conduct a controlled study of progressive reasoning externalization, sequentially providing raw SHAP sequences, deterministic temporal descriptors, and feature relations. Each generated claim is verified against provenance-linked evidence. Across Qwen3, externalizing numerical and relational reasoning improved evidence faithfulness as well as temporal and relational accuracy. Evidence faithfulness increased from 0.696 to 0.996 for Qwen3-32B-Instruct. While historical analogs did not improve structured automatic faithfulness, they received higher human-rated usefulness scores. These results suggest that externalizing verifiable reasoning enhances narrative faithfulness and that historical context adds interpretive value.
| Subjects: | Artificial Intelligence (cs.AI); Computational Engineering, Finance, and Science (cs.CE) |
| Cite as: | arXiv:2609.38869 [cs.AI] |
| (or arXiv:2609.38869v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2609.38869 arXiv-issued DOI via DataCite |
Submission history
From: Sangjin Park [view email]
[v1]
Wed, 30 Sep 2026 03:20:17 UTC (1,030 KB)
[v2]
Tue, 6 Oct 2026 09:02:14 UTC (2,621 KB)
来源：arXiv cs.AI · arxiv.org