最新预印本
查看全部 →Construction and Validation of a Multidimensional Tragedy Evaluation Model A Computational Narratology Approach
Quantitative assessment of tragic aesthetics has remained methodologically fragmented, hindering systematic cross-work and cross-tradition comparisons in literary studies.To address this,we constructed and validated the Multidimensional Tragedy Evaluation Model (MDTEM),a computational framework that operationalizes tragic intensity along four core dimensions plus a redemption coefficient. Drawing on a curated corpus of 39 canonical tragedies (20 Greek classical,19 modern Western),we decomposed each work into discrete narrative nodes with sentiment-coded emotional values using the NRC Emotion Lexicon combined with expert Delphi calibration.We then calibrated MDTEM hyperparameters using ordinary least squares and aggregated node-level values via a peak-end rule informed by Kahneman’s cognitive-psychological theory.The model achieved a mean absolute error (MAE)of 3.18,a root mean squared error (RMSE)of 3.92,and a Pearson correlation of r =0.978 between predicted and expert-rated Tragedy Index (TI),with tier classification accuracy reaching 97.4% (38/39 correct).Cross-tradition comparison revealed that Greek classical tragedies exhibited significantly higher TI (Mdn =78.6)than modern Western tragedies (Mdn =67.4;Mann-Whitney U =88.5,p =0.021, r =0.36),driven by higher Scope (S)and lower Redemption (R)coefficients in the Greek corpus.An AI assisted refactoring case study on Romeo and Juliet demonstrated that the model can serve as both a “compass” and“validator” for targeted aesthetic manipulation, successfully shifting the tragedy level from T1 (Extreme,83.4)to T3 (Sorrowful,54.7)with expert agreement.MDTEM thus provides a reproducible,mathematically grounded framework for cross-tradition tragedy research and offers a structured pathway for AI-assisted literary analysis and generation,translating qualitative notions of “cosmic fate” versus“psychological interiority” into measurable aesthetic regularities.
计算机应用研究-张量动力学作家数字人写作模型：四组对照实验设计与实证分析
针对大模型风格模拟的“画皮”与人格崩坏瓶颈，本研究构建高阶人格张量动力学框架。通过四组对照实验验证：势能阱使状态偏差方差降低86.8%；在180个测试单元中，本方法四项指标显著优于基线（p<1e-10）；消融实验证实势能阱与惯性模块的核心贡献；跨时空语境下核心人格维度完全锁定。噪声敏感性测试显示强鲁棒性。研究表明，将作家风格建模为受控连续时间动力学系统，是解决人格崩坏与可解释性问题的有效路径。
Persona Recovery:When and How Self-Authorial Recovery Mechanisms Restore Long-Form Authorial Consistency
Long-form generation by authorial language models suffers from the well documented "persona collapse" problem: over sufficiently long outputs, the author's distinctive voice, narrative logic, and stylistic fingerprint drift away from the persona's core profile. We present a controlled study of three recovery mechanisms — none, threshold-based, and learned predictive — applied to three high-modernist writers (Joyce, Woolf, Proust) under three noise regimes (σ=0.5,5.0,20.0). The state evolves under a 7-dimensional second-order tensor-dynamics system; recovery is a real-time intervention that snaps the state back when predicted or observed deviation exceeds ε=0.5.We find that (i) in the moderate-noise regime,recovery cuts collapse rate by 50– 70% (Joyce: 0.247→0.083;Woolf: 0.380→0.103; Proust: 0.342→0.131) with Hedges' g = -1.8 to -4.4; (ii) the learned predictor modestly outperforms threshold-only (g=-0.3 to-0.9) but with much lower computational cost; and (iii) in the high-noise regime, the system spends so much time collapsed that even the learned predictor cannot meaningfully reduce collapse (Joyce: 0.987→0.974; Woolf: 0.990→0.974; Proust: 0.989→0.961). We recommend threshold-based recovery as a default for moderate noise and learned recovery when computational headroom permits,and we identify σ≈5 as the practical ceiling beyond which persona consistency cannot be maintained.
三步完成分享
上传 PDF
填写标题、作者与摘要，选择要分享的 PDF 文件。
获得专属主页
系统为每篇论文生成独立的详情页，含在线预览。
复制链接分享
把论文链接发给任何人，无需注册即可查看与下载。