LA-CPD: Local-Evidence-Aware Change-Point Detection for Human-LLM Authorship Segmentation

Qing Yang; Zhenyu Mao; Zixiang Luo; Zezheng Wu; Xinghe Cheng; Qinggang Zhang; Jingwei Zhang; Jiapu Wang

cs.CLSep 27, 2026

LA-CPD: Local-Evidence-Aware Change-Point Detection for Human-LLM Authorship Segmentation

Authors: Qing Yang, Zhenyu Mao, Zixiang Luo, Zezheng Wu, Xinghe Cheng, Qinggang Zhang, Jingwei Zhang, Jiapu Wang

Organizations: Guilin University of Electronic Technology, China · Jinan Unaiversity, China · Jilin University, China · Nanjing University of Science and Technology, China

Abstract

As LLM-generated text becomes increasingly human-like, accurately localizing LLM-authored spans in human-LLM co-authored documents is important for attribution and accountability in cases involving copyright infringement, fraud, and other harmful uses of AI-generated content. Sentence-level detectors provide local authorship evidence, but content variation can cause score fluctuations even among sentences from the same source, creating spurious boundaries. Recovering a coherent document partition therefore remains challenging when both the number and locations of authorship transitions are unknown. We propose Local-Evidence-Aware Change-Point Detection (LA-CPD), a structured method that transforms noisy sentence-level score sequences into coherent authorship segments. Given scores from a frozen local detector, LA-CPD combines a length-weighted within-segment residual with a windowed two-mean contrast to capture segment consistency and sustained changes around candidate cut points. Dynamic programming optimizes cut locations for each candidate count, while an AIC-style criterion selects the final partition, yielding sentence labels, authorship boundaries, and maximal LLM-authored spans. On a held-out human-LLM co-authored test set, LA-CPD outperforms WCP+AIC, increasing sentence-level accuracy from 0.747 to 0.796 while improving boundary localization and LLM-span delineation.

Figures & tables

Figure 1: LA-CPD training. Human–LLM sentence pairs are used to adapt a LoRA-augmented scoring model with a frozen reference model through a sampling-discrepancy separation loss. The adapted scorer is then frozen to produce sentence-level scores $q_{i}=\Phi(u_{i})$ for structured segmentation.

Figure 2: LA-CPD inference. Top left: a frozen local detector produces ordered sentence-level scores, while sentence lengths determine the fitting weights. Top right: $V(s,e)$ and $E(c;h)$ define the candidate-partition objective. Bottom right: dynamic programming optimizes cut locations for each fixed cut count $k$ . Bottom middle: an AIC-style criterion selects the final statistical cut set. Bottom left: two-cluster grouping of segment means delineates Human/LLM labels, authorship boundaries, and contiguous LLM-authored spans.

Methods	Overall			By condition
Methods	Acc $\uparrow$	WD $\downarrow$	AI-F1 $\uparrow$	H0 Acc $\uparrow$	B1 Acc / AI-F1 $\uparrow$	B2 Acc / AI-F1 $\uparrow$	B3-2 Acc / AI-F1 $\uparrow$	B3-3 Acc / AI-F1 $\uparrow$
SenPred	0.598	0.874	0.086	0.404	0.690 / 0.086	0.696 / 0.091	0.527 / 0.072	0.673 / 0.135
Voting	0.711	0.581	0.317	0.466	0.836 / 0.419	0.829 / 0.402	0.642 / 0.208	0.784 / 0.434
TextTiling	0.724	0.413	0.506	0.534	0.839 / 0.726	0.805 / 0.667	0.681 / 0.334	0.760 / 0.567
PaLD-scores	0.713	0.880	0.037	0.753	0.707 / 0.041	0.700 / 0.036	0.730 / 0.034	0.672 / 0.048
VCP (MAD)	0.729	0.257	0.427	0.891	0.699 / 0.642	0.725 / 0.673	0.711 / 0.106	0.617 / 0.335

Table 1: Performance comparison on the GPT-5.5 human–LLM co-authored test set. H0 reports accuracy, whereas mixed conditions report accuracy and AI-F1. Best and second-best results among methods with unknown boundary counts are shown in bold and underlined , respectively. Oracle-

K

, which uses the ground-truth boundary count, is reported only for diagnostic purposes and is excluded from ranking.

Figure 3: Cross-model generalization performance of LA-CPD in terms of sentence-level accuracy across different generation models.

Methods	Acc $\uparrow$	WD $\downarrow$	AI-F1 $\uparrow$	B3-2 F1 $\uparrow$
LA-CPD (Full)	0.796	0.231	0.600	0.338
LA-CPD w/o $E(c;h)$	0.715	0.279	0.519	0.241
LA-CPD w/o Length Weighting	0.781	0.223	0.516	0.185
LA-CPD w/o AIC	0.789	0.250	0.605	0.325

Table 2: Ablation study of LA-CPD. Best and second-best results are shown in bold and underlined , respectively.

Metric	$h$				$\alpha$			$\beta$			$r$
Metric	1	3	$6^{\dagger}$	8	0.25	$0.75^{\dagger}$	1.5	0.75	$1.25^{\dagger}$	2.0	1.0	2.0	$2.5^{\dagger}$	3.0
Acc $\uparrow$	0.702	0.766	0.796	0.793	0.794	0.796	0.786	0.790	0.796	0.793	0.769	0.795	0.796	0.786
AI-F1 $\uparrow$	0.494	0.561	0.600	0.567	0.592	0.600	0.575	0.577	0.600	0.588	0.614	0.607	0.600	0.544
WD $\downarrow$	0.301	0.277	0.231	0.247	0.255	0.231	0.259	0.257	0.231	0.255	0.321	0.267	0.231	0.244

Table 3: Hyperparameter sensitivity of LA-CPD on the GPT-5.5 human–LLM co-authored dataset.

\dagger

denotes the default configuration. Best and second-best results within each parameter group are shown in bold and underlined , respectively.

Appendix figures & tables9 assets

Supplementary material from the paper’s appendix.

Appendix

Component	Setting
Scoring and reference backbone	Gemma-3-1B
LoRA target modules	q_proj , k_proj , v_proj , o_proj
LoRA rank	4
LoRA scaling parameter	16
LoRA dropout	0.05
Continued-adaptation training pairs	4,000

Appendix

Table 4: Configuration of the continued local-detector adaptation.

Operation	Time	Main storage
Prefix statistics	$O(n)$	$O(n)$
Interval residual cache	$O(n^{2})$	$O(n^{2})$
Windowed evidence	$O(nh)$	$O(n)$
Shared dynamic programming	$O((K_{\mathrm{eff}}+1)n^{2})$	$O((K_{\mathrm{eff}}+1)n)$
Candidate backtracking and residual accumulation	$O((K_{\mathrm{eff}}+1)^{2})$	Does not change the overall bound
Authorship delineation	$O(n\log n+m^{2})$	$O(n)$

Appendix

Table 5: Time and storage costs of structured segmentation, excluding detector forward passes.

\widehat{\mathcal{C}}\leftarrow\{i\in\{1,\ldots,n-1\}:\widehat{y}_{i}\neq\widehat{y}_{i+1}\};

Appendix

Algorithm 1 Full LA-CPD Inference

Stratum	Relative position	Target proportion
Early	25%–40%	Approximately one third
Middle	40%–60%	Approximately one third
Late	60%–75%	Approximately one third

Appendix

Table 6: Target boundary-position strata during corpus construction.

H0	B1	B2	B3-2	B3-3	Total
12,000	20,000	19,999	10,998	6,000	68,997

Appendix

Table 7: Condition counts in the cleaned candidate pool, before removing the two B1 records with zero-length AI spans.

Split	Total	H0	B1	B2	B3-2	B3-3
Training	48,297	8,434	14,073	14,018	7,639	4,133
Validation	6,900	1,198	1,924	1,996	1,169	613
Test	13,798	2,368	4,001	3,985	2,190	1,254
Total	68,995	12,000	19,998	19,999	10,998	6,000

Appendix

Table 8: Statistics of the complete GPT-5.5 corpus after removing the two invalid records and grouping examples by source family.

Variant	Acc $\uparrow$	WD $\downarrow$	AI-F1 $\uparrow$	B3-2 F1 $\uparrow$
LA-CPD (Full)	0.796	0.231	0.600	0.338
LA-CPD w/o $E(c;h)$	0.715	0.279	0.519	0.241
LA-CPD w/o Length Weighting	0.781	0.223	0.516	0.185
LA-CPD w/o AIC	0.789	0.250	0.605	0.325
WCP+AIC (reference)	0.747	0.270	0.552	0.280

Appendix

Table 9: Complete component ablation results on the held-out GPT-5.5 test subset of 1,000 documents. Best and second-best results among the reported methods are shown in bold and underlined , respectively.

Figure 4: Complete cross-distribution generalization results on Hybrid Essays, RoFT, Claude–SQuAD, Claude–WritingPrompts, and Claude–XSum. The top and bottom panels report sentence-level accuracy and AI-fragment F1, respectively. Bars are grouped by method, and colors identify the evaluation datasets. Missing results are left blank. Higher Acc and AI-F1 indicate better performance. LA-CPD oracle- $K$ uses the ground-truth boundary count and is included only as a diagnostic setting.

Metric	$\rho=1.00$	$\rho=0.75$	$\rho=0.50$	$\rho=0.35$
Acc $\uparrow$	0.756	0.747	0.737	0.731
H0 Acc $\uparrow$	0.852	0.840	0.829	0.810
AI-F1 $\uparrow$	0.519	0.512	0.510	0.513
B3-2 AI-F1 $\uparrow$	0.237	0.249	0.264	0.276

Appendix

Table 10: Sensitivity to the regime-return discount

\rho

under the MAD-based complexity penalty. Other parameters are fixed to

h=3

\alpha=0.5

\beta=1

, and

c=2

Explore similar work

Segmenting Human-LLM Co-authored Text via Change Point Detection
May 5, 2026Mengchu Li, Jin Zhu, Jinglai Li +1Machine-Generated Text Detection Document Segmentation
School of Mathematics University of Birmingham Birmingham, UK · Department of Statistics London School of Economics and Political Science London, UK
Detecting LLM-Generated Tokens in Human--LLM Coauthored Text
Jul 23, 2026Yangjun Lu, Hongyi Zhou, Fabian Spill +3Machine-Generated Text Detection Authorship
School of Mathematics, University of Birmingham, Birmingham, UK · School of Statistics and Data Science, Shanghai University of Finance and Economics, Shanghai, China · Department of Statistics, The London School of Economics and Political Science, London, UK
Beyond Sub-Gaussian Detector Scores: Robust Weighted Profile-Loss Change Point Detection for Human-LLM Text Segmentation
Sep 29, 2026Wan Tian, Zhongyi Li, Yawen Li +3Change-Point Detection Document Segmentation
Peking University · Beihang University · Capital Normal University +2

May 5, 2026, cs.CLJ/K move · Enter open · S save

Segmenting Human-LLM Co-authored Text via Change Point Detection

Mengchu Li, Jin Zhu, Jinglai Li, Chengchun Shi

School of Mathematics University of Birmingham Birmingham, UK · Department of Statistics London School of Economics and Political Science London, UK

Machine-Generated Text Detection Document Segmentation Text Analysis and Detection Change-Point Detection

Open paper AskPDF arXiv

The rise of large language models (LLMs) has created an urgent need to distinguish between human-written and LLM-generated text to ensure authenticity and societal trust. Existing detectors typically provide a binary classification for an entire passage; however, this is insufficient for human--LLM co-authored text, where the objective is to localize specific segments authored by humans or LLMs. To bridge this gap, we propose algorithms to segment text into human- and LLM-authored pieces. Our key observation is that such a segmentation task is conceptually similar to classical change point detection in time-series analysis. Leveraging this analogy, we adapt change point detection to LLM-generated text detection, develop a weighted algorithm and a generalized algorithm to accommodate heterogeneous detection score variability, and establish the minimax optimality of our procedure. Empirically, we demonstrate the strong performance of our approach against a wide range of existing baselines. The python implementation of our proposal is available at https://github.com/Mamba413/DetectLLMSegmentation.

Abstract

Explore similar work

Segmenting Human-LLM Co-authored Text via Change Point Detection

Detecting LLM-Generated Tokens in Human--LLM Coauthored Text

Beyond Sub-Gaussian Detector Scores: Robust Weighted Profile-Loss Change Point Detection for Human-LLM Text Segmentation