AI-generated text detectors often achieve high in-domain accuracy but remain sensitive to changes in domain, generator, and writing style. This limitation partly arises because conventional detectors rely predominantly on latent language-model representations and fixed decision boundaries, without explicitly characterizing sample-level structural and semantic variation. We introduce S2E-AL, an end-to-end score-aware structure–semantic entropy adversarial learning framework for AI-generated text detection. S2E-AL encodes local dependency, global contextual, and semantic entropy features into a structure–semantic representation. A gradient-reversal source classifier discourages this representation from retaining source-specific information, while a score-aware threshold estimator conditions on both the entropy representation and a detached base-detector logit to produce a sample-specific logit adjustment. The adjusted logit is used to estimate the final probability that a text is AI-generated. Across HC3, HC3_Plus, LLMTrace, RAID, and XSum, the score-aware model achieved a macro Accuracy of 90.90% and a macro F1-score of 91.42%. Its performance remained lower on XSum, indicating that substantial domain and stylistic shifts remain challenging. These results suggest that jointly modeling structure–semantic variation and detector confidence can improve the overall performance–stability trade-off of AI-generated text detection across heterogeneous test distributions.
This paper presents a novel algorithm, the Drug Pair-based Fuzzy Decision Tree (DP-FDT), designed to improve classification performance on imbalanced drug combination datasets. Unlike traditional decision trees, the DP-FDT integrates drug combination feature pairs into its feature selection and node construction processes to better capture the underlying relationships between features from different drugs. The algorithm employs a fuzzy splitting rule that allows samples to be assigned to both child nodes with certain probabilities, thereby enhancing the model’s fault tolerance. In addition, logistic regression classifiers are incorporated at impure leaf nodes to provide an additional means of separating samples that cannot be perfectly partitioned. Experimental results on two cell line datasets (A375 and A549) demonstrate that the DP-FDT achieves superior performance, particularly in terms of AUC and F1 score, compared with several state-of-the-art oblique decision trees and a deep learning baseline. It also improves recall for minority classes. Ablation studies confirm the positive contributions of its key components. Furthermore, a preliminary bioinformatics analysis provides insights into the biological processes associated with the most influential features identified by the model. Overall, the DP-FDT offers an effective and interpretable approach for analyzing complex drug combination data.
Multi-stream graph convolutional networks improve skeleton action recognition by combining representations, but each branch adds inference cost. We evaluate one fixed stream-removal operating point in a CTR-GCN pipeline using a paired non-inferiority protocol on NTU RGB+D 120. A temporal configuration reduced source-frame entries requested from the skeleton store by 53.52%, but lost 3.191 percentage points of accuracy and provided no latency saving after reopening the data mapping. A three-stream configuration then omitted bone motion while retaining 64 temporal positions. Historical RTX 4090 measurements at batch size one, using development checkpoints, showed 24.30–24.94% lower latency. After development confirmation, the fixed pair was trained on 54,468 examples and evaluated on 59,477 cross-setup test examples. Overall accuracy was 89.1050% for four streams and 89.0277% for three streams; paired setup-bootstrap lower bounds exceeded all three prespecified non-inferiority margins. Stream removal introduced 605 errors and corrected 559, while predicted labels changed for 1778 examples. Thus, a net loss of 46 correct decisions concealed broader redistribution. Recall decreased for 61 of 120 actions. These findings support assessing a branch by its contribution to the complete system, measured cost, and explicit quality tolerances. They do not imply preservation of every action. A separate CPU experiment with the final checkpoints measured 24.67% mean latency saving across three processes on one host. The contribution is a controlled empirical assessment, not a new architecture.
Self-refinement can improve generated code without additional training, but without execution feedback, revision can also turn already correct code into incorrect code. We study how self-refinement protocol design affects code correctness using six locally deployable open-weight models, three Python code-generation benchmarks, and 9984 model-task pairs. Using the same initial code for each protocol, we compare direct revision with refinement that adds critique and planning, test whether asking the model if revision is needed before refining affects correctness, and compare separate calls with a single response. Among the evaluated refinement protocols, direct revision achieved the highest overall correctness, whereas adding separate Critique and Planning stages increased repairs but caused substantially more regressions. Direct revision produced 55 repairs and 20 regressions; adding Critique produced 220 repairs and 605 regressions, and adding Planning after Critique produced 206 repairs and 719 regressions. Decision filtering and Single-call organization improved the regression-prone protocols by reducing harmful changes. They improved pass rate by 3.89 to 5.10 and 3.22 to 3.92 percentage points, respectively, but neither approach surpassed direct revision. Repeated direct generation produced many repairs simply by generating another candidate, while direct revision maintained a much more favorable repair-regression balance. These results suggest that, without execution feedback, refinement designs should prioritize preserving already correct code over adding reasoning stages.
Engagement with generative artificial intelligence (GenAI) involves practical literacy as well as learning, professional, sociotechnical, and configuration concerns. This mixed-methods study examined whether these responses were differentiated among Chinese arts-university students. Primary inference used a quality-screened cross-sectional sample of 355 questionnaire records from 351 participant-identifier clusters after excluding 105 of 460 eligible records (22.8%) showing exact whole-battery straightlining across all 59 structured items; the full eligible sample of 460 records from 455 clusters was retained as a sensitivity analysis. GenAI literacy was assessed with a 16-item music-context-adapted self-report instrument covering cognitive (CL), behavioral (BL), affective (AL), and ethical (EL) literacy, and AI Anxiety with 21 items spanning Learning Anxiety (LA), Job-Replacement Anxiety (JR), Sociotechnical Blindness Anxiety (ST), and AI Configuration Anxiety (AC). Of six prespecified associations, only EL-JR and EL-ST met the Holm-adjusted criterion in the primary sample and remained supported in the full-sample and prior-use-adjusted sensitivity analyses. CL-LA and BL-LA were unsupported in both samples, whereas AL-LA and AL-AC were sample-specification sensitive. Measurement models showed mixed fit; CL-BL separation improved after quality screening but remained incomplete. A secondary 47-person exact-identifier matched panel showed higher later-wave CL, BL, and AL scores, whereas EL and all four AI Anxiety dimensions showed no Holm-adjusted difference. These literacy comparisons are exploratory because the administered literacy items were music-framed while the documented learning activities involved broadcasting scripts and AI-supported video production. Independently analyzed qualitative material provided construct-level consistency with the EL-ST interpretation and tentative complementarity for EL-JR, but was not participant-linked to the quantitative samples. Overall, the evidence supports a bounded interpretation of differentiated, calibrated rather than anxiety-free adaptation; it does not establish causal intervention effects or participant-level mixed-method mechanisms.
