cs.CLNov 2, 2026

SAGE (Subjectivity-Aware Generation Estimator): A Linguistic Framework for AI-Generated Review Detection *

Authors: Gonzalo Freijedo Aduna, Charles Yaiche, Patrick Bas, Alda Mari

Organizations: IJN · CRIStAL

Abstract

We propose a simple, linguistically motivated approach to detecting AI-generated peer reviews based on the inherently human trait of subjectivity. We capture subjectivity at two levels: the semantic level of evaluative language and modality and the pragmatic level of speech acts. Using features derived from speech act annotations and model-predicted subjectivity scores, we train a lightweight classifier that achieves an AUROC of 0.9566 with only these linguistic features. Combining its output with the state-of-the-art detector Sem-Detect raises the TPR at 1% FPR of Sem-Detect on the APRDB benchmark from 0.874 to 0.985, outperforming the current state of the art. Our results show that linguistic features of subjectivity provide an explainable source of information for detecting AI-generated peer reviews and can be readily integrated into existing detection methods. The code and models are publicly available on GitHub (https://github.com/Arys02/generated\_ text_detection_hub) and Hugging Face (https://huggingface.co/Arys02), respectively.

Explore similar work

May 20, 2026

Sem-Detect: Semantic Level Detection of AI Generated Peer-Reviews

How can we distinguish whether a peer review was written by a human or generated by an AI model? We argue that, in this setting, authorship should not be attributed solely from the textual features of a review, but also from the ideas, judgments, and claims it expresses. To this end, we propose Sem-Detect, an authorship detection method for peer reviews that operationalizes this principle by combining textual features with claim-level semantic analysis. Sem-Detect compares a target review against multiple AI-generated reviews of the same paper, leveraging the observation that different AI models tend to converge on similar points, while human reviewers introduce more unique and diverse ones. As a result, Sem-Detect is able to distinguish fully AI reviews from authentic human-written ones, including those that have been refined using an LLM but still reflect human judgment. Across a dataset of over 20,000 peer reviews from ICLR and NeurIPS conferences, Sem-Detect improves over the strongest baseline by 25.5% in [email protected]% FPR in the binary setting. Moreover, in the three-class scenario, we empirically show that LLM refinement preserves the semantic signals of human reviews, which remain distinct from the patterns exhibited by fully AI-generated text; as a result, fewer than 3.5% of LLM-refined human reviews are misclassified as AI-generated.
Apr 16, 2026

PeerPrism: Peer Evaluation Expertise vs Review-writing AI

Large Language Models (LLMs) are increasingly used in scientific peer review, assisting with drafting, rewriting, expansion, and refinement. However, existing peer-review LLM detection methods largely treat authorship as a binary problem-human vs. AI-without accounting for the hybrid nature of modern review workflows. In practice, evaluative ideas and surface realization may originate from different sources, creating a spectrum of human-AI collaboration. In this work, we introduce PeerPrism, a large-scale benchmark of 20,690 peer reviews explicitly designed to disentangle idea provenance from text provenance. We construct controlled generation regimes spanning fully human, fully synthetic, and multiple hybrid transformations. This design enables systematic evaluation of whether detectors identify the origin of the surface text or the origin of the evaluative reasoning. We benchmark state-of-the-art LLM text detection methods on PeerPrism. While several methods achieve high accuracy on the standard binary task (human vs. fully synthetic), their predictions diverge sharply under hybrid regimes. In particular, when ideas originate from humans but the surface text is AI-generated, detectors frequently disagree and produce contradictory classifications. Accompanied by stylometric and semantic analyses, our results show that current detection methods conflate surface realization with intellectual contribution. Overall, we demonstrate that LLM detection in peer review cannot be reduced to a binary attribution problem. Instead, authorship must be modeled as a multidimensional construct spanning semantic reasoning and stylistic realization. PeerPrism is the first benchmark evaluating human-AI collaboration in these settings. We release all code, data, prompts, and evaluation scripts to facilitate reproducible research at https://github.com/Reviewerly-Inc/PeerPrism.
Oct 21, 2026

Building Auditable AI Text Detectors with Sparse Autoencoders

AI-generated text detectors are increasingly deployed in consequential settings---peer review, education, admissions, journalism, and content moderation---where their outputs directly affect individuals. Yet the strongest detectors are fine-tuned neural classifiers whose decisions are opaque, while transparent alternatives (stylometric and linguistic-feature classifiers) underperform them by wide margins. We present a pipeline that converts a fine-tuned neural detector into a classifier that operates on human-readable concepts. We accomplish this by training sparse autoencoders (SAEs) directly on a detector's residual stream, assigning auto-interpretation labels to each SAE feature, and retaining only features whose generated descriptions pass a global fidelity threshold. We then train a lightweight classifier on just the interpretable features. We find that our SAEs recover latent concepts that correspond to highly specific and predictive linguistic "tells" of AI-generated text. Classifiers built on these extracted features are not only highly interpretable but exhibit stronger performance on out-of-distribution texts than either neural or stylometric alternatives.