Evaluating 13-21 models across six presentation operationalizations and four task-domain operationalizations suggests that, despite confounds, some models possess practical SGTR capabilities, and that SGTR should be monitored and considered in the design of safety-critical AI applications.
Abstract
Self-Generated Text Recognition (SGTR)--the ability of an LLM to identify its own outputs--poses risks to AI safeguards that rely on LLMs as evaluators or monitors: an LLM may recognize outputs from other copies of the same model and make biased judgments or collude outright. Prior work has drawn conflicting conclusions about whether current models possess significant SGTR capabilities. We explain these disagreements by identifying key experimental design choices--which we term operationalizations--that drive divergent results. Evaluating 13-21 models across six presentation operationalizations and four task-domain operationalizations, we find that accuracy varies substantially with evaluation format (pairwise vs individual assessments of text), conversation format (presenting candidate text in user tags vs assistant tags), and the domain of the task used to generate candidate text (e.g., coding vs summarization). We corroborate previous observations that a quality heuristic--models attributing authorship to text they perceive as higher quality--is a dominant confound. We also find that improving a model's SGTR performance via supervised fine-tuning (SFT) on one operationalization can generalize to others, and can increase the model's preference for its own outputs when it acts as a judge in the AlpacaEval framework. Our results suggest that, despite confounds, some models possess practical SGTR capabilities, and that SGTR should be monitored and considered in the design of safety-critical AI applications.
It is proposed that persona-based evaluation can serve as a scalable diagnostic of what generative systems value and prioritize when depicting humanity, and that persona generations are far from neutral.
N. Corrêa, Rafaela Weber Mallmann, David Kaczér et al.· Artificial Intelligence Revi...· 0 citations
LLM-based evaluators of natural language generation (NLG) quality are widely deployed as scoring tools and as automated training signals, yet the internal procedure by which they assign a rating remains poorly understood. We investigate this procedure mechanistically through an eight-attack perturbation taxonomy across the Readability and Adequacy dimensions of NLG quality, a generation pipeline that produces paired clean and corrupt summaries with controlled error intensity and explicit token-level modification maps, and a four-experiment battery of causal tracing, logit-lens vocabulary projection, and attention-head knockout applied to Themis (Llama-3-8B) and Prometheus (Mistral-7B). Both evaluators implement a structured, coherent evaluation pipeline operating in two stages: below layer 15, attention performs local error comparison and routes the result to the final input position; above it, the MLP cascade integrates the signal and writes the rating, with the decision crystallizing in the residual stream at a sharp late layer (L = 26 on Themis, L = 25 on Prometheus). Furthermore, a base-model control at the same scale (Llama-3-8B) reproduces the routing architecture and crystallization but not the stage separation, isolating the two mechanisms that fine-tuning specifically installs, suppression of below-L15 MLP contribution at the last position and a two-layer advance of the crystallization depth, indicating that fine-tuning sculpts an existing substrate rather than building the pipeline from scratch. We release the source code and data at https://github.com/himil-v/judge-mech
This survey provides a comprehensive overview of recent advances in LLM-based evaluation, covering techniques, applications, and challenges across domains, with future directions emphasizing standardized protocols, uncertainty estimation, and human–AI collaboration.
M. Nadăş· Artificial Intelligence Revi...· 0 citations
This study evaluates two multimodal LLMs, Qwen2.5-VL-72B and Pixtral-Large-124B, as reviewers across 165 submissions to the 2026 International Conference on Learning Representations, a venue that postdates both models'training cutoffs.
Large language models (LLMs) struggle to classify text into taxonomies with many semantically similar labels, as the distinctions are domain-specific and not captured by pre-training. To handle large label spaces, a common approach retrieves top-$K$ candidate labels by embedding similarity and prompt the LLM to choose among them. However, top-$K$ retrieval reduces the number of candidates but does not help the model tell similar ones apart. When two similar labels both appear as candidates, the model lacks the signal to choose correctly between them. We propose a framework that (1) identifies which label pairs the model struggles to distinguish, (2) expands the candidate set to include confusable labels, and (3) generates targeted rules to differentiate between similar candidates. The framework requires no fine-tuning, and the generated rules transfer to smaller, cheaper models. On three benchmarks (WOS, Flipkart, LEDGAR), our approach improves Macro F1 by up to 10.0pp over retrieval baselines, with smaller models (2B--20B) gaining up to 11.5pp via cross-model transfer.
Results show that detector performance measured on the conventional human-vs-LLM benchmark does not transfer to human-authored text revised by an LLM, even though the same detectors remain largely robust to LLM-only rewriting.
