Log in
Back to #small language model
#small language model Open access

From perceptual rule transformation to listener attribution judgments: a blind-listening experiment on AI-generated, human–AI collaborative, and human-composed music

Junfang Chang Yue-Qi Jing
Aug 2026 · Frontiers in Psychology · 0 citations · 17 references

TL;DR

Findings suggest that, in the absence of external authorship labels, listeners spontaneously form judgments about the creative agent of music that are stably associated with aesthetic evaluation and may constitute an endogenous perceptual bias in the reception of AI-generated music.

Abstract

This study examines listeners’ creator attribution judgments and aesthetic evaluations of AI-generated, human–AI collaborative, and human-composed music under blind-listening conditions. Within a framework of personal compositional style modeling, compositional experience was transformed into executable sampling constraints through natural-language interaction. This process is conceptualized in the present study as “perceptual rule transformation” and was used to generate 18 melodic excerpts. Seventy-one participants with music training completed tasks involving creator attribution judgment, attribution confidence rating, and aesthetic evaluation. The results partially supported H1: a statistically significant but very small association was observed between the actual compositional condition and listeners’ attribution judgments, χ 2 (4) = 23.076, p  < 0.001, Cramér’s V = 0.095. Attribution judgments in the AI-generated and human–AI collaborative conditions were close to a random distribution, and even in the human-composed condition the majority of excerpts (54.7%) were misattributed. Thus, musically trained listeners could not reliably identify the compositional source of the excerpts, with only a modest attribution advantage for human-composed music. Significant differences in aesthetic evaluation were also found across the three compositional conditions. After controlling for actual compositional condition, attribution confidence, and individual participant differences, creator attribution judgment remained independently associated with aesthetic evaluation, and this association held even within the AI-generated condition, where the actual source of all excerpts was constant. These findings suggest that, in the absence of external authorship labels, listeners spontaneously form judgments about the creative agent of music. Such judgments are stably associated with aesthetic evaluation and may constitute an endogenous perceptual bias in the reception of AI-generated music.

DOI Semantic Scholar
Read PDF

Similar papers

#small language model Open access Aug 2026

LifeSciBench: Evaluating Language Models on Realistic, Expert-Level Tasks in the Life Sciences

LifeSciBench is introduced, a benchmark of 750 expert-authored tasks designed to evaluate whether language models can handle realistic life science research work, with each constituent task paired with a human expert-written rubric.

Amelia Liu, Andrew Ho, Anne Marie Droste et al. · 2 citations
#small language model Preprint Aug 2026

Apodex 1.1: Scaling Agentic Intelligence for Complex Work

General-purpose language models can reason and synthesize knowledge, but complex work also requires sustained interaction with files, information sources, and executable code, together with state maintenance, failure recovery, and verifiable delivery. We call this \emph{working capability}: sustained, verifiable progress toward a real-world objective. Apodex 1.1 develops this capability along two complementary dimensions. \emph{Environment Scaling} expands the diversity and verifiability of executable file, search, and code environments, while \emph{Agentic Coordination Scaling} trains agents to decompose long-horizon tasks, delegate parallel work, integrate asynchronous results, and replan. A shared execution harness and AgentOS maintain task state and provenance across tools and agents, and training turns environment trajectories and coordination traces into reliable behavior. Across complex professional work, finance, scientific research, mathematics, coding, and search, Apodex 1.1 reaches the leading performance band despite using a substantially smaller model than many frontier systems. The 35B-parameter Apodex 1.1 Mini further retains strong working capability in a locally deployable form. These results ground agentic intelligence in useful, verifiable work completed over time and advance our goal of building a \emph{Heavy-Duty Solver} for ambitious, long-running tasks.

Apodex Team B. An, B. Li, B. Wang et al. · 1 citation
#small language model Sep 2026

Low Carbon Scheduling of Integrated Energy System Based on Large Language Model-Embedded Multi-Agent Reinforcement Learning

The complex multi-energy coupling characteristics inherent to integrated energy system (IES) present unprecedented challenges for the implementation of low-carbon scheduling. Existing optimization methods often exhibit limitations in system scalability, algorithm adaptivity, and carbon reduction efficacy for complex IES. This paper proposes a Large Language Model (LLM)-Embedded Multi-Agent Reinforcement Learning (LEMARL) to address the aforementioned issues. The proposed method integrates the global perception capability of LLMs with the dynamic optimization capability of MARL. Specifically, the LLM-Embedded module generates high-quality reward functions and policy frameworks from a global perspective, while the MARL module leverages these LLM-generated strategies for distributed interactive iterations—greatly enhancing computation efficiency and scalability. Simulation results demonstrate that LEMARL reduces carbon emissions by 7.76% and simultaneously decreases operating costs by 4.49% in a small-scale IES. Furthermore, LEMARL also exhibits superior applicability and scalability in large-scale IES of the IEEE 141-bus power grid integrated with 51-node thermal system.

Chen Xia, Tong Gou, Yinliang Xu et al. · 1 citation
#small language model Preprint Aug 2026

CritICL: Inference-Time Weak-to-Strong Generalization from Small Language Model Failure Modes

Experimental results show that CritICL consistently outperforms standard in-context learning and achieves performance competitive with or superior to test-time scaling methods, while requiring significantly fewer generations and lower token cost.

Yu-Fan Wu, Yinghui He, Zhengyi Hu et al. · 1 citation
#artificial intelligence Preprint Aug 2026

TestifAI: Tomography-Based Testing for Deep Learning Systems

TestifAI, a deep learning testing framework for efficient and accurate estimation of robustness against combinations of perturbations, is proposed and partial model tomography is introduced, a novel approach to reconstructing model behaviour in a multi-perturbation space from tests that apply only a small number of perturbations.

Arooj Arif, T. Hartung, E. Botoeva et al. · 1 citation

Related blog posts