Skip to content
Log in
#small language model Open access

What students ask matters: LLM interaction depth, task quality, and immediate recall in higher education

V. Tsiligkiris
Aug 2026 · International Journal of Educational Technology in Higher Education · Vol 23 · 0 citations · 40 references

Abstract

Large Language Models (LLMs) are rapidly transforming higher education, yet evidence on how interaction patterns affect learning remains limited. This study examines whether explanation-seeking dialogue with an LLM is associated with task quality and immediate recall in a controlled learning session. Twenty-two postgraduate students completed a pre-test, an LLM-assisted neuroeconomics case task, and an immediate post-test. Fine-grained interaction logs captured per-turn telemetry, enabling extraction of Depth (proportion of "why/how/explain" prompts), Volume, and Pacing features. Results showed large immediate pre–post score gains (Cohen’s dz = 2.12, p < .001). In the task-quality regression model, Depth was positively associated with task quality beyond baseline and Volume (β = 6.27, p = .006), indicating that a one-standard-deviation increase in explanation-seeking prompts was associated with approximately six additional marks on a 0–100 scale. Depth was not associated with immediate recall (β =  − 0.014, p = .728); instead, gain scores were strongly associated with baseline knowledge, consistent with reduced headroom for improvement among students starting from a higher pre-test score (β =  − 0.161, p < .001). The findings indicate a dissociation between performance quality and short-term recall in LLM-supported study. This aligns with cognitive-psychology evidence that elaboration improves comprehension while retrieval practice consolidates retention. Pedagogically, the dissociation suggests that depth-oriented dialogue may need to be paired with deliberate memory-strengthening activities such as self-testing or spaced retrieval if comprehension gains are to translate into recall, although the present design does not test such activities directly. Methodologically, it contributes a replicable, privacy-preserving instrumentation pipeline linking conversational telemetry to learning outcomes. Limitations include single-group design, small sample (n = 22), immediate testing only, and keyword-based depth proxies. Future work should randomise assistance styles, incorporate delayed retention tests, and refine depth measurement through semantic coding.

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

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
#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

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

Related blog posts