Lev Sukherman

Dr. Lev Sukherman

orcid

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications

6 June 2026 in The Education and science journal
Verified
Journal Article
Open access
PDF file
MKMarina Konkol
Marina Konkol
LSLev Sukherman
Lev Sukherman
Introduction. The widespread adoption of generative language models raises questions regarding user competence. Existing component models of AI literacy identify a range of necessary competencies but do not organise them hierarchically. Level-based digital competence frameworks fail to account for the specific characteristics of generative AI, while Bloom’s taxonomy does not encompass the metacognitive, pragmatic, and ethico-social dimensions of human-AI interaction. Aim. The present study aims to develop and theoretically substantiate a hierarchical model of meta-digital competence specific to interactions with generative language models. Methodology and research methods. The study employs theoretical modelling. The hierarchical model was developed through a critical analysis of component AI literacy models, level-based digital competence frameworks, Bloom’s taxonomy, and theories of distributed cognition, metacognition, situated learning, and responsible AI principles. Results. A five-level MDC model was developed, comprising the operational, cognitive, reflexive, pragmatic, and ethico-social levels. Behavioural indicators and achievement criteria were defined for each level. The ethico-social level is positioned at the top of the hierarchy, as ethical responsibility integrates all preceding levels. Scientific novelty . The scientific novelty lies in the development of a hierarchical model specific to generative AI interaction; demonstrating the insufficiency of Bloom’s taxonomy for a comprehensive description of this competence; identifying the pragmatic level, which has no equivalent in existing component AI literacy models; and theoretically justifying the placement of the ethico-social level at the apex of the hierarchy. Practical significance. The practical significance resides in the potential application of the model for designing educational programmes, developing diagnostic tools, and creating assessment procedures aimed at the sequential development of meta-digital competence.

27 April 2026 in Computers
Verified
Journal Article
Open access
PDF file
LSLev Sukherman
Lev Sukherman
YFYetunde Folajimi
Yetunde Folajimi
MKMarina Konkol
Marina Konkol
This paper investigates whether structured collaboration between multiple large language models (LLMs), each assigned a distinct cognitive role grounded in psychological theory, produces benefits beyond simple answer aggregation. We propose the Parallel Synthesis architecture, in which three cognitively specialized roles Analyzer (hierarchical decomposition), Creative (divergent thinking), and Critic (critical evaluation) process each task independently and in parallel, and a Synthesizer integrates their outputs into a final response. To evaluate collaborative reasoning, we introduce the Emergent Reasoning Score (ERS), a composite metric that separates perspective integration (Synthesis Effectiveness) from novel concept generation (Emergent Value). Experiments on the AI2 Reasoning Challenge (ARC-Challenge) (1172 questions) and and the Massive Multitask Language Understanding benchmark (MMLU) (1531 questions) show two consistent findings. First, the architecture achieves high Synthesis Effectiveness (SE=0.711–0.744), indicating reliable integration of all three cognitive perspectives. Second, Emergent Value remains low (EV=0.096–0.112), indicating that synthesis primarily recombines existing concepts rather than generating substantial novel content. A Majority Voting baseline achieves comparable or slightly higher answer accuracy than the Synthesizer on both benchmarks, showing that the architecture’s main contribution lies not in answer selection but in producing integrated reasoning traces that draw on multiple perspectives. These findings suggest that the practical value of cognitively grounded multi-agent architectures lies in reliable perspective integration, while ERS provides a reusable framework for distinguishing integration from genuinely novel reasoning in multi-agent LLM systems. The empirical results reported here constitute a pilot validation of the proposed framework on closed-form benchmarks, intended to establish a proof of concept and motivate larger-scale evaluation.

26 July 2024
Verified
Conference Paper
Open access
MMMixer Mark
Mixer Mark
SMSteven Morrow
Steven Morrow
LSLev Sukherman
Lev Sukherman