Stefano Rini

National Yang Ming Chiao Tung University · Italy

Unclaimed · AI-structured profileORCID OpenAlex

Knowledge-graph QA researcher focused on traceable multi-hop graph navigation and small-language-model evaluation.

papers here

reviewed by Evolve

first-author

corresponding

—

works · OpenAlex

h-index · OpenAlex

Bibliometric figures come from OpenAlex (2026-10-15) and cover the author's entire public record, not just the papers listed here. Reproducibility signal comes from Evolve's own reviews.

Research profile

Output—

Not scored: this snapshot lacks the complete publication and venue data the output axis needs. It would measure total works, annual counts, and venue types; partial counts-by-year figures exist in the raw record, but a full output assessment is pending.

Impact—

Not scored: citation-based impact data is insufficient in this snapshot. The axis would measure citation totals, citation trajectory, and influence on follow-up work; raw indicators such as h-index 16 exist, but no full impact score has been computed here.

Focus0/100

Focus scores 0, reflecting exceptional breadth rather than concentration. The OpenAlex topic profile spans at least five clusters — cooperative communication and network coding (67 works), distributed sensor networks and detection (40), wireless communication security (37), error-correcting codes (28), and stochastic gradient optimization (27) — and the current field split (0.5 knowledge-graph NLP, 0.5 KGQA/LLM reasoning) layers a newer NLP direction on top of the communications base.

Collaboration32/100

Collaboration scores 32, indicating a modest recent co-authorship footprint: 4 distinct coauthors appear in the three-year window. The recurring partners are Eduin E. Hernandez, Luis F. Garcia, Nurassyl Askar, and Sergio A. Diaz. Both reviewed 2026 papers are multi-author works in which Rini sits in the last author position without a corresponding-author flag, consistent with a senior supervisory role within a small, stable team rather than large consortium work.

Activity100/100

Activity scores 100, the maximum on the radar. The OpenAlex counts-by-year record runs unbroken from 2009 (4 works) to 2026 (15 works), with 16 works in both 2022 and 2025, and 12 in each of 2020 and 2021; the last three years total 44 works. This is sustained, near-constant publishing across nearly two decades rather than a single burst, and output remains high through 2025–2026.

Reproducibility78/100

Reproducibility scores 78, computed from Evolve's review of 2 papers (reproMean 78). Readiness is rated high for 'Theseus in the Graph', which ships gold paths and fidelity metrics as a usable benchmark, and medium for 'The Path Matters'. The verdicts temper the score: the Theseus benchmark excludes the closest prior baselines, and the SLM study runs on small benchmarks without significance tests.

Research summary

Stefano Rini is a researcher at National Yang Ming Chiao Tung University whose OpenAlex-resolved record spans 2009–2026 with an h-index of 16 and an i10-index of 35, sustained annual output, and 44 works in the last three years. The bibliometric topic profile is rooted in information and communication sciences — cooperative communication and network coding (67 works), distributed sensor networks and detection (40), wireless communication security (37), error-correcting codes (28), and stochastic gradient optimization (27) — while the current field distribution in this snapshot splits evenly between natural language processing on knowledge graphs and KGQA/LLM reasoning. That split tracks a recent, active line on knowledge graph question answering: two 2026 arXiv papers reviewed by Evolve list Rini as last author. 'Theseus in the Graph' reframes KGQA as question-conditioned graph navigation with gold paths and fidelity metrics, rated high for reproducibility readiness; 'The Path Matters' shows path-based evaluation can rank small language models differently from answer accuracy alone, rated medium. Evolve credits controlled, sound designs while flagging excluded baselines and small benchmarks without significance tests. Radar scores capture peak activity (100), broad topical focus (0), moderate collaboration (32; 4 distinct recent coauthors), and reproducibility (78) across 2 scored papers.

Collaboration pattern

Rini's collaboration score is 32, with 4 distinct coauthors in the recent three-year OpenAlex window. The recurring names are Eduin E. Hernandez, Luis F. Garcia, Nurassyl Askar, and Sergio A. Diaz. Both catalog papers are multi-author arXiv works where Rini occupies the last author slot without a corresponding-author flag — a pattern consistent with senior supervision of a small, stable team rather than large consortium projects or solo authorship.

Research keywords

knowledge graph question answeringmulti-hop reasoninggraph navigationpath fidelitysmall language modelsLLM evaluationtraceabilityreinforcement learning agentscooperative communicationsnetwork codingerror-correcting codesstochastic gradient optimization

Highlights

Theseus in the Graph: KGQA as traceable graph navigation
The 2026 arXiv paper reframes knowledge graph question answering as question-conditioned graph navigation, pairing gold paths with fidelity metrics; Evolve rates reproducibility readiness high and calls the benchmark solid, while noting that the closest prior baselines were excluded.
Path fidelity can reorder small-language-model rankings
In 'The Path Matters', Evolve's controlled-study verdict finds that KG navigation fidelity and answer accuracy can rank small language models differently, with Path Edit Distance among the metrics; the design is sound, though benchmarks are small and significance tests absent (medium readiness).

Timeline

2026'Theseus in the Graph: Towards Traceable Multi-Hop Graph Navigation' appears on arXiv; Evolve's review highlights gold paths, fidelity metrics, and high reproducibility readiness.
2026'The Path Matters: Evaluating Small Language Models Beyond Answer Accuracy in KGQA' follows, showing that path fidelity can rank small language models differently from answer accuracy alone.
2026Both reviewed works list Stefano Rini as last author, marking a senior supervisory role within this knowledge-graph reasoning line of the Evolve catalog.

Papers in the Evolve catalog

Frequent co-authors

Eduin E. Hernandez×2 Luis F. Garcia×2 Nurassyl Askar×2 Sergio A. Diaz×2

About this profile

This profile is AI-structured by Evolve.Pub from public bibliographic data (OpenAlex) and Evolve's own paper reviews. It is not written or endorsed by the researcher unless the profile is claimed. Radar dimensions are computed from real bibliometrics with fixed formulas — raw figures are shown alongside; missing dimensions mean insufficient data, never zero. Errors? Claim the profile to correct or hide it.

Bibliometric indicators (h-index 16, i10-index 35, counts-by-year 2009–2026, topics) come from an OpenAlex identity resolution; citation-based output and impact axes are not scored in this snapshot.
The Evolve catalog contains only two reviewed papers, both 2026 arXiv works in knowledge-graph NLP; the earlier communications and coding output is not covered by reviewed-paper detail.
This profile is name-resolved from public bibliographic data (ORCID 0000-0003-1681-3316); no biography, awards, or career history beyond the snapshot is included.
The source record lists country code 'IT' while the named institution is National Yang Ming Chiao Tung University; affiliation details should be verified before reuse.