Early-career researcher validating LLM synthetic social research: benchmarks for 'silicon sampling'
2
papers here
0
reviewed by Evolve
1
first-author
0
corresponding
—
works · OpenAlex
2
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-09-28) 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—
Output cannot be scored yet: publication volume beyond the recent three-year window is unavailable, and indexed works are too few to characterize productivity. This axis would measure total publication count, output consistency, and venue quality across the full career.
Impact—
Impact cannot be scored yet: with an i10-index of 0 and an h-index of 2, citation-based signals are too thin to assess influence. This axis would measure citation counts, citing works, and uptake of the benchmark framework by other groups.
Focus—
Focus cannot be scored yet: OpenAlex topic labels, led by 'Embodied and Extended Cognition' (5 works), diverge from the reviewed LLM-validation papers, so no coherent focus profile emerges. This axis would measure thematic concentration across the full publication record.
Collaboration44/100
The collaboration score of 44 reflects a compact, early-stage network: 8 distinct coauthors are recorded, with Min Jun Jung, Florian Kutzner, Celina Kacperski, Laura de Molière, Felix P. S. Wallis, and James K. He recurring across works. Both the first-author benchmark paper and the framework paper are multi-authored, suggesting team-based computational social science rather than solo work; repeated co-authorship with Wallis and He points to a small stable core.
Activity67/100
The activity score of 67 is supported by 9 works over the recent three-year window: 2 in 2024, rising to 4 in 2025, and 3 so far in 2026, including both arXiv papers reviewed by Evolve. This steady, recent cadence indicates an actively publishing early-career researcher, though the short window means long-term productivity cannot yet be judged.
Reproducibility—
Reproducibility cannot be scored yet: only two papers are rated, one 'medium' and one 'low' on repro-readiness — too small a sample. This axis would measure code release, data availability, and protocol transparency across outputs.
Research summary
Edoardo Chidichimo is an early-career researcher whose reviewed work centres on validating LLM-based synthetic research — 'silicon sampling', in which large language models stand in for human survey respondents. Evolve has reviewed two of his 2026 arXiv papers. In the Artificial Societies Benchmark he is first author, proposing an 11-test scorecard that validates synthetic-survey research against 20 human sources across 9 models; Evolve judges the design credible but notes key cells rest on single generations. A second, co-authored paper argues that synthetic samples be validated against real behaviour and subgroup fairness; it is well argued but untested. The OpenAlex snapshot records an h-index of 2, an i10-index of 0, and 9 works in the recent three-year window (2 in 2024, 4 in 2025, 3 in 2026), with 8 distinct coauthors. Its topic labels — led by 'Embodied and Extended Cognition' (5 publications), followed by complex systems and decision making, cognitive mapping, education research, and action observation — sketch a broader cognitive-science profile than the reviewed papers alone, which may reflect genuine breadth or name ambiguity; no institution or country is recorded. Only collaboration (44) and activity (67) are currently scoreable. Overall this is a full but early-stage profile: a small, recent, methodologically ambitious body of work whose citation footprint has yet to develop.
Collaboration pattern
Chidichimo publishes in small multi-author teams. The OpenAlex record lists 8 distinct coauthors over the recent window, with Min Jun Jung, Florian Kutzner, Celina Kacperski, Laura de Molière, Felix P. S. Wallis, and James K. He recurring — Wallis and He appear twice in the co-author list, suggesting repeated collaboration within a stable core. His first-author benchmark paper shows he can lead collective projects, while the second paper is noted by Evolve as vendor-authored.
Research keywords
Highlights
First-author benchmark for validating synthetic research
Framework for behavioural validation and subgroup fairness
Steady three-year publication cadence
OpenAlex records 9 works in the recent three-year window — 2 in 2024, 4 in 2025, 3 in 2026 — supporting an activity score of 67.
Cognitive-science topic base
OpenAlex topics are led by 'Embodied and Extended Cognition' (5 publications), with complex systems and decision making, cognitive mapping, education research, and action observation also represented — broader than the two reviewed LLM-validation papers.
Timeline
- 20242 indexed works appear in the OpenAlex record, marking the start of the visible publication record.
- 2025Indexed output doubles to 4 works in the year.
- 20263 indexed works so far, including both Evolve-reviewed arXiv papers: the first-author Artificial Societies Benchmark and the co-authored validation-framework paper.
Papers in the Evolve catalog
Frequent co-authors
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 2, i10-index 0, per-year counts, topics) come from OpenAlex only; no other citation source is represented.
- Institution and country are absent from the input data; this profile is name-resolved from public bibliographic data and the provided ORCID.
- OpenAlex topic labels (e.g. embodied and extended cognition) diverge from the reviewed papers' LLM-validation focus, which may reflect name ambiguity or genuine breadth.
- Only two Evolve-reviewed papers are available; quality and repro-readiness assessments cover this sample alone.