Jie Song — Sichuan University researcher bridging RL dexterous manipulation and biomedical NLP ontology normalization
2
papers here
2
reviewed by Evolve
1
first-author
0
corresponding
—
works · OpenAlex
—
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-16) 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—
Insufficient data: this axis would measure total publication volume and productivity — paper counts by year, venue, and author role. Only the two reviewed 2026 arXiv papers are available, so no output score can be computed yet.
Impact—
Insufficient data: this axis measures citation-based influence, such as citation counts, h-index, and uptake by later works. None of these metrics are provided for Jie Song, so the axis is left unscored rather than estimated.
Focus0/100
Scores 0/100 because the observed field distribution splits exactly 50/50 between biomedical natural language processing (computational biology) and robotics (dexterous manipulation and animation): the first-authored OntologyAligner sits on the NLP side while the last-authored ConGraspXL sits on the robotics side. With only two reviewed 2026 papers, the perfectly even spread yields the minimum focus score; additional papers concentrated in either field would move it.
Collaboration51/100
Scores 51/100 with 12 distinct coauthors in the aggregate pool, including Zhichuan Xu, Ziyu Lu, Meng Xiao, Cheng Bi, Bairong Shen, and Mirko Meboldt. The collaborator set spans both research lines, pointing to team-based work rather than solo authorship, though the data does not map individual names to specific papers. The mid-range score fits a moderately sized, likely lab-centered network around the two 2026 releases.
Activity—
Insufficient data: this axis measures publishing cadence — papers per year and recency of output over recent years. Only the 2026 papers are observed, so current activity levels cannot be scored honestly.
Reproducibility55/100
Reproducibility mean is 55/100 across the 2 papers rated, both at medium readiness. ConGraspXL reports strong simulated gains (92.7% vs 82.2%) but its quantitative evaluation is simulation-only, limiting external verification; OntologyAligner applies rigorous statistics on a 13,390-sample benchmark, though LLM inference cost and a small HGR gain temper practical reproduction. Code and data availability beyond these Evolve assessments is not captured in the input.
Research summary
Jie Song, affiliated with Sichuan University, has an early-stage profile built from two 2026 arXiv papers reviewed by Evolve, one in each of two distinct fields. On the biomedical side, Song first-authored OntologyAligner, which couples ontology-aligned retrieval with hierarchy-guided large language model reranking and refinement for biomedical ontology normalization, reporting state-of-the-art HPO normalization on a 13,390-sample benchmark with rigorous statistics, tempered by LLM inference cost and a small HGR gain. On the robotics side, Song last-authored ConGraspXL, an extension of GraspXL that adds hierarchical maskable constraint control to reinforcement-learning-based dexterous grasping, achieving strong simulated gains (92.7% vs 82.2%) whose quantitative evaluation remains simulation-only. The observed field distribution splits exactly 50/50 between biomedical natural language processing (computational biology) and robotics (dexterous manipulation and animation), driving the focus score to 0/100. Collaboration scores 51/100 with 12 distinct coauthors in the pool. Reproducibility averages 55/100 across the two papers, both rated medium readiness. Output, impact, and activity axes lack sufficient bibliometric data and are left unscored. Because 'Jie Song' is a common name and no ORCID is linked, this profile is name-resolved from public bibliographic data.
Collaboration pattern
The collaboration score is 51/100, based on 12 distinct coauthors including Zhichuan Xu, Ziyu Lu, Meng Xiao, Cheng Bi, Bairong Shen, and Mirko Meboldt. The pool spans both research lines, suggesting team-based work rather than solo authorship, but per-paper coauthor attribution is absent from the data, so specific collaborators cannot be tied to either 2026 paper. The mid-range score is consistent with a moderately sized, likely lab-centered network.
Research keywords
Highlights
First-authored OntologyAligner advances HPO normalization
Last-authored ConGraspXL adds controllable dexterous grasping
Timeline
- 2026Publishes OntologyAligner (arXiv) as first author, combining ontology-aligned retrieval with hierarchy-guided LLM reranking and refinement for HPO normalization on a 13,390-sample benchmark.
- 2026Publishes ConGraspXL (arXiv) as last author, extending GraspXL with hierarchical maskable constraint control for RL-based dexterous grasping; simulated success rises from 82.2% to 92.7%, with simulation-only evaluation.
- 2026Both releases land in the same year across two fields, producing an exact 50/50 field split (focus 0/100) and a reproducibility mean of 55/100 over 2 papers.
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 scores come from OpenAlex-derived radar data; output, impact, and activity axes lack sufficient data and are left unscored.
- 'Jie Song' is a common name and no ORCID is linked; the profile is name-resolved from public bibliographic data and may conflate same-named researchers.
- Coverage is limited to the two arXiv papers reviewed by Evolve; other publications, venue history, and citations are not reflected.
- Coauthor counts are aggregate only; the data does not map individual coauthors to specific papers.