Robotics researcher advancing demonstration-driven manipulation and language-guided social navigation
2
papers here
1
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
—
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—
Publication output metrics such as paper counts, venue types, and productivity over time are not yet available in the underlying data, so this axis is unscored; it would measure cumulative and per-year publication volume across indexed sources.
Impact—
Impact indicators such as citation counts, h-index, and downstream adoption are not available in the current snapshot, so this axis is unscored; it would measure how often and how widely this researcher's work is cited and built upon by others.
Focus100/100
The focus axis scores 100 out of 100: the field distribution registers a single field, robotics (robot manipulation and learning from demonstration). Both reviewed 2026 papers stay within robotics — one on manipulation from demonstration via agentic programming, the other on social navigation and human-robot interaction — indicating a tightly concentrated agenda around language- and demonstration-guided robot behavior.
Collaboration41/100
The collaboration axis scores 41, based on 7 distinct coauthors in the underlying data. Both reviewed 2026 papers are team efforts: David Hsu appears in a middle author position on RAPID and in the last position on the group-joining paper without a corresponding-author flag. Recurring collaborators include Zilin Fang, Zishuo Wang, Gim Hee Lee, Yuyao Liu, Jiayuan Mao, Leslie Pack Kaelbling, and Tomás Lozano-Pérez, suggesting small, stable robotics teams.
Activity—
Recent-activity data beyond the two reviewed 2026 arXiv preprints is insufficient — there is no multi-year publication cadence to evaluate — so this axis is unscored; it would measure ongoing output and research engagement in recent years.
Reproducibility20/100
The reproducibility axis scores 20, the mean over 1 assessed paper. Evolve rates RAPID's reproducibility readiness as low, citing thin statistics and reliance on proprietary models, and the group-joining paper as medium, noting a small 42-scene benchmark and no released code. Together these indicate promising results whose methods and benchmarks are not yet straightforward to reproduce independently.
Research summary
David Hsu is a robotics researcher whose reviewed work centers on making robot behavior programmable through demonstration and socially competent through language. The Evolve catalog holds two 2026 arXiv papers. The first, RAPID (Robot Agentic Programming from Demonstrations), infers the task specification, manipulation primitives, and a simulator-based verifier from a single demonstration, reaching 75.9% against a 14.6% baseline across 8 tasks; Evolve cautions that thin statistics and proprietary model dependencies temper certainty. The second formulates language-guided robot group joining, predicting joining goals from natural-language instructions over F-formation structure with spectral clustering, and reports 89.76% pose validity against vision-language-model baselines on 42 scenes alongside a favorable F1/latency trade-off. Bibliometrically the profile is extremely focused: the field distribution registers a single field, robotics (robot manipulation and learning from demonstration), and the focus score is 100. The collaboration score is 41 with 7 distinct coauthors, and the reproducibility score is 20, reflecting low-to-medium reproducibility readiness across the two papers. Since "David Hsu" is a common name, this profile is name-resolved from public bibliographic data; output, impact, and activity metrics are not yet available.
Collaboration pattern
Both reviewed papers are multi-author efforts with 7 distinct coauthors overall (collaboration score 41). On RAPID (2026) David Hsu sits in a middle author slot; on the group-joining paper, in the last slot without a corresponding-author flag. The recurring coauthor list — Zilin Fang, Zishuo Wang, Gim Hee Lee, Yuyao Liu, Jiayuan Mao, Leslie Pack Kaelbling, Tomás Lozano-Pérez — suggests small, recurring robotics teams spanning manipulation and social navigation.
Research keywords
Highlights
RAPID: programming robots from a single demonstration
Language-guided robot group joining
A language-and-demonstration research program
Taken together, the two 2026 papers use natural language and human demonstration as the interface to robot behavior: one compiles a demonstration into verifiable manipulation programs, the other grounds language goals into socially valid motion. Both are early arXiv-stage results with reproducibility caveats.
Timeline
- 2026RAPID: Robot Agentic Programming from Demonstrations appears on arXiv, with David Hsu in a middle author position.
- 2026Where Should I Join? Robot Group Joining via Language-Guided Goal Prediction appears on arXiv, with David Hsu in the last author position.
- 2026The catalog's collaboration snapshot records 7 distinct coauthors, including Zilin Fang, Jiayuan Mao, Leslie Pack Kaelbling, and Tomás Lozano-Pérez.
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 are computed from an OpenAlex-based snapshot only; the output, impact, and activity dimensions lack sufficient data and are left unscored.
- "David Hsu" is a common name; this profile is name-resolved from public bibliographic data and the two reviewed arXiv papers, with no ORCID or affiliation provided to confirm identity.
- The reproducibility score rests on a single assessed paper (reproPapers = 1) and may not represent the broader body of work.
- Both catalog entries are 2026 arXiv preprints; no peer-reviewed venue information is included in the data.