Nuo Chen
The Hong Kong University of Science and Technology (Guangzhou)
Machine learning researcher at HKUST (Guangzhou) spanning LLM post-training and multimodal AI for physics
2
papers here
2
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
—
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-02) 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: no publication-count record is available in this snapshot, so output is unscored. This axis would measure the volume and venue distribution of the researcher's publications over time.
Impact—
Insufficient data: no citation counts or h-index are available, so impact is unscored. This axis would measure how widely the reviewed papers are cited and taken up by subsequent work.
Focus0/100
The focus score is 0, reflecting an even 50/50 split in the field distribution between LLM post-training with reinforcement learning and AI for physics with scientific multimodal learning. Across the two reviewed papers, no single subfield dominates, which suggests a deliberately dual-track agenda rather than a concentrated specialty; as more papers are catalogued, this spread may either consolidate or widen.
Collaboration53/100
The collaboration score is 53, based on 13 distinct coauthors across the two catalogued papers. Names such as Henan Sun, Haitao Hu and Jianfeng Zhang recur across both works, pointing to a small stable core team. Both papers place this researcher in a middle author position rather than first or corresponding, so the profile reads as a consistent contributor within group projects rather than a solo driver.
Activity—
Insufficient data: only two 2026 preprints are on record, with no full publication timeline, so activity is unscored. This axis would measure recent publication cadence and research tempo across years.
Reproducibility20/100
The reproducibility score is 20, the mean over 2 scored papers. Evolve rates both the DCRL post-training paper and the PhyMo multimodal-modality paper as low reproducibility readiness: DCRL is flagged for thin theory and missing key ablations, and PhyMo for trivial theory and no released code. Releasing code, ablation tables and experimental detail would materially raise future scores.
Research summary
Nuo Chen is a machine-learning researcher at The Hong Kong University of Science and Technology (Guangzhou). The Evolve catalog holds two 2026 arXiv preprints with this name, and the field distribution splits evenly (0.5/0.5) between LLM post-training with reinforcement learning and AI for physics with scientific multimodal learning. The first paper, DCRL, proposes decoupling and coupling reinforcement learning via policy-reward manifold alignment: policy and reward co-evolve through shared weights and evolving rubrics, and Evolve notes a 4B model outperforming a 32B one, while flagging thin theory and missing key ablations. The second, PhyMo, frames PDE-structured physical fields as a new modality for multimodal AI4Physics, with promising multi-dataset gains but thin margins, trivial theory and no released code. Neither paper lists this researcher as first or corresponding author; both are middle-position contributions. Collaboration data records 13 distinct coauthors, with Henan Sun, Haitao Hu and Jianfeng Zhang recurring across both works, and a moderate collaboration score of 53. Output, impact and activity axes remain unscored because the underlying bibliometric data is insufficient. As the name is shared by multiple researchers, this profile is name-resolved from public bibliographic data. Overall it is an early-stage profile of a contributor moving along two parallel machine-learning tracks — post-training methods and scientific multimodality — whose reproducibility footprint (score 20 across 2 papers) currently depends on code and ablation release.
Collaboration pattern
Across 13 distinct coauthors on two 2026 arXiv papers, a recurring core of Henan Sun, Haitao Hu and Jianfeng Zhang appears on both works, suggesting one or two stable teams. This researcher sits in middle author positions on both, neither first nor corresponding, consistent with a contributing member rather than project lead. The collaboration score of 53 reflects a moderate, team-based pattern rather than a large distributed network.
Research keywords
Highlights
DCRL: a 4B model beats 32B by co-evolving policy and reward
PhyMo: physical fields as a new modality for AI4Physics
Timeline
- 2026Co-authored DCRL on arXiv, a reinforcement-learning post-training method that aligns policy and reward through shared weights and evolving rubrics.
- 2026Co-authored PhyMo on arXiv, framing PDE-structured physical fields as a new multimodal modality for AI4Physics.
- 2026Both catalogued papers place this researcher as a middle author, with the field distribution split evenly between LLM post-training and scientific multimodal learning.
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.
- Radar dimensions and co-author statistics derive from public bibliometric records only; no citation counts, h-index or venue-tier data is included in this snapshot.
- Output, impact and activity are unscored because the underlying publication, citation and cadence data was insufficient.
- The catalog holds only two 2026 arXiv preprints, both with middle author positions, so this researcher's specific contribution to each work cannot be attributed from the data.
- "Nuo Chen" is a common name; this profile is name-resolved from public bibliographic data and may merge or split same-name researchers.