Early-career researcher in LLM-agent long-term memory and graph-based retrieval-augmented reasoning (2026)
2
papers here
2
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
—
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-09-20) 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—
No publication counts or productivity totals are available yet, so output cannot be scored. This axis would measure the volume and cadence of publications, separating arXiv preprints from peer-reviewed venue papers, once a fuller record is linked to this ORCID.
Impact—
Citation counts and h-index are absent from the input data, so impact is not scored. This axis would measure how widely the two reviewed 2026 preprints are cited and taken up by the LLM-agent and graph-learning communities.
Focus0/100
The focus score is 0, reflecting an even 0.5/0.5 split across two field labels: NLP on LLM-agent long-term memory and personalization, and machine learning on graph representation learning and retrieval-augmented generation. In practice the two reviewed 2026 papers share heavy topical overlap — graph neural networks and RAG appear in both keyword lists — so the low score signals breadth across field labels rather than scattered unrelated work.
Collaboration56/100
Collaboration scores 56, based on 16 distinct coauthors in the raw record. Twelve recurring partners are listed, including Siyu An, Junnan Dong, Xing Sun, Cai Ke, Xiaoyu Shen, Ruifeng Xu, Linhao Luo, and Keyu Chen. Both catalog papers carry middle-position authorship without corresponding role, consistent with contributing specialist work inside larger LLM-agent and graph-learning teams rather than leading an independent group.
Activity—
No yearly publication series beyond the two 2026 preprints is provided, so activity cannot be scored. This axis would measure research cadence — papers per year and gaps between releases — once a complete publication history is attached.
Reproducibility20/100
Reproducibility scores 20 (reproMean over 2 papers), and both reviewed papers carry low reproReadiness. Evolve's verdicts cite concrete gaps: the latent-memory paper lacks significance tests, a stated judge, and released code, while the WFM paper reports a claimed 10.5x speedup with absent statistics and a leaked 'planning values' note. No code artifact is referenced for either.
Research summary
Ruizhi Qiao (ORCID 0000-0002-3663-0149) is an early-stage researcher whose profile is name-resolved from public bibliographic data; the current snapshot records no institution or country. The Evolve catalog holds two 2026 arXiv preprints, both with middle-position authorship and no corresponding role. The work spans two evenly weighted fields (0.5 and 0.5): natural language processing on LLM-agent long-term memory and personalization, and machine learning on graph representation learning and retrieval-augmented generation. The first paper, Disentangling Long-Term Memory via Latent Neuro-Symbolic Reasoning, builds on sparse autoencoders and query-conditioned latent graphs; Evolve's review reports large gains on personalization benchmarks but flags missing significance tests, an unstated judge, and no released code. The second, WFM: Wiki Foundation Model for Complex Agentic Reasoning, proposes a graph foundation model over dense LLM-Wiki graphs with a claimed 10.5x speedup and state-of-the-art QA gains; the review cites a leaked 'planning values' note and absent statistics as trust concerns. Both papers carry low reproducibility readiness (reproMean 20 across 2 papers). Collaboration is moderate (score 56) with 16 distinct coauthors, including recurring partners such as Siyu An, Junnan Dong, Xing Sun, and Linhao Luo. Output, impact, and activity metrics are not yet available.
Collaboration pattern
Ruizhi Qiao works within sizeable teams: the raw record lists 16 distinct coauthors and 12 recurring partners, including Siyu An, Junnan Dong, Xing Sun, Cai Ke, Xiaoyu Shen, Ruifeng Xu, Linhao Luo, and Keyu Chen. Collaboration scores 56. In both catalog papers the researcher appears in a middle author position without corresponding role, suggesting a contributor-specialist pattern inside LLM-agent and graph-learning groups rather than a lab-lead profile.
Research keywords
Highlights
Query-conditioned latent graphs for agent long-term memory
WFM: a graph foundation model over dense LLM-Wiki graphs
Timeline
- 2026Co-authored (middle position) 'Disentangling Long-Term Memory via Latent Neuro-Symbolic Reasoning', posted on arXiv in the LLM-agent memory and personalization field.
- 2026Co-authored (middle position) 'WFM: Wiki Foundation Model for Complex Agentic Reasoning', extending into graph representation learning and retrieval-augmented generation.
- 2026Both Evolve-reviewed 2026 papers were assigned low reproReadiness, giving a reproMean of 20 across the 2-paper record.
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.
- Bibliometrics and radar scores are computed from OpenAlex-based bibliometric data only; institution and country are unrecorded in this snapshot.
- Identity is name-resolved from public bibliographic data (ORCID 0000-0002-3663-0149); no biographical or affiliation details are verified.
- Output, impact, and activity dimensions are unscored because citation counts, publication totals, and yearly activity data were unavailable.
- Qualitative assessments rest on the two Evolve-reviewed 2026 arXiv preprints; no peer-reviewed venue record is included in this profile.