Yu Ma

Beijing Institute of Technology · China

Unclaimed · AI-structured profileOpenAlex

Yu Ma (Beijing Institute of Technology): early-stage work in 6G token communication and LLM enterprise data agents

papers here

reviewed by Evolve

first-author

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—

Publication volume and career output trajectory cannot yet be scored: the snapshot lists no verified total publication count, and the indexed record rests on two 2026 arXiv preprints. This axis would measure cumulative output, venue spread, and authorship roles.

Impact—

Citation impact cannot be scored: the snapshot reports h-index 0 and i10-index 0, which for freshly released 2026 preprints reflects a citation window too short rather than measured influence. This axis would track citations, h-index growth, and downstream uptake.

Focus0/100

Evolve's focus score is 0, mirroring an exactly even field distribution (0.5 wireless/semantic communication vs 0.5 NLP/enterprise information systems) across the indexed record. Topic counts — wireless (3), video coding (2), video analysis (2), optical wireless (1), IoT/data (1) — confirm a deliberately dual-track, breadth-first portfolio rather than a single narrow specialty.

Collaboration62/100

The collaboration score is 62, based on 21 distinct coauthors in the indexed record and a frequent-coauthor list of 12, including senior wireless figures such as Kaibin Huang, Jiangzhou Wang, Tony Q. S. Quek, Rahim Tafazolli, and Sheng Chen. The pattern suggests team-based work bridging two communities, though corresponding-author roles are not yet evident in the two reviewed papers.

Activity40/100

The activity score is 40: the snapshot records 3 works in the last three years, all concentrated in 2026 (countsByYear shows 2026: 3), with no indexed output in earlier years. Recent activity is real but thin and front-loaded into a single year, leaving the longer-term publishing cadence unverified.

Reproducibility20/100

Reproducibility scores 20, a reproMean of 20 averaged over 2 papers assessed. Both reviewed works — the UniDataAgent system and the token-communication survey — carry low repro-readiness ratings: the former for a small, single-site evaluation, the latter because a conceptual paradigm survey has no artifact to rerun.

Research summary

Yu Ma is a researcher at Beijing Institute of Technology whose indexed record splits evenly (0.5/0.5) between two fields: communication engineering — wireless, 6G, and semantic communication — and natural language processing applied to enterprise information systems. The 2026 record contains three works (recent3yWorks: 3; countsByYear 2026: 3) built with 21 distinct coauthors. The most visible first-author output is an arXiv survey, 'From Semantic to Token Communication,' which argues that tokens, not bits, are the missing communication unit for large-model-driven 6G intelligent connectivity; Evolve rates its taxonomy thorough and timely while noting the paradigm claim remains conceptual. The second cataloged work, UniDataAgent, is a co-authored ontology-grounded LLM agent for enterprise question-to-report automation, with reported gains of 95.0% versus 72.5% for a RAG baseline, though the evaluation is small, single-site, and baseline-thin. Topic counts (wireless 3, video coding 2, video analysis 2, optical wireless 1, IoT/data 1) show depth in advanced wireless communication with secondary multimedia activity. Impact metrics are not yet meaningful: h-index and i10-index both read 0 for freshly posted preprints. Reproducibility is low (reproMean 20 over 2 assessed papers), and focus scores 0 given the exact two-field split. This is an early-stage, name-resolved profile whose trajectory is worth re-checking as venue-filtered publications accumulate.

Collaboration pattern

Yu Ma's indexed record shows 21 distinct coauthors and a 12-person frequent-coauthor core spanning two communities: a wireless/6G cluster including Zhen Gao, Wenjun Xu, Xiaodong Xu, Yin Xu, Mahdi Boloursaz Mashhadi, and senior figures Kaibin Huang, Jiangzhou Wang, Rahim Tafazolli, Sheng Chen, and Tony Q. S. Quek, plus Li Qiao and Xiaoyuan Zhang. Collaboration scores 62; the pattern is large-team and senior-anchored, with no corresponding-author role evident in the two reviewed papers.

Research keywords

semantic communicationtoken communication6G intelligent connectivitylarge-model-driven communicationjoint source-channel codingmultimodal tokenizationLLM agentontology groundingquestion-to-report automationvideo coding and compressionvideo analysis and summarizationoptical wireless communication

Highlights

First-author survey stakes out 'token communication' for 6G
In a 2026 arXiv survey, Yu Ma (first author) argues that tokens — not bits — are the missing communication unit above bits for large-model-driven 6G intelligent connectivity, with a taxonomy spanning multimodal tokenization and joint source-channel coding. Evolve's verdict: thorough and timely taxonomy, but the paradigm claim is conceptual and flagship results are co-authored.
Co-authored UniDataAgent reports 95.0% vs 72.5% over RAG
The second cataloged paper positions Yu Ma as a middle author on UniDataAgent, an ontology-grounded LLM agent for enterprise question-to-report automation, deployed with a reported 95.0% result versus 72.5% for a RAG baseline. Evolve flags the evaluation as small, single-site, and baseline-thin.
A deliberately dual-track 2026 portfolio
The indexed record splits exactly 50/50 between communication engineering (wireless/6G, semantic communication) and NLP/enterprise information systems, spanning topics from token communication to video coding and analysis (3/2/2/1/1 across five topic clusters) — a breadth that explains the focus score of 0.

Timeline

2026First-author arXiv survey 'From Semantic to Token Communication' proposes tokens as the communication unit above bits for 6G.
2026Co-authors UniDataAgent, an ontology-grounded agent automating enterprise question-to-report workflows, reporting 95.0% versus 72.5% for RAG.
2026The indexed year totals 3 works, 21 distinct coauthors, and an even 50/50 field split between wireless communication and NLP/enterprise systems.

Papers in the Evolve catalog

2026-09-23
UniDataAgent: An Ontology-Grounded Agent for Enterprise Question-to-Report Automation
2026-09-09first author
From Semantic to Token Communication: The Next Paradigm for Large-Model-Driven 6G Intelligent Connectivity

Frequent co-authors

Zhen Gao×1 Li Qiao×1 Xiaoyuan Zhang×1 Mahdi Boloursaz Mashhadi×1 Yin Xu×1 Wenjun Xu×1 Xiaodong Xu×1 Kaibin Huang×1 Jiangzhou Wang×1 Rahim Tafazolli×1 Sheng Chen×1 Tony Q. S. Quek×1

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 (coauthor, topic, and year counts, h-index, reproducibility scores) come from an OpenAlex-derived snapshot only; 'Yu Ma' is a common name, so this profile is name-resolved from public bibliographic data and namesake conflation is possible (no ORCID linked).
Both cataloged papers are 2026 arXiv preprints; citation and venue filters have not accrued, so h-index 0 and i10-index 0 reflect timing rather than measured impact.
The output and impact axes are unscored (insufficient data): no verified total publication count or career trajectory exists in this snapshot.
The reproducibility mean of 20 averages only 2 assessed papers, so it should be read as a coarse estimate.