Researchers Develop Automatic Detection Pipeline for Chinese Web Neologisms Using Prompt Engineering
Researchers have published a four-stage automated pipeline for detecting Chinese web neologisms by encoding traditional linguistic principles as prompt-engineering skills for large language models. The system was applied to a 267-million-document corpus, yielding nearly 5,000 labeled neologisms, and evaluated against an existing lexicon using a novel per-stage recall decomposition method. The work identifies key bottlenecks in automated neologism detection and releases all code, outputs, and evaluation protocols publicly.
A team of researchers has proposed an automated Chinese web neologism detection pipeline that translates classical linguistic identification principles into prompt-engineering skills for LLMs. The four-stage system begins with tokenizer-independent character n-gram candidate generation, followed by dictionary anchoring with a Pointwise Mutual Information pre-filter, a well-formedness check grounded in Chinese word-formation principles, and a final combined rule and three-way classification stage distinguishing neologisms, named entities, and non-words. Applied to the BAAI CCI 3.0 corpus of 267 million documents, the pipeline produced 226,959 classified candidates, including 4,853 labeled neologisms. Evaluation against the Hou (2023) lexicon of 4,199 entries using a per-stage conditional recall decomposition revealed two primary bottlenecks: Stage 1 candidate coverage (recall 41.5%) and Stage 4B LLM semantic judgment (recall 60.0%), while intermediate stages were near-lossless. A length-stratified analysis showed that the structural well-formedness skill is robust across word lengths (≥96.9%), whereas semantic novelty classification degrades notably with longer candidates (65.6%, 59.0%, and 44.1% for 2-, 3-, and 4-character words respectively), delineating a current ceiling for skill-based linguistic operationalization. All methods, pipeline outputs, and evaluation protocols are released as open resources.
What's missing
The paper does not report inter-annotator agreement for the neologism labels assigned to the 4,853 candidates, leaving the reliability of the gold-standard output uncertain.
What different sources said
- arXiv cs.CLCenter
Operationalizing Linguistic Methods through Prompt-Engineering Skills: An Automatic Chinese Web Neologism Detection Pipeline
Related
Gut Bacteria Enzyme Found to Break Down Heat-Processed Food Compounds, Producing Novel Biogenic Amines
Researchers have discovered that an enzyme in common gut bacteria can degrade N-epsilon-carboxymethyllysine (CML), a compound formed during thermal food processing, producing previously unknown biogenic amines. The enzyme, ornithine decarboxylase SpeC from enterobacteria, acts on CML and related modified lysine derivatives through a low-level 'underground' catalytic activity. This finding suggests a previously unrecognized communication axis between thermally processed dietary compounds and gut microbial physiology, with potential implications for host health.
Full-Length Gene Sequencing Reveals Two Distinct Bacterial Communities in Black-Legged Ticks Expanding Into Canada
Researchers used Oxford Nanopore full-length 16S rRNA gene sequencing to characterize the microbiome of Ixodes scapularis black-legged ticks collected in Nova Scotia, Canada, distinguishing between tick-adapted bacteria and environmentally acquired bacteria. The study comes as I. scapularis — the primary vector of Lyme disease — is rapidly expanding northward into Canada due to climate change. The findings suggest that environmentally derived bacteria in tick microbiomes are not mere contamination, which has implications for how tick microbiome data is collected and interpreted across surveillance studies.
Study Identifies Metabolic Link Between Cell Envelope Stress and Biofilm Formation in Bacteria
Researchers have discovered that the metabolite acetyl-CoA directly inhibits enzymes that degrade the bacterial signaling molecule c-di-GMP, connecting cell envelope biosynthesis stress to biofilm formation in Pseudomonas aeruginosa. The study found that sub-inhibitory concentrations of antibiotics targeting early peptidoglycan biosynthesis — but not other antibiotic classes — elevate c-di-GMP levels by reducing phosphodiesterase activity, with acetyl-CoA competing for the enzyme active site. Because the relevant enzyme domain is broadly conserved across bacterial species, this checkpoint mechanism may be widespread and could have implications for understanding antibiotic-induced biofilm responses.