Chengzhi Zhang

Collaboration between academic and industrial institutions significantly boosts the novelty of NLP research outputs, challenging the effectiveness of isolated industrial efforts. WHY_IT MATTERS: This insight could reshape strategies for fostering impactful research collaborations between academia and industry, ultimately enhancing the quality and innovation of academic publications.

The shift from conceptual to empirical research in LIS reveals a growing focus on user-centered topics, reshaping the landscape of information science inquiry.

Combining text, images, and audio from academic papers can dramatically boost keyword extraction accuracy, revealing the untapped potential of multimodal data.

Research papers in Chinese LIS are becoming more novel, with collaboration patterns revealing that solo authorship is linked to lower novelty.

More than half of algorithm mentions in NLP papers are for direct use, signaling a significant shift in how researchers engage with algorithms over time.

The rise of pre-trained language models has not only reshaped NLP innovation but also intensified the knowledge demands on researchers, with implications for future research directions.

An optimal number of thought leaders can enhance team impact, but too many may stifle innovative ideas.

Formulaic expression desensitization not only boosts dataset scale but also enhances model generalization in extracting key sentences from scientific literature.

EEG signals can dramatically enhance keyphrase extraction from microblogs, outperforming traditional eye-tracking methods alone.

Moderately difficult research in NLP achieves greater academic impact, revealing a critical balance for researchers to target.

ChatGPT can generate high-quality academic highlights without the need for extensive labeled datasets, outperforming traditional methods with just a few examples.

Data resources are not just tools; they are the driving force behind the evolution of research methods in Library and Information Science.

Mixed-gender research teams achieve significantly higher citation counts, revealing an optimal gender ratio that could reshape collaborative practices in academia.

As peer reviews progress, positive sentiment rises significantly, challenging the notion that critique intensifies with scrutiny. WHY_IT MATTERS: This insight could transform how we understand the peer review process, potentially leading to improved evaluation strategies in scientific publishing.

Classic algorithms maintain their dominance in influence, but their decline reveals a predictable loss of network centrality that could inform future research trajectories.

Methodological insights are often buried in the middle of research papers, but this study shows how to unearth them for better classification and retrieval.

Authors' promotional language can significantly sway peer reviewers' perceptions of novelty, but only in the murky middle ground of innovation.

Fine-tuning LLMs for automated detection of quotation errors reveals that using source abstracts dramatically boosts accuracy in citation integrity.

LIS scholars get more basic as they age: bibliometric methods dominate the twilight of their careers.

LLMs are subtly reshaping peer review, leading to longer, more superficially polished reports that prioritize clarity over critical assessment of originality and replicability.

Academic paper "highlights" sections are a surprisingly rich source of keywords, boosting unsupervised extraction when combined with abstracts.