Related papers: Tailor Based Allocations for Multiple Authorship: …
We address the question how citation-based bibliometric indicators can best be normalized to ensure fair comparisons between publications from different scientific fields and different years. In a systematic large-scale empirical analysis,…
The predictive power of the h-index has been shown to depend for a long time on citations to rather old publications. This has raised doubts about its usefulness for predicting future scientific achievements. Here I investigate a variant…
The self-attention mechanism, while foundational to modern Transformer architectures, suffers from a critical inefficiency: it frequently allocates substantial attention to redundant or noisy context. Differential Attention addressed this…
Understanding the mechanisms driving the distribution of scientific citations is a key challenge in assessing the scientific impact of authors. We investigate the influence of the preferential attachment rule (PAR) in this process by…
The $K$-index is an easily computable centrality index in complex networks, such as a scientific citations network. A researcher has a $K$-index equal to $K$ if he or she is cited by $K$ articles that have at least $K$ citations. The…
I introduce a decomposition of the h-index, which is nowadays the leading criterion to assess the relevance of a scientist in his/her research field. According to the proposed decomposition, the h-index is the product of two indicators, the…
How to quantify the impact of a researcher's or an institution's body of work is a matter of increasing importance to scientists, funding agencies, and hiring committees. The use of bibliometric indicators, such as the h-index or the…
The pursuit of reducing the memory footprint of the self-attention mechanism in multi-headed self attention (MHA) spawned a rich portfolio of methods, e.g., group-query attention (GQA) and multi-head latent attention (MLA). The methods…
The recently proposed Euclidean index offers a novel approach to measure the citation impact of academic authors, in particular as an alternative to the h-index. We test if the index provides new, robust information, not covered by existing…
The h-index (Hirsch, 2005) is robust, remaining relatively unaffected by errors in the long tails of the citations-rank distribution, such as typographic errors that short-change frequently-cited papers and create bogus additional records.…
Training Data Attribution (TDA) seeks to trace model predictions back to influential training examples, enhancing interpretability and safety. We formulate TDA as a Bayesian information-theoretic problem: subsets are scored by the…
In the literature and on the Web we can readily find research excellence rankings for organizations and countries by either total number of highly-cited articles (HCAs) or by ratio of HCAs to total publications. Neither are indicators of…
Journal Impact Factors (IFs) can be considered historically as the first attempt to normalize citation distributions by using averages over two years. However, it has been recognized that citation distributions vary among fields of science…
In the case of the scientometric evaluation of multi- or interdisciplinary units one risks to compare apples with oranges: each paper has to assessed in comparison to an appropriate reference set. We suggest that the set of citing papers…
Evaluation of researchers' output is vital for hiring committees and funding bodies, and it is usually measured via their scientific productivity, citations, or a combined metric such as h-index. Assessing young researchers is more critical…
How much is the h-index of an editor of a well ranked journal improved due to citations which occur after his or her appointment? Scientific recognition within academia is widely measured nowadays by the number of citations or h-index. Our…
Are existing ways of measuring scientific quality reflecting disadvantages of not being part of giant collaborations? How could possible discrimination be avoided? We propose indices defined for each discipline (subfield) and which count…
With such increasing popularity and availability of digital text data, authorships of digital texts can not be taken for granted due to the ease of copying and parsing. This paper presents a new text style analysis called natural frequency…
While computer modeling and simulation are crucial for understanding scientometrics, their practical use in literature remains somewhat limited. In this study, we establish a joint coauthorship and citation network using preferential…
This paper analyzes publication efficiency in terms of Hirsch-index or h-index and total citations, with an analogy to the Carnot efficiency used in thermodynamics. Such publication efficiency, with typical value of 30%, can be utilized to…