• Publications
  • Influence
Fill in the BLANC: Human-free quality estimation of document summaries
TLDR
Evidence is presented that BLANC scores have as good correlation with human evaluations as do the ROUGE family of summary quality measurements, and the method does not require human-written reference summaries, allowing for fully human-free summary quality estimation.
Capture of vapor molecules by a realistic attraction potential of a drop
We derive the rate of the capture of a molecule from the surrounding dilute vapor by a drop with a realistic attraction potential. For a small drop, which is of interest in the vapor phase nucleation
Effect of the attractive potential of a drop in vapor phase nucleation.
  • Oleg V. Vasilyev, Reiss
  • Materials Science, Medicine
    Physical review. E, Statistical physics, plasmas…
  • 1 October 1996
Headline Generation: Learning from Decomposable Document Titles
TLDR
A novel method for generating titles for unstructured text documents is proposed and the results of a randomized double-blind trial in which subjects were unaware of which titles were human or machine-generated are presented.
Deviations from electrodynamics: sun and laser
The existing data on the constraints imposed on the simplest deviation from standard electrodynamics are collected into a picture. The solar constraints are calculated and the possibilities of a
Headline Generation: Learning from Decomposed Document Titles
TLDR
A novel method for generating titles for unstructured text documents is proposed and the results of a randomized double-blind trial in which subjects were unaware of which titles were human or machine-generated are presented.
Sensitivity of BLANC to human-scored qualities of text summaries
TLDR
The case is made for optimal BLANC parameters, at which the BLANC sensitivity to almost all of summary qualities is about as good as the sensitivity of a human annotator.
Is Human Scoring the Best Criteria for Summary Evaluation?
TLDR
The authors' observations for the BLANC family of measures suggest that the criterion of selecting the best measure not relying on correlations with human scores is universal across very different styles of summaries.
Play the Shannon Game With Language Models: A Human-Free Approach to Summary Evaluation
TLDR
New reference-free summary evaluation metrics that use a pretrained language model to estimate the information shared between a document and its summary are introduced, a modern take on the Shannon Game.
...
1
2
3
...