Analysis of GraphSum's Attention Weights to Improve the Explainability of Multi-Document Summarization

•Download as PPTX, PDF•

0 likes•68 views

The document analyzes the explainability of GraphSum, an abstractive multi-document summarization model, by examining its attention weights. It finds that GraphSum's attention weights from later decoding layers correlate more strongly with the relevance of input text segments, improving explainability. It also finds that GraphSum performs better when using paragraphs rather than sentences as input for the news domain, as paragraphs aid structure rather than topic separation for news articles. The document concludes that attention weights and expert annotations may provide better insight into abstractive summarization than ROUGE scores alone.

Science

Analysis of GraphSum’s Attention
Weights to Improve the
Explainability of Multi-Document
Summarization
06.04.2022
M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp 1
M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner,
J. Töllich and A. Scherp

Extractive vs. Abstractive MDS
06.04.2022
M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp 2
Input Documents
Model
Summary
Model
Extractive:
Abstractive:

Abstractive Graph-based MDS
06.04.2022
M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp 3
Documents
Model
Summary
Sentences
Explainability ?

Research Questions
06.04.2022
M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp 4
Model1
Sentences Paragraphs
Model2
Quality?
Documents
Model
Summary
Explainability?

GraphSum
06.04.2022
M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp 5
Source: Li et al. “Leveraging Graph to Improve Abstractive Multi-Document Summarization” (2020)

Textual Unit Comparison
M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp

Build TF-IDF
Graph
Train
GraphSum
Model
Evaluate
Performance
Approach for Textual Units Comparison
M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp 7
06.04.2022

Pre-Processing
06.04.2022
M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp 8
EXTRACTION
TRUNCATION
/
PADDING
TF-IDF
GRAPH
Build TF-IDF
Graph
Train
GraphSum
Model
Evaluate
Performance

GraphSum Training Procedure
06.04.2022
M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp 9
Build TF-IDF
Graph
Train
GraphSum
Model
Evaluate
Performance
 Architecture and hyper-parameters as suggested by
Li et. al “Leveraging Graph to Improve Abstractive Multi-Document Summarization” (2020)
 Use similarity graph generated by pre-processing
 Use multiple batch-sizes
 Same number of input tokens
 Train / validation / test split

ROUGE Score
 ROUGE-2: Overlapping bi-grams
06.04.2022
M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp 10
 ROUGE-L: Longest common subsequence
 Final score based on F-score as proposed by
Chin-Yew Lin, “ROUGE: A Package for Automatic Evaluation of Summaries” (2004)
Reference Reference
Candidate Candidate
Build TF-IDF
Graph
Train
GraphSum
Model
Evaluate
Performance

Explainability Analysis
M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp

Approach for Explainability Improvement
06.04.2022
M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp 12

Data Sets
MultiNews WikiSum
Sentence vs Paragraphs x
Explainability Analysis x x
06.04.2022
M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp 13
MultiNews:
Human written news summaries from professionals (60.000 Documents)
WikiSum:
Wikipedia articles and their references as MDS task (2.3 Million Arcticles)

Results: Textual Unit Comparison
M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp

Sentences vs Paragraphs
06.04.2022
M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp 15
MultiNews

Usage of Paragraphs in News Domain
06.04.2022
M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp 16
MultiNews

Results: Explainability Analysis
M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp

Attention Weights Correlation
06.04.2022
M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp 18
Decoding Layer Multi-Heads

Correlation between Attention Weights and Reference Metric
06.04.2022
M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp 19
MultiNews
Layer 6 (High Correlation)
Reference Metric
Attention
Weights
Reference Metric
Attention
Weights
Layer 3 (Low Correlation)

Positional Bias (MultiNews)
06.04.2022
M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp 20

Conclusion
 Paragraphs perform better than sentences for news domain
 Paragraphs are used as structural aid, not for topic separation
 Other domains may show different behaviour
 Attention weights improve explainability of MDS
 Attention weights provide source origin information
 Latter decoding layers more suitable
 ROUGE score might not be fully applicable as metric for abstractive MDS
 ROUGE score not suitable for e.g., paraphrased sentences
 Expert annotated source information could provide better insights
06.04.2022
M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp 21
Code available on GitHub: https://github.com/arnelochner/GBTBMDS

Text extraction from scientific figures has been addressed in the past by different unsupervised approaches due to the limited amount of training data. Motivated by the recent advances in Deep Learning, we propose a two-step neural-network-based pipeline to localize and extract text using Fully Convolutional Networks. We improve the localization of the text bounding boxes by applying a novel combination of a Residual Network with the Region Proposal Network based on Faster R-CNN. The predicted bounding boxes are further pre-processed and used as input to the of-the-shelf optical character recognition engine Tesseract 4.0. We evaluate our improved text localization method on five different datasets of scientific figures and compare it with the best unsupervised pipeline. Since only limited training data is available, we further experiment with different data augmentation techniques for increasing the size of the training datasets and demonstrate their positive impact. We use Average Precision and F1 measure to assess the text localization results. In addition, we apply Gestalt Pattern Matching and Levenshtein Distance for evaluating the quality of the recognized text. Our extensive experiments show that our new pipeline based on neural networks outperforms the best unsupervised approach by a large margin of 19-20%.

A Comparison of Approaches for Automated Text Extraction from Scholarly Figures

Ansgar Scherp

So far, there has not been a comparative evaluation of different approaches for text extraction from scholarly figures. In order to fill this gap, we have defined a generic pipeline for text extraction that abstracts from the existing approaches as documented in the literature. In this paper, we use this generic pipeline to systematically evaluate and compare 32 configurations for text extraction over four datasets of scholarly figures of different origin and characteristics. In total, our experiments have been run over more than 400 manually labeled figures. The experimental results show that the approach BS-4OS results in the best F-measure of 0.67 for the Text Location Detection and the best average Levenshtein Distance of 4.71 between the recognized text and the gold standard on all four datasets using the Ocropy OCR engine.

About Multimedia Presentation Generation and Multimedia Metadata: From Synthe...

Ansgar Scherp

Mining and Managing Large-scale Linked Open Data

Ansgar Scherp

Linked Open Data (LOD) is about publishing and interlinking data of different origin and purpose on the web. The Resource Description Framework (RDF) is used to describe data on the LOD cloud. In contrast to relational databases, RDF does not provide a fixed, pre-defined schema. Rather, RDF allows for flexibly modeling the data schema by attaching RDF types and properties to the entities. Our schema-level index called SchemEX allows for searching in large-scale RDF graph data. The index can be efficiently computed with reasonable accuracy over large-scale data sets with billions of RDF triples, the smallest information unit on the LOD cloud. SchemEX is highly needed as the size of the LOD cloud quickly increases. Due to the evolution of the LOD cloud, one observes frequent changes of the data. We show that also the data schema changes in terms of combinations of RDF types and properties. As changes cannot capture the dynamics of the LOD cloud, current work includes temporal clustering and finding periodicities in entity dynamics over large-scale snapshots of the LOD cloud with about 100 million triples per week for more than three years.

Knowledge Discovery in Social Media and Scientific Digital Libraries

Ansgar Scherp

The talk presents selected results of our research in the area of text and data mining in social media and scientific literature. (1) First, we consider the area of classifying microblogging postings like tweets on Twitter. Typically, the classification results are evaluated against a gold standard, which is either the hashtags of the tweets’ authors or manual annotations. We claim that there are fundamental differences between these two kinds of gold standard classifications and conducted an experiment with 163 participants to manually classify tweets from ten topics. Our results show that the human annotators are more likely to classify tweets like other human annotators than like the tweets’ authors (i. e., the hashtags). This may influence the evaluation of classification methods like LDA and we argue that researchers should reflect the kind of gold standard used when interpreting their results. (2) Second, we present a framework for semantic document annotation that aims to compare different existing as well as new annotation strategies. For entity detection, we compare semantic taxonomies, trigrams, RAKE, and LDA. For concept activation, we cover a set of statistical, hierarchy-based, and graph-based methods. The strategies are evaluated over 100,000 manually labeled scientific documents from economics, politics, and computer science. (3) Finally, we present a processing pipeline for extracting text of varying size, rotation, color, and emphases from scholarly figures. The pipeline does not need training nor does it make any assumptions about the characteristics of the scholarly figures. We conducted a preliminary evaluation with 121 figures from a broad range of illustration types. URL: https://www.ukp.tu-darmstadt.de/ukp-home/news-singleview/artikel/guest-speaker-ansgar-scherp/

A Comparison of Different Strategies for Automated Semantic Document Annotation

Ansgar Scherp

We introduce a framework for automated semantic document annotation that is composed of four processes, namely concept extraction, concept activation, annotation selection, and evaluation. The framework is used to implement and compare different annotation strategies motivated by the literature. For concept extraction, we apply entity detection with semantic hierarchical knowledge bases, Tri-gram, RAKE, and LDA. For concept activation, we compare a set of statistical, hierarchy-based, and graph-based methods. For selecting annotations, we compare top-k as well as kNN. In total, we define 43 different strategies including novel combinations like using graph-based activation with kNN. We have evaluated the strategies using three different datasets of varying size from three scientific disciplines (economics, politics, and computer science) that contain 100, 000 manually labeled documents in total. We obtain the best results on all three datasets by our novel combination of entity detection with graph-based activation (e.g., HITS and Degree) and kNN. For the economic and political science datasets, the best F-measure is .39 and .28, respectively. For the computer science dataset, the maximum F-measure of .33 can be reached. The experiments are the by far largest on scholarly content annotation, which typically are up to a few hundred documents per dataset only. Gregor Große-Bölting, Chifumi Nishioka, and Ansgar Scherp. 2015. A Comparison of Different Strategies for Automated Semantic Document Annotation. In Proceedings of the 8th International Conference on Knowledge Capture (K-CAP 2015). ACM, New York, NY, USA, , Article 8 , 8 pages. DOI=http://dx.doi.org/10.1145/2815833.2815838

We propose a pipeline for text extraction from infographics that makes use of a novel combination of data mining and computer vision techniques. The pipeline defines a sequence of steps to identify characters, cluster them into text lines, determine their rotation angle, and apply state-of-the-art OCR to recognize the text. In this paper, we formally define the pipeline and present its current implementation. In addition, we have conducted preliminary evaluations over a data corpus of 121 manually annotated infographics from a broad range of illustration types such as bar charts, pie charts, and line charts, maps, and others. We assess the results of our text extraction pipeline by comparing it with two baselines. Finally, we sketch an outline for future work and possibilities for improving the pipeline. - http://ceur-ws.org/Vol-1458/

A Framework for Iterative Signing of Graph Data on the Web

Ansgar Scherp

Existing algorithms for signing graph data typically do not cover the whole signing process. In addition, they lack distinctive features such as signing graph data at different levels of granularity, iterative signing of graph data, and signing multiple graphs. In this paper, we introduce a novel framework for signing arbitrary graph data provided, e g., as RDF(S), Named Graphs, or OWL. We conduct an extensive theoretical and empirical analysis of the runtime and space complexity of different framework configurations. The experiments are performed on synthetic and real-world graph data of different size and different number of blank nodes. We investigate security issues, present a trust model, and discuss practical considerations for using our signing framework. We released a Java-based open source implementation of our software framework for iterative signing of arbitrary graph data provided, e. g., as RDF(S), Named Graphs, or OWL. The software framework is based on a formalization of different graph signing functions and supports different configurations. It is available in source code as well as pre-compiled as .jar-file. The graph signing framework exhibits the following unique features: - Signing graphs on different levels of granularity - Signing multiple graphs at once - Iterative signing of graph data for provenance tracking - Independence of the used language for encoding the graph (i. e., the signature does not break when changing the graph representation) The documentation of the software framework and its source code is available from: http://icp.it-risk.iwvi.uni-koblenz.de/wiki/Software_Framework_for_Signing_Graph_Data

Smart photo selection: interpret gaze as personal interest

Ansgar Scherp

Manually selecting subsets of photos from large collections in order to present them to friends or colleagues or to print them as photo books can be a tedious task. Today, fully automatic approaches are at hand for supporting users. They make use of pixel information extracted from the images, analyze contextual information such as capture time and focal aperture, or use both to determine a proper subset of photos. However, these approaches miss the most important factor in the photo selection process: the user. The goal of our approach is to consider individual interests. By recording and analyzing gaze information from the user's viewing photo collections, we obtain information on user's interests and use this information in the creation of personal photo selections. In a controlled experiment with 33 participants, we show that the selections can be significantly improved over a baseline approach by up to 22% when taking individual viewing behavior into account. We also obtained significantly better results for photos taken at an event participants were involved in compared with photos from another event.

Events in Multimedia - Theory, Model, Application

Ansgar Scherp

Can you see it? Annotating Image Regions based on Users' Gaze Information

Ansgar Scherp

Linked open data - how to juggle with more than a billion triples

Ansgar Scherp

SchemEX -- Building an Index for Linked Open Data

Ansgar Scherp

SchemEX -- Building an Index for Linked Open Data

Ansgar Scherp

A Model of Events for Integrating Event-based Information in Complex Socio-te...

Ansgar Scherp

SchemEX - Creating the Yellow Pages for the Linked Open Data Cloud

Ansgar Scherp

strukt - A Pattern System for Integrating Individual and Organizational Knowl...

Ansgar Scherp

Identifying Objects in Images from Analyzing the User‘s Gaze Movements for Pr...

Ansgar Scherp

Linked Open Data (Entwurfsprinzipien und Muster für vernetzte Daten)

Ansgar Scherp

Hemostasis_importance& clinical significance.pptx

muralinath2

GBSN- Microbiology (Lab 3) Gram Staining

Areesha Ahmad

Leaf Initiation, Growth and Differentiation.pdf

RenuJangid3

Nucleic Acid-its structural and functional complexity.

Nistarini College, Purulia (W.B) India

What is greenhouse gasses and how many gasses are there to affect the Earth.

moosaasad1975

extra-chromosomal-inheritance[1].pptx.pdfpdf

DiyaBiswas10

Slide 1: Title Slide Extrachromosomal Inheritance Slide 2: Introduction to Extrachromosomal Inheritance Definition: Extrachromosomal inheritance refers to the transmission of genetic material that is not found within the nucleus. Key Components: Involves genes located in mitochondria, chloroplasts, and plasmids. Slide 3: Mitochondrial Inheritance Mitochondria: Organelles responsible for energy production. Mitochondrial DNA (mtDNA): Circular DNA molecule found in mitochondria. Inheritance Pattern: Maternally inherited, meaning it is passed from mothers to all their offspring. Diseases: Examples include Leber’s hereditary optic neuropathy (LHON) and mitochondrial myopathy. Slide 4: Chloroplast Inheritance Chloroplasts: Organelles responsible for photosynthesis in plants. Chloroplast DNA (cpDNA): Circular DNA molecule found in chloroplasts. Inheritance Pattern: Often maternally inherited in most plants, but can vary in some species. Examples: Variegation in plants, where leaf color patterns are determined by chloroplast DNA. Slide 5: Plasmid Inheritance Plasmids: Small, circular DNA molecules found in bacteria and some eukaryotes. Features: Can carry antibiotic resistance genes and can be transferred between cells through processes like conjugation. Significance: Important in biotechnology for gene cloning and genetic engineering. Slide 6: Mechanisms of Extrachromosomal Inheritance Non-Mendelian Patterns: Do not follow Mendel’s laws of inheritance. Cytoplasmic Segregation: During cell division, organelles like mitochondria and chloroplasts are randomly distributed to daughter cells. Heteroplasmy: Presence of more than one type of organellar genome within a cell, leading to variation in expression. Slide 7: Examples of Extrachromosomal Inheritance Four O’clock Plant (Mirabilis jalapa): Shows variegated leaves due to different cpDNA in leaf cells. Petite Mutants in Yeast: Result from mutations in mitochondrial DNA affecting respiration. Slide 8: Importance of Extrachromosomal Inheritance Evolution: Provides insight into the evolution of eukaryotic cells. Medicine: Understanding mitochondrial inheritance helps in diagnosing and treating mitochondrial diseases. Agriculture: Chloroplast inheritance can be used in plant breeding and genetic modification. Slide 9: Recent Research and Advances Gene Editing: Techniques like CRISPR-Cas9 are being used to edit mitochondrial and chloroplast DNA. Therapies: Development of mitochondrial replacement therapy (MRT) for preventing mitochondrial diseases. Slide 10: Conclusion Summary: Extrachromosomal inheritance involves the transmission of genetic material outside the nucleus and plays a crucial role in genetics, medicine, and biotechnology. Future Directions: Continued research and technological advancements hold promise for new treatments and applications. Slide 11: Questions and Discussion Invite Audience: Open the floor for any questions or further discussion on the topic.

Body fluids_tonicity_dehydration_hypovolemia_hypervolemia.pptx

muralinath2

Orion Air Quality Monitoring Systems - CWS

Columbia Weather Systems

nodule formation by alisha dewangan.pptx

alishadewangan1

Recently uploaded

Hemostasis_importance& clinical significance.pptx

muralinath2

GBSN- Microbiology (Lab 3) Gram Staining

Areesha Ahmad

Leaf Initiation, Growth and Differentiation.pdf

RenuJangid3

Nucleic Acid-its structural and functional complexity.

Nistarini College, Purulia (W.B) India

What is greenhouse gasses and how many gasses are there to affect the Earth.

moosaasad1975

extra-chromosomal-inheritance[1].pptx.pdfpdf

DiyaBiswas10

Body fluids_tonicity_dehydration_hypovolemia_hypervolemia.pptx

muralinath2

Orion Air Quality Monitoring Systems - CWS

Columbia Weather Systems

nodule formation by alisha dewangan.pptx

alishadewangan1

Lateral Ventricles.pdf very easy good diagrams comprehensive

silvermistyshot

in vitro propagation of plants lecture note.pptx

yusufzako14

THE IMPORTANCE OF MARTIAN ATMOSPHERE SAMPLE RETURN.

Sérgio Sacani

The return of a sample of near-surface atmosphere from Mars would facilitate answers to several first-order science questions surrounding the formation and evolution of the planet. One of the important aspects of terrestrial planet formation in general is the role that primary atmospheres played in influencing the chemistry and structure of the planets and their antecedents. Studies of the martian atmosphere can be used to investigate the role of a primary atmosphere in its history. Atmosphere samples would also inform our understanding of the near-surface chemistry of the planet, and ultimately the prospects for life. High-precision isotopic analyses of constituent gases are needed to address these questions, requiring that the analyses are made on returned samples rather than in situ.

Richard's aventures in two entangled wonderlands

Richard Gill

Since the loophole-free Bell experiments of 2020 and the Nobel prizes in physics of 2022, critics of Bell's work have retreated to the fortress of super-determinism. Now, super-determinism is a derogatory word - it just means "determinism". Palmer, Hance and Hossenfelder argue that quantum mechanics and determinism are not incompatible, using a sophisticated mathematical construction based on a subtle thinning of allowed states and measurements in quantum mechanics, such that what is left appears to make Bell's argument fail, without altering the empirical predictions of quantum mechanics. I think however that it is a smoke screen, and the slogan "lost in math" comes to my mind. I will discuss some other recent disproofs of Bell's theorem using the language of causality based on causal graphs. Causal thinking is also central to law and justice. I will mention surprising connections to my work on serial killer nurse cases, in particular the Dutch case of Lucia de Berk and the current UK case of Lucy Letby.

bordetella pertussis.................................ppt

kejapriya1

(May 29th, 2024) Advancements in Intravital Microscopy- Insights for Preclini...

Scintica Instrumentation

Intravital microscopy (IVM) is a powerful tool utilized to study cellular behavior over time and space in vivo. Much of our understanding of cell biology has been accomplished using various in vitro and ex vivo methods; however, these studies do not necessarily reflect the natural dynamics of biological processes. Unlike traditional cell culture or fixed tissue imaging, IVM allows for the ultra-fast high-resolution imaging of cellular processes over time and space and were studied in its natural environment. Real-time visualization of biological processes in the context of an intact organism helps maintain physiological relevance and provide insights into the progression of disease, response to treatments or developmental processes. In this webinar we give an overview of advanced applications of the IVM system in preclinical research. IVIM technology is a provider of all-in-one intravital microscopy systems and solutions optimized for in vivo imaging of live animal models at sub-micron resolution. The system’s unique features and user-friendly software enables researchers to probe fast dynamic biological processes such as immune cell tracking, cell-cell interaction as well as vascularization and tumor metastasis with exceptional detail. This webinar will also give an overview of IVM being utilized in drug development, offering a view into the intricate interaction between drugs/nanoparticles and tissues in vivo and allows for the evaluation of therapeutic intervention in a variety of tissues and organs. This interdisciplinary collaboration continues to drive the advancements of novel therapeutic strategies.

Observation of Io’s Resurfacing via Plume Deposition Using Ground-based Adapt...

Sérgio Sacani

Since volcanic activity was first discovered on Io from Voyager images in 1979, changes on Io’s surface have been monitored from both spacecraft and ground-based telescopes. Here, we present the highest spatial resolution images of Io ever obtained from a groundbased telescope. These images, acquired by the SHARK-VIS instrument on the Large Binocular Telescope, show evidence of a major resurfacing event on Io’s trailing hemisphere. When compared to the most recent spacecraft images, the SHARK-VIS images show that a plume deposit from a powerful eruption at Pillan Patera has covered part of the long-lived Pele plume deposit. Although this type of resurfacing event may be common on Io, few have been detected due to the rarity of spacecraft visits and the previously low spatial resolution available from Earth-based telescopes. The SHARK-VIS instrument ushers in a new era of high resolution imaging of Io’s surface using adaptive optics at visible wavelengths.

Seminar of U.V. Spectroscopy by SAMIR PANDA

SAMIR PANDA

erythropoiesis-I_mechanism& clinical significance.pptx

muralinath2

DERIVATION OF MODIFIED BERNOULLI EQUATION WITH VISCOUS EFFECTS AND TERMINAL V...

Wasswaderrick3

In this book, we use conservation of energy techniques on a fluid element to derive the Modified Bernoulli equation of flow with viscous or friction effects. We derive the general equation of flow/ velocity and then from this we derive the Pouiselle flow equation, the transition flow equation and the turbulent flow equation. In the situations where there are no viscous effects , the equation reduces to the Bernoulli equation. From experimental results, we are able to include other terms in the Bernoulli equation. We also look at cases where pressure gradients exist. We use the Modified Bernoulli equation to derive equations of flow rate for pipes of different cross sectional areas connected together. We also extend our techniques of energy conservation to a sphere falling in a viscous medium under the effect of gravity. We demonstrate Stokes equation of terminal velocity and turbulent flow equation. We look at a way of calculating the time taken for a body to fall in a viscous medium. We also look at the general equation of terminal velocity.

NuGOweek 2024 Ghent - programme - final version

pablovgd

Recently uploaded (20)

Hemostasis_importance& clinical significance.pptx

GBSN- Microbiology (Lab 3) Gram Staining

Leaf Initiation, Growth and Differentiation.pdf

Nucleic Acid-its structural and functional complexity.

What is greenhouse gasses and how many gasses are there to affect the Earth.

extra-chromosomal-inheritance[1].pptx.pdfpdf

Body fluids_tonicity_dehydration_hypovolemia_hypervolemia.pptx

Orion Air Quality Monitoring Systems - CWS

nodule formation by alisha dewangan.pptx

Lateral Ventricles.pdf very easy good diagrams comprehensive

in vitro propagation of plants lecture note.pptx

THE IMPORTANCE OF MARTIAN ATMOSPHERE SAMPLE RETURN.

Richard's aventures in two entangled wonderlands

bordetella pertussis.................................ppt

(May 29th, 2024) Advancements in Intravital Microscopy- Insights for Preclini...

Observation of Io’s Resurfacing via Plume Deposition Using Ground-based Adapt...

Seminar of U.V. Spectroscopy by SAMIR PANDA

erythropoiesis-I_mechanism& clinical significance.pptx

DERIVATION OF MODIFIED BERNOULLI EQUATION WITH VISCOUS EFFECTS AND TERMINAL V...

NuGOweek 2024 Ghent - programme - final version

Analysis of GraphSum's Attention Weights to Improve the Explainability of Multi-Document Summarization

1. Analysis of GraphSum’s Attention Weights to Improve the Explainability of Multi-Document Summarization 06.04.2022 M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp 1 M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp

2. Extractive vs. Abstractive MDS 06.04.2022 M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp 2 Input Documents Model Summary Model Extractive: Abstractive:

3. Abstractive Graph-based MDS 06.04.2022 M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp 3 Documents Model Summary Sentences Explainability ?

4. Research Questions 06.04.2022 M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp 4 Model1 Sentences Paragraphs Model2 Quality? Documents Model Summary Explainability?

5. GraphSum 06.04.2022 M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp 5 Source: Li et al. “Leveraging Graph to Improve Abstractive Multi-Document Summarization” (2020)

6. Textual Unit Comparison M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp

7. Build TF-IDF Graph Train GraphSum Model Evaluate Performance Approach for Textual Units Comparison M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp 7 06.04.2022

8. Pre-Processing 06.04.2022 M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp 8 EXTRACTION TRUNCATION / PADDING TF-IDF GRAPH Build TF-IDF Graph Train GraphSum Model Evaluate Performance

9. GraphSum Training Procedure 06.04.2022 M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp 9 Build TF-IDF Graph Train GraphSum Model Evaluate Performance  Architecture and hyper-parameters as suggested by Li et. al “Leveraging Graph to Improve Abstractive Multi-Document Summarization” (2020)  Use similarity graph generated by pre-processing  Use multiple batch-sizes  Same number of input tokens  Train / validation / test split

10. ROUGE Score  ROUGE-2: Overlapping bi-grams 06.04.2022 M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp 10  ROUGE-L: Longest common subsequence  Final score based on F-score as proposed by Chin-Yew Lin, “ROUGE: A Package for Automatic Evaluation of Summaries” (2004) Reference Reference Candidate Candidate Build TF-IDF Graph Train GraphSum Model Evaluate Performance

11. Explainability Analysis M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp

12. Approach for Explainability Improvement 06.04.2022 M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp 12

13. Data Sets MultiNews WikiSum Sentence vs Paragraphs x Explainability Analysis x x 06.04.2022 M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp 13 MultiNews: Human written news summaries from professionals (60.000 Documents) WikiSum: Wikipedia articles and their references as MDS task (2.3 Million Arcticles)

14. Results: Textual Unit Comparison M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp

15. Sentences vs Paragraphs 06.04.2022 M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp 15 MultiNews

16. Usage of Paragraphs in News Domain 06.04.2022 M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp 16 MultiNews

17. Results: Explainability Analysis M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp

18. Attention Weights Correlation 06.04.2022 M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp 18 Decoding Layer Multi-Heads

19. Correlation between Attention Weights and Reference Metric 06.04.2022 M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp 19 MultiNews Layer 6 (High Correlation) Reference Metric Attention Weights Reference Metric Attention Weights Layer 3 (Low Correlation)

20. Positional Bias (MultiNews) 06.04.2022 M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp 20

21. Conclusion  Paragraphs perform better than sentences for news domain  Paragraphs are used as structural aid, not for topic separation  Other domains may show different behaviour  Attention weights improve explainability of MDS  Attention weights provide source origin information  Latter decoding layers more suitable  ROUGE score might not be fully applicable as metric for abstractive MDS  ROUGE score not suitable for e.g., paraphrased sentences  Expert annotated source information could provide better insights 06.04.2022 M.L. Hickmann, F. Wurzberger, M. Hoxhalli, A. Lochner, J. Töllich and A. Scherp 21 Code available on GitHub: https://github.com/arnelochner/GBTBMDS

Editor's Notes

Paragraphs: - Leveraging inter-paragraph relations can provide the model additional information for detecting contextual relations between topics. Sentences: - Our rationale is that with sentences as textual units, the graph structure represents inter-sentence relations, which may provide more detailed information within topics and thus may improve the results.
Batch Sizes GraphSum Model Hyperparamter as proposed by Li et al
Use tokenzier for extraction Same number of tokens
Wir haben ROUGE Scores als referenz verwendet Pearson Correlation
WikiSum nicht für Snetence vs Paragraphs aus resource limitations
Averaged Runs
Multi News Example
Basierend auf diesen Erkentnissen haben wir die Attention weights der Multi-heads aggregiert im weiteren Vorgehen
ROUGE Score ist Reference metric

Analysis of GraphSum's Attention Weights to Improve the Explainability of Multi-Document Summarization

Recommended

Recommended

More Related Content

More from Ansgar Scherp

More from Ansgar Scherp (13)

Recently uploaded

Recently uploaded (20)

Analysis of GraphSum's Attention Weights to Improve the Explainability of Multi-Document Summarization

Editor's Notes