forked from wis-delft/wis-delft.github.io
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathinformation-retrieval.html
156 lines (139 loc) · 8.3 KB
/
information-retrieval.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
---
layout: default_style
title: Information Retrieval
---
<section id="{{page.team}}" class="section-global-wrapper">
<div class="container content-space">
<div class="row justify-content-center blog-post">
<h2> {{page.title}} </h2>
</div>
<div class="row">
<div class="col-md-12">
<div class="row pr-3 pl-3 blog-post">
<p class="text-center"> Our work brings together research expertise in information retrieval (IR),
recommender systems, natural language processing and machine learning. We focus on core IR
topics, from algorithms to applications.
</p>
</div>
</div>
<div class="col-md-12">
<h4> Neural Information Retrieval </h4>
<div class="row pr-3 pl-3 blog-post">
<p>
Our is aim to create more efficient, effective, and interpretable neural models for information
retrieval. In our research on neural information retrieval, we are exploring the utility of
large language models to enhance retrieval and ranking algorithms. We focus mostly on the issues
of efficiency, including runtime and training efficiency, as well as explainability. To overcome
these challenges, we are actively investigating methods to improve the runtime performance of
large models, optimize training processes, and develop techniques for interpreting and
explaining the decisions made by neural retrieval systems.
</p>
</div>
</div>
<div class="col-md-12">
<h4> Explainable IR </h4>
<div class="row pr-3 pl-3 blog-post">
<p>
Predictive models based on complex large language models are widely used in various domains like
search engines, recommender systems, health, legal, and finance. However, they often function as
black boxes, providing predictions or rankings without fully understanding how different factors
influence their outputs. Our research focuses on developing explainable AI techniques for
information retrieval tasks from various angles. We are researching interpretability approaches
to assess ranking models in web search, question-answering, and fact checking.
</p>
</div>
</div>
<div class="col-md-12">
<h4> Complex Question Answering and Fact checking </h4>
<div class="row pr-3 pl-3 blog-post">
<p> Despite significant advancements in question answering brought about by large language models,
they still have limitations when it comes to certain types of questions. These include questions
involving numerical proficiency, compound questions, and complex reasoning, such as fact
verification across diverse sources. Our research is dedicated to comprehending these
limitations of large language models in complex question answering tasks. Moreover, we are
actively developing innovative approaches to address these challenges and make substantial
advancements in real-world question answering applications.
</div>
</div>
<div class="col-md-12">
<h4> Children Information Retrieval </h4>
<div class="row pr-3 pl-3 blog-post">
<p>
IR technology embedded in our lives (Google or Amazon recommendations) responds to average
users. Children have particular cognitive, social, physical, and emotional needs that make the
information they seek, their experiences, sense making, and skills different from those of
adults. children are not simply short users; they are unique users. As such, they use IR tools
differently than adults. With research in this are we aim to empower children so they can
proficiently conduct information discovery tasks. This involved better modeling their needs and
expectations when interacting with IR systems, identifying challenges they face, and building
the algorithmic bridges needed to address them: from new ranking models and novel ways to
generate SERP to how to model.
</p>
</div>
</div>
<div class="col-md-12">
<h4> Information Access for All </h4>
<div class="row pr-3 pl-3 blog-post">
<p> In our quest to access information, we regularly interact with search, recommender, and
question-answering systems. In theory, these systems serve a broad range of users in different
domains attempting to address different tasks. But how do they fare when faced with users,
contexts, and tasks they were not originally designed for? In this area of research, we focus on
identifying limitations inherent in non-traditional, i.e., non-expected, ecosystems and
designing the algorithms needed to address them. For this, we study how users interact with IR
systems using multiple lenses; we also identify how to leverage human traits for modeling users
and how that can impact IR technology design. Along the way, we question how we evaluate novel
solutions in this area, in the absence of benchmarks or given other existing restrictions, e.g.,
federal regulations and fairness objectives.
</p>
</div>
</div>
</div>
<!--The list of people is AUTOMATICALLY computed - DO NOT MODIFY-->
<div class="row">
<h4 class="col-md-12 blog-post">People</h4>
<div class="col-md-12 margin-left-3">
{% include theme-members.html team1='lambda' %}
</div>
</div>
<div class="row">
<h4 class="col-md-12 blog-post">Thesis topics</h4>
<ul class="col-md-12 margin-left-3">
<li>Grounding fact-checking claims</li>
<li>Blind spots in Large Language Models (LLMs)</li>
<li>Personal question-answering using LLMs</li>
<li>Information pollution detection</li>
<li>Explaining personalized ranking</li>
<li>Personalized ranking for non-traditional users</li>
<li>Recommender systems and stereotype propagation</li>
<li>The impact of readability on the web search process</li>
</ul>
</div>
<!--The list of projects is automatically retrieved from _data/lambda.yml -->
<!--Please fill in the yml file with the data about your projects. -->
<div class="row margin-top-1">
<h4 class="col-md-12 blog-post">Projects</h4>
<ul class="col-md-12">
{% for project in site.data.lambda.projects %}
<li class="margin-left-3">
{% assign project_link = project.link | strip %}
{% if project_link == '' %}
<h5>{{project.title}}</h5>
{% else %}
<h5><a href="{{project.link}}" target="_blank"> {{project.title}}</a></h5>
{% endif %}
<p>{{project.description}}</p>
</li>
{% endfor %}
</ul>
</div>
<!--Create link to PURE to retrieve the publications -->
<div class="row" hidden>
<h4 class="col-md-12 blog-post">Publications</h4>
<ul class="col-md-12">
<li class="margin-left-3">
List of publications
</li>
</ul>
</div>
</div>
</section>