Added author page patrick-haller (HU Berlin) next to patrick-haller-zurich

data/xml/2023.emnlp.xml

@@ -14758,7 +14758,7 @@ The experiments were repeated and the tables and figures were updated. Changes a
     <paper id="1">
       <title>Fabricator: An Open Source Toolkit for Generating Labeled Training Data with Teacher <fixed-case>LLM</fixed-case>s</title>
       <author><first>Jonas</first><last>Golde</last><affiliation>Humboldt-University of Berlin</affiliation></author>
-      <author><first>Patrick</first><last>Haller</last><affiliation>Machine Learning Group - Humboldt University of Berlin</affiliation></author>
+      <author id="patrick-haller"><first>Patrick</first><last>Haller</last><affiliation>Machine Learning Group - Humboldt University of Berlin</affiliation></author>
       <author><first>Felix</first><last>Hamborg</last><affiliation>University of Konstanz</affiliation></author>
       <author><first>Julian</first><last>Risch</last><affiliation>deepset</affiliation></author>
       <author><first>Alan</first><last>Akbik</last><affiliation>Humboldt University of Berlin</affiliation></author>

data/xml/2024.blackboxnlp.xml

@@ -183,7 +183,7 @@
     <paper id="14">
       <title>On the alignment of <fixed-case>LM</fixed-case> language generation and human language comprehension</title>
       <author><first>Lena Sophia</first><last>Bolliger</last><affiliation>University of Zurich</affiliation></author>
-      <author orcid="0000-0002-8968-7587"><first>Patrick</first><last>Haller</last><affiliation>University of Zurich</affiliation></author>
+      <author orcid="0000-0002-8968-7587" id="patrick-haller-zurich"><first>Patrick</first><last>Haller</last><affiliation>University of Zurich</affiliation></author>
       <author orcid="0000-0001-9018-9713"><first>Lena Ann</first><last>Jäger</last><affiliation>University of Zurich and Universität Potsdam</affiliation></author>
       <pages>217-231</pages>
       <abstract>Previous research on the predictive power (PP) of surprisal and entropy has focused on determining which language models (LMs) generate estimates with the highest PP on reading times, and examining for which populations the PP is strongest. In this study, we leverage eye movement data on texts that were generated using a range of decoding strategies with different LMs. We then extract the transition scores that reflect the models’ production rather than comprehension effort. This allows us to investigate the alignment of LM language production and human language comprehension. Our findings reveal that there are differences in the strength of the alignment between reading behavior and certain LM decoding strategies and that this alignment further reflects different stages of language understanding (early, late, or global processes). Although we find lower PP of transition-based measures compared to surprisal and entropy for most decoding strategies, our results provide valuable insights into which decoding strategies impose less processing effort for readers. Our code is available via https://github.com/DiLi-Lab/LM-human-alignment.</abstract>

data/xml/2024.conll.xml

@@ -604,7 +604,7 @@
     </paper>
     <paper id="7">
       <title><fixed-case>B</fixed-case>aby<fixed-case>HGRN</fixed-case>: Exploring <fixed-case>RNN</fixed-case>s for Sample-Efficient Language Modeling</title>
-      <author><first>Patrick</first><last>Haller</last><affiliation>Humboldt Universität Berlin</affiliation></author>
+      <author id="patrick-haller"><first>Patrick</first><last>Haller</last><affiliation>Humboldt Universität Berlin</affiliation></author>
       <author><first>Jonas</first><last>Golde</last><affiliation>Department of Computer Science, Humboldt University Berlin, Humboldt Universität Berlin</affiliation></author>
       <author><first>Alan</first><last>Akbik</last><affiliation>Humboldt Universität Berlin</affiliation></author>
       <pages>82-94</pages>

data/xml/2024.lrec.xml

@@ -13104,7 +13104,7 @@
     </paper>
     <paper id="1111">
       <title><fixed-case>PECC</fixed-case>: Problem Extraction and Coding Challenges</title>
-      <author><first>Patrick</first><last>Haller</last></author>
+      <author id="patrick-haller"><first>Patrick</first><last>Haller</last></author>
       <author><first>Jonas</first><last>Golde</last></author>
       <author><first>Alan</first><last>Akbik</last></author>
       <pages>12690–12699</pages>

data/xml/2024.naacl.xml

@@ -8082,7 +8082,7 @@
     </paper>
     <paper id="8">
       <title><fixed-case>O</fixed-case>pinion<fixed-case>GPT</fixed-case>: Modelling Explicit Biases in Instruction-Tuned <fixed-case>LLM</fixed-case>s</title>
-      <author><first>Patrick</first><last>Haller</last><affiliation>Humboldt Universität Berlin</affiliation></author>
+      <author id="patrick-haller"><first>Patrick</first><last>Haller</last><affiliation>Humboldt Universität Berlin</affiliation></author>
       <author><first>Ansar</first><last>Aynetdinov</last><affiliation>Department of Computer Science, Humboldt University Berlin, Humboldt Universität Berlin</affiliation></author>
       <author><first>Alan</first><last>Akbik</last><affiliation>Humboldt Universität Berlin</affiliation></author>
       <pages>78-86</pages>

data/xml/2025.acl.xml

@@ -17605,7 +17605,7 @@
     </paper>
     <paper id="1205">
       <title>Leveraging In-Context Learning for Political Bias Testing of <fixed-case>LLM</fixed-case>s</title>
-      <author orcid="0000-0002-8968-7587"><first>Patrick</first><last>Haller</last><affiliation>University of Zurich</affiliation></author>
+      <author id="patrick-haller-zurich" orcid="0000-0002-8968-7587"><first>Patrick</first><last>Haller</last><affiliation>University of Zurich</affiliation></author>
       <author orcid="0009-0002-1821-1837"><first>Jannis</first><last>Vamvas</last><affiliation>University of Zurich</affiliation></author>
       <author orcid="0000-0002-1438-4741"><first>Rico</first><last>Sennrich</last><affiliation>University of Zurich</affiliation></author>
       <author orcid="0000-0001-9018-9713"><first>Lena Ann</first><last>Jäger</last><affiliation>University of Zurich</affiliation></author>

data/xml/2025.babylm.xml

@@ -181,7 +181,7 @@
     </paper>
     <paper id="14">
       <title>Sample-Efficient Language Modeling with Linear Attention and Lightweight Enhancements</title>
-      <author><first>Patrick</first><last>Haller</last><affiliation>Humboldt Universität Berlin</affiliation></author>
+      <author id="patrick-haller"><first>Patrick</first><last>Haller</last><affiliation>Humboldt Universität Berlin</affiliation></author>
       <author><first>Jonas</first><last>Golde</last><affiliation>Department of Computer Science, Humboldt University Berlin, Humboldt Universität Berlin</affiliation></author>
       <author><first>Alan</first><last>Akbik</last><affiliation>Humboldt Universität Berlin</affiliation></author>
       <pages>175-191</pages>

data/xml/2025.l2m2.xml

@@ -52,7 +52,7 @@
       <title>From Data to Knowledge: Evaluating How Efficiently Language Models Learn Facts</title>
       <author><first>Daniel</first><last>Christoph</last></author>
       <author orcid="0009-0007-4593-2353"><first>Max</first><last>Ploner</last><affiliation>Humboldt Universität Berlin</affiliation></author>
-      <author><first>Patrick</first><last>Haller</last><affiliation>Humboldt Universität Berlin</affiliation></author>
+      <author id="patrick-haller"><first>Patrick</first><last>Haller</last><affiliation>Humboldt Universität Berlin</affiliation></author>
       <author><first>Alan</first><last>Akbik</last><affiliation>Humboldt Universität Berlin</affiliation></author>
       <pages>29-46</pages>
       <abstract>Sample efficiency is a crucial property of language models with practical implications for training efficiency. In real-world text, information follows a long-tailed distribution. Yet, we expect models to learn and recall frequent and infrequent facts. Sample efficient models are better equipped to handle this challenge of learning and retaining rare information without requiring excessive exposure. This study analyzes multiple models of varying architectures and sizes, all trained on the same pre-training data. By annotating relational facts with their frequencies in the training corpus, we examine how model performance varies with fact frequency. Our findings show that most models perform similarly on high-frequency facts but differ notably on low-frequency facts. This analysis provides new insights into the relationship between model architecture, size, and factual learning efficiency.</abstract>

data/xml/2025.naacl.xml

@@ -515,7 +515,7 @@
     <paper id="37">
       <title>Familiarity: Better Evaluation of Zero-Shot Named Entity Recognition by Quantifying Label Shifts in Synthetic Training Data</title>
       <author><first>Jonas</first><last>Golde</last></author>
-      <author><first>Patrick</first><last>Haller</last></author>
+      <author id="patrick-haller"><first>Patrick</first><last>Haller</last></author>
       <author orcid="0009-0007-4593-2353"><first>Max</first><last>Ploner</last></author>
       <author><first>Fabio</first><last>Barth</last></author>
       <author><first>Nicolaas</first><last>Jedema</last></author>

data/yaml/name_variants.yaml

@@ -4015,6 +4015,11 @@
 - canonical: {first: Mark, last: Hall}
   variants:
   - {first: Mark Michael, last: Hall}
+- canonical: {first: Patrick, last: Haller}
+  id: patrick-haller
+  comment: HU Berlin
+  degree: Humboldt Universität zu Berlin
+  orcid: 0009-0006-0445-4765
 - canonical: {first: Patrick, last: Haller}
   id: patrick-haller-zurich
   comment: University of Zurich