Update README.md

README.md

@@ -44,6 +44,6 @@ pinned: false
     <p class="lg:col-span-3">
         The ESB datasets are sourced from 11 different domains and cover a range of audio and text distributions (speaking styles, background noise, transcription requirements). There is no restriction on architecture or training data: any system capable of processing audio inputs and generating the corresponding transcriptions is eligible to participate. The only constraint is that the same training and evaluation algorithms must be used across datasets and systems may not use any dataset-specific pre- or post-processing. The objective of ESB is to encourage the research of more generalisable, multi-domain ASR systems. <br />
         <br />
-        ESB was proposed in ESB: A Benchmark For Multi-Domain End-to-End Speech Recognition by ... For more information, see the official submission on <a  href="https://openreview.net/forum?id=9OL2fIfDLK"  class="underline">OpenReview.net</a> or the blog post at <a href="https://openreview.net/forum?id=9OL2fIfDLK" class="underline">ESB Benchmark (TODO)</a>.
+        ESB was proposed in <i> ESB: A Benchmark For Multi-Domain End-to-End Speech Recognition </i>. For more information, see the official paper on <a  href="https://arxiv.org/abs/2210.13352"  class="underline">Arxiv</a>.
 	</p>
 </div>