ddiruscio/logseq
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

[logseq-plugin-git:commit] 2025-06-09T15:15:16.623Z

Browse Source
This commit is contained in:
Davide Di Ruscio
2025-06-09 17:15:17 +02:00
parent 477d636969
commit 9e24b7b6b8
1 changed files with 2 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
pages/RecSys_2025_Reproducibility_711.md
+2 -1
View File
@@ -17,4 +17,5 @@ collapsed:: true
- Overall presentation: Overall, the paper is well written and structured. Both the goal and the proposed solution are properly described. I have only a few concerns related to some missing justifications or explanations. In particular, I'm not convinced on why a mobile setup is considered and why such a setup has been used to compare the original environments with that presented in this work.
- Technical Soundness: The [[Paper]] defines interesting research questions especially that concerning the inference phase. The methodological pipeline is clearly described in the online material, The insights of the first research questions, while valid, are obvious and expected to me (i[[.]]e., newer hardware being more efficient). It is necessary to stress what's the take away message and why it is [[IMPORTANT]], and especially what's can be built on top of it.
- Reproducibility of methods: The online appendix contains all the artifacts that are needed to reproduce the experiments. I've played a bit with the docker image made available by the authors. Even though I could not execute the whole process until the end, I managed to easly start the pipelines by following the instructions made available by the authors.
- Impact: Carbon footprint in RS is indeed a relevant and interesting topic. While some results are insightful (e.g., training/inference trafe-offs) other are rather expected may not be impactfull for the community. However, the emphasis on model validation overhead adds useful considerations for future research direction related to sustainable RS.
- Impact: Carbon footprint in RS is indeed a relevant and interesting topic. While some results are insightful (e[[.]]g., training/inference trafe-offs) other ones are rather expected and may not be impactfull for the community. However, the emphasis on model validation overhead adds useful considerations for future research direction related to sustainable RS.
- Artifacts (Availability and Documentation):