Pablo Castells
@pcastells
Followers
815
Following
686
Media
10
Statuses
845
Principal Research Scientist at Amazon
Madrid, Spain
Joined June 2009
Photos from @ChenglongM's presentation of "Temporal Conformity-aware Hawkes Graph Network for Recommendations" at the #TheWebConf2004 last week. Preprint available here https://t.co/PCghK2eXeH
@pcastells @RMITComputing @AdmsCentre
Huge congratulations to @ChenglongM for acceptance of his #TheWebConf2024 paper "Temporal Conformity-aware Hawkes Graph Network for Recommendations" with Yongli Ren @pcastells @RMITComputing @AdmsCentre
0
3
10
On arXiv now is a draft chapter on "Generative Information Retrieval Evaluation", examining both GenIR evaluation and using generative for evaluation. Love to hear from you all of references we missed. Co-authored by @Marwah_k @claclarke @NegarEmpr
arxiv.org
In this chapter, we consider generative information retrieval evaluation from two distinct but interrelated perspectives. First, large language models (LLMs) themselves are rapidly becoming tools...
2
10
60
Are you into generative AI for document retrieval, grounded answer generation, recommendation, knowledge graphs and other modalities? Our paper submission deadline is coming up: 🗓 April 25th Gen-IR #SIGIR2024
https://t.co/ej3iCcvKyA
coda.io
The Second Workshop on Generative Information Retrieval
0
4
13
💡Important PSA: are you a seasoned #RecSys veteran who already has their single-column ACM template ready to go? We're aligning formatting with other ACM conferences to use the double-column format! Spread the word so no one has to reformat in panic minutes before the deadline.
0
6
11
'More work, less talk' may be a day-to-day motto, but we actually are looking for talks now! The call for industry talks is out: https://t.co/6h9o32jeaD.
#RecSys2024 is looking for industry practitioners who tackle challenges and find solutions to real-problems. Share your work!
0
6
11
Huge congratulations to @ChenglongM for acceptance of his #TheWebConf2024 paper "Temporal Conformity-aware Hawkes Graph Network for Recommendations" with Yongli Ren @pcastells @RMITComputing @AdmsCentre
0
3
19
Interested to know how we can combine effective confidence modelling with competitive recommendation performance and why that matters? Then join the presentation of our #RecSys2023 work with @HarrieOos in Hall406CX at 14:40 Singapore time! Paper:
dl.acm.org
1
2
18
Listening to @yashonwu presenting our work at #recsys2023 - the paper is entitled “Goal-Oriented Multi-Modal Interactive Recommendation with Verbal and Non-Verbal Relevance Feedback”; Now in Session 7: Interactive Recommendation 1
0
6
26
.@asash is presenting our work gSASRec: reducing overconfidence in sequential recommendation trained with negative sampling @ACMRecSys #recsys2023
1
3
31
Extended deadline for LERI23 at #RecSys2023 Submission due on August 10th, 2023! We welcome papers on all aspects of impressions for #recsys: models including sequence and context-aware, evaluation, OPE, perspectives, industry... Check our CFP https://t.co/YV65ePDCA5
@ACMRecSys
1
3
3
Still two weeks available to submbit to LERI23, the Workshop on Learning and Evaluating Recommendations with Impressions at #RecSys2023 Submissions may cover all aspects related to the use of impressions for #recsys Check our CFP here https://t.co/YV65ePEapD
@ACMRecSys
1
5
2
Happy to be co-organizer of the Workshop on Learning and Evaluating Recommendations with Impressions, that will be held at #RecSys2023 this September. @ACMRecSys @crmpla67 @pcastells @JustinBasilico Paper submission is in early august, check our CFP here
0
5
9
In the LLM-science discussion, I see a common misconception that science is a thing you do and that writing about it is separate and can be automated. I’ve written over 300 scientific papers and can assure you that science writing can’t be separated from science doing. Why? 1/18
38
468
2K
"NEST: Simulating Pandemic-like Events for Collaborative Filtering by Modeling User Needs Evolution" by my student Chenglong Ma, along with @pcastells and Yongli Ren, presented at #cikm2022, is now available for download https://t.co/yX3N18MNOc
@rmit_cidda @RMITComputing
0
3
7
Stellar relevance dataset combining classic LTR features with text queries and documents. Classic LTR with GBDT is still alive. From @macavaney @ntonellotto et al. Thanks to @srchvrs for pointing me to the paper and dataset. https://t.co/s1a0UcJftR
https://t.co/rPDDdrN4W3
1
5
36
Back home from #recsys2022? Then it's time to think about your next project and keep in mind that #ACM_TORS has currently three special issues with open calls relating to #trust, #CausalInference and #Evaluation of #recsys: https://t.co/CzYRUZa3JF
#CfP
0
6
16
What a wonderful week at #recsys2022 this was! Thanks and kudos to the General Chairs, and all the organizing team. Great program, awesome workshops and tutorials. Looking forward to #recsys2023 in Singapore!
0
2
33
Now @JavierSanzCruza is giving an overview of @Infinitech_EU to the #finrec #recsys2022 project - including hearing a snippet by @jsoldatos et al
0
6
21
I'm psyched to be giving a keynote at the @ACMRecSys Workshop for Online Recommender Systems. I'll be talking about ... ... juice. And a bit about real-time recommender systems. Friday, Sep 23rd, 2pm. Come say hi! Details: https://t.co/pC4q45eEUZ,
@orsum_ws
#RecSys2022
5
7
43