ELLIS ESSIR 2025, the European Summer School on Information Retrieval, is taking place in Wolverhampton, UK, July 7–11. Don't miss exciting lectures and tutorials around IR, AI, UX, etc!
More info 👉 https://2025.essir.eu
#InformationRetrieval
Registration for ELLIS ESSIR 2025, the European Summer School on Information Retrieval, is now open! It’s happening from July 7th to 11th in Wolverhampton, UK.
More exciting news – @elsa-ai.bsky.social, the European Lighthouse on Secure and Safe AI, is offering travel support! 🎉
Check out the registration and support links:
https://2025.essir.eu/attending/registration
https://2025.essir.eu/attending/travel-support
#IR #AI #InformationRetrieval #SummerSchool
@ELLISforEurope
#jobs #stellenangebote #GESISjobs #jobfairy #ResearchAssociate #InformationRetrieval #KnowledgeGraphs
Starting on the next possible date our Team Human Information Interaction in #Cologne is looking for a Research Associate Information Retrieval / Knowledge Graphs
(Salary group 13 TV-L, working time 75 %, limited to 4 years):
https://gesis.jobs.personio.de/job/2062129?language=de&display=en
AI Search Tools: Can They Be Manipulated to Return False or Malicious Results?
AI search tools are improving, but hidden manipulation is a concern. Malicious actors can use "prompt injection" to influence results, potentially leading to inaccurate information, especially in reviews. Be cautious when relying on these tools! #AISearch #AIPitfalls #HiddenContent #PromptInjection #AIReliability #InformationRetrieval
https://tech-champion.com/general/ai-search-tools-can-they-be-manipulated-to-r...
[Veille] Les consortiums @HumaNum PictorIA et MASAplus, le GT DoReMITI du @cnrs organisent 2 journées sur la thématique :
"Images et IA : faire dialoguer les communautés" https://imagia2025.sciencesconf.org
Super initiative 👍
#IA #ESR #recherche #digitalhumanties #STEM #SHS #images #deeplearning #ML #researchdata #imageprocessing #informationretrieval #machinelearning #openscience
AI Search Tools: Can They Be Manipulated to Return False or Malicious Results?
AI search tools are improving, but hidden manipulation is a concern. Malicious actors can use "prompt injection" to influence results, potentially leading to inaccurate information, especially in reviews. Be cautious when relying on these tools! #AISearch #AIPitfalls #HiddenContent #PromptInjection #AIReliability #InformationRetrieval
https://tech-champion.com/general/ai-search-tools-can-they-be-manipulated-to-r...
#ausdembildarchiv
Heterogene Angaben der 128 eingebundenen Bilddatenbanken in den Datierungsfeldern sind teilweise nicht direkt maschinenlesbar und müssen quasi übersetzt werden: Wie viele Jahre meint ein ca.-Zeitraum? ...
https://prometheus-bildarchiv.de/blog/suchergebnisse-gezielt-nach-datierung-filtern
#Suchergebnisse #filtern #Datierung #Datierungsfilter #ErweiterteSuche
#Retrieval #InformationRetrieval
Hilbert spaces are mathematical constructs used in information retrieval to represent and analyze data in high-dimensional vector spaces. They provide a framework for modeling the relationships and similarities between documents, queries, and other information elements, enabling advanced techniques like latent semantic analysis and vector space models to improve the accuracy and effectiveness of search and recommendation systems.
Meet the newly elected BCS IRSG committee! We have a new chair, treasurer and several other new committee members. Interested in shaping the future of the British and European IR community while meeting wonderful people? Join us in this year's election!
https://medium.com/bcs-irsg-informer/meet-the-newly-elected-bcs-irsg-committee-d4627ffab919
Mitarbeiter suchen 10 Stunden pro Woche nach Informationen
https://www.heise.de/news/Ineffizienz-im-Buero-Mitarbeiter-suchen-10-Stunden-pro-Woche-nach-Informationen-10309833.html
Vorsicht, ist eine #Atlassian-Studie. Dahinter steckt eine klare Marketing-Agenda.
Ad #Jira und #Confluence: hätte ich nicht meinen #orgmode in Kombination mit Browser-Bookmarks, ich würde sehr viele Informationen ständig nicht mehr wiederfinden. 🤷 Da hat Atlassian noch einen langen Weg vor sich und ich erkenne aus den letzten 10 Jahren keine großartigen Verbesserungen in dieser Richtung.
Original: https://atlassianblog.wpengine.com/wp-content/uploads/2025/03/the-state-of-teams-2025.pdf
Google is innovating search in an amazing and yet predictable way — it’s removing the search results. There will only be the “AI Overview,” much loved by nobody.From https://circumstances.run/users/davidgerard/statuses/114122009704468173
You might think the point of search was to find a thing, then go to it. But then Google doesn’t make enough ad money. Trying to wheedle an answer out of a hallucinating robot might keep you on the page longer.
Edit: Reading the Google blog post about this change more carefully (thanks @picklish@weirder.earth ), it seems more of an experimental feature than a default. I think the worry that Google will eventually make "AI overview" the default is a real and valid one, but I don't it's the case right now. If it ever does happen we will need to stop using the word "Google" as a synonym for "search" and declare RIP Google, 1998 - 2025
#IR #InformationRetrieval #search #Google #RIPGoogle #AISlop
How the Index Card Cataloged the World (2017) — https://www.theatlantic.com/technology/archive/2017/12/how-the-index-card-catalogued-the-world/547271/
#HackerNews #IndexCardHistory #CatalogingTechnology #InformationRetrieval #TheAtlantic
I keep hearing that the popular search engine Solr / Elastic Search is really heaving on compute resources.
For example, mastodon admins saying search takes a big chunk of cpu.
I'm no expert - is this view correct?
And why aren't lightweight (non-Java) search engines more popular? Do they even exist?
Would it be good to create one?
Xu Z, Mo F, Huang Z, Zhang C, Yu P, Wang B, Lin J, Srikumar V. A Survey of Model Architectures in Information Retrieval. arXiv preprint arXiv:2502.14822. 2025 Feb 20. https://arxiv.org/abs/2502.14822
Good overview. learned about people trying to use state space models to replace transformers. But at this point, review papers in IR should be a website, or a wiki style article. we have too many review papers. #InformationRetrieval
Quite a good list of LLM evaluation metrics (with papers!) by Parea AI: https://docs.parea.ai/blog/eval-metrics-for-llm-apps-in-prod
#llms #rag #evaluation #eval #metrics #faithfulness #relevance #informationretrieval
Aaron Tay, heading down the rabbit hole of Retrieval Augmented Generation (RAG) systems...😮
When is a Hallucination Not a Hallucination? The Role of Implicit Knowledge in RAG
https://musingsaboutlibrarianship.blogspot.com/2025/02/when-is-hallucination-not-hallucination.html @aarontay #RAG #Hallucinations #GenAI #LLM #retrieval #InformationRetrieval
"Sammeln, bewahren, forschen, ausstellen, vermitteln: Eine Metadatenanalyse digitaler Sammlungen ausgewählter Museen aus dem DACH-Raum im Kontext musealer Kernaufgaben": Masterarbeit von Katja Weingartshofer @IBI_HU: https://doi.org/10.18452/31462
#Museum #Metadaten #InformationRetrieval
☝️Remember the SIGIR 2024 Conference in Washington where our colleague Hidir Aras presented the fifth PatentSemTech workshop together with researchers from TU Wien, ZBW Kiel and other institutions?
👉You can find the workshop proceedings online now!
Link: https://ceur-ws.org/Vol-3775/
#SIGIR24
#PatentSemTech
#IP
#InformationRetrieval
#Patents
#ACM
Information Retrieval: Advanced Topics and Techniques
ACM Books, December 2024
If you are into #DiamondOpenAccess or into #InformationRetrieval (or both!) then you should definetely folliow: @IRRJ
Client Info
Server: https://mastodon.social
Version: 2025.04
Repository: https://github.com/cyevgeniy/lmst