Thanks, Hannah. Appreciate it. If I understand right, you are flagging existing gaps and what a future AI system that is built on WD can help curate in the hopes they can be easily fixed. These are indeed some of the examples we are thinking about for a future system can help reveal (though not the only ones). We'll hopefully get there, but this is only the first step.
For now, the best help the community can give us is help sharing queries, so we can train the model to answer them accurately. We'll keep everyone posted. Let me know if you'd like additional details on AI-BRIGDES. There are also monthly Open Forum meetings on the last Friday of each month (15:00 UK time) if the topic is of interest. More details on speakers etc are in the dedicated google group <https://groups.google.com/g/ai-bridges>. Best, Shani. On Mon, Oct 12, 2026 at 4:30 PM Hannah Clover via Wikimedia-l < [email protected]> wrote: > Also I've noticed that sometimes Wikimedia Commons will have categories > for surnames (https://commons.wikimedia.org/wiki/Category:Surnames) that > don't always have wikidata items. Some way of checking that en masse > instead of manually sounds like a massive use case. Assuming I'm > understanding what you're trying to do here correctly. > > On Mon, Oct 12, 2026, 11:15 a.m. Hannah Clover < > [email protected]> wrote: > >> Regarding things being mentioned in one place but not others, I think >> this may have potential in other contexts. For example, something I've done >> off and on throughout the years is create a few thousand redirects about >> people's names. Sometimes people have middle names and it's pretty common >> for women to have different last names at birth vs when they got married. A >> list being created for community review to skip the most time lengthy part >> of finding these gaps might be useful. >> >> On Mon, Oct 12, 2026, 11:08 a.m. Shani Evenstein Sigalov via Wikimedia-l < >> [email protected]> wrote: >> >>> Hello all, >>> >>> I'm happy to share the first open release from AI-BRIDGES >>> <https://ai-bridges.org>, funded by Wikimedia Switzerland >>> <https://meta.wikimedia.org/wiki/Wikimedia_CH/Grant_apply/AI-BRIDGES:_Synthetic_Reasoning_Corpus> >>> and delivered by Pleias <https://pleias.ai/>, with support from >>> Wikimedia Deutschland. >>> >>> We built a harness that lets a language model explore a frozen, >>> six-language Wikidata snapshot step by step, reading entities and citing >>> the statements it used, rather than stuffing large amounts of data into its >>> context, taking a Recursive Language Model approach. From it we generated a >>> Synthetic Reasoning Corpus of exploration traces, released under CC0. >>> >>> A few results from the technical report: >>> >>> - >>> >>> 74 of 100 questions answered correctly by Qwen3.8-27B, an >>> open-weight model on a single GPU, vs 61 for gpt-6-luna in the same >>> harness >>> - >>> >>> 10,235 solving traces released under CC0, over single-entity and >>> multi-hop questions >>> - >>> >>> The harness lifts both models we ran under both interfaces, with the >>> gain mostly on multi-hop questions >>> >>> Of particular interest to editors: exploration surfaced cross-language >>> inconsistencies and misplaced information (for example, nationality >>> appearing in descriptions but not in statements). Nothing was edited >>> automatically; the findings are in the report for community review. >>> >>> - >>> >>> Harness, code and benchmark (GitHub): >>> https://github.com/aibridges/RLM_Wikidata >>> >>> >>> - >>> >>> Corpus (Hugging Face): >>> https://huggingface.co/datasets/AI-BRIDGES/wikidata-search-traces >>> - >>> >>> Technical report: >>> https://arxiv.org/abs/2610.06650 >>> >>> >>> Next, we plan to train open Wikidata search models on this corpus and we >>> would appreciate your help: if you work with GLAMs, Education, >>> Research, especially in the Humanities, please send us real questions >>> you'd want to ask Wikidata in plain language or in SPARQL >>> <https://docs.google.com/document/d/1ipIzwSwf-xCtU7R7ewNCCND7vk0yha6byWMNs3OyIEc/edit?tab=t.0>, >>> or via email: [email protected]. >>> >>> I'd like to take a moment to thank not just Pleias, WMCH & WMDE, but >>> also our other core partners, WMUK & WM Brasil, as well as other >>> partners -- organizations, institutions and so many wonderful individuals >>> from this community -- for being part of the AI-BRIDGES journey and >>> supporting it. >>> >>> Feedback, questions and critique are all very welcome on-list, on wiki >>> <https://meta.wikimedia.org/wiki/Research:AI-BRIDGES> or privately. >>> >>> All best wishes, Shani. >>> >>> >>> ----------------------- >>> Dr. Shani Evenstein Sigalov >>> Marie Curie Postdoctoral Research Fellow & AI-BRIDGES >>> <https://ai-bridges.org/> lead >>> Digital Humanities Research Hub, School of Advanced Study, University of >>> London. >>> _______________________________________________ >>> Wikimedia-l mailing list -- [email protected], guidelines >>> at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and >>> https://meta.wikimedia.org/wiki/Wikimedia-l >>> Public archives at >>> https://lists.wikimedia.org/hyperkitty/list/[email protected]/message/54MO3NDOYVUGV5D4G2AW3VZ2TZJ52IZR/ >>> To unsubscribe send an email to [email protected] >> >> _______________________________________________ > Wikimedia-l mailing list -- [email protected], guidelines > at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and > https://meta.wikimedia.org/wiki/Wikimedia-l > Public archives at > https://lists.wikimedia.org/hyperkitty/list/[email protected]/message/CKQLILYA67FY3QGD2TWWKRQH7D46ETX7/ > To unsubscribe send an email to [email protected]
_______________________________________________ Wikimedia-l mailing list -- [email protected], guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l Public archives at https://lists.wikimedia.org/hyperkitty/list/[email protected]/message/EJT7ESZJNUB73MV7T5N22H3OLKE6JARY/ To unsubscribe send an email to [email protected]
