Tootfinder

Helsingille uusi luonnonsuojelualueohjelma – suojeltavaksi esitetään yli 100 uutta kohdetta https://www.hel.fi/fi/uutiset/helsingille-uusi-luonnonsuojelualueohjelma-suojeltavaksi-esitetaan-yli-100-uutta-kohdetta

Helsingille uusi luonnonsuojelualueohjelma – suojeltavaksi esitetään yli 100 uutta kohdetta
Kaupunginhallitus hyväksyi kevätkauden viimeisessä kokouksessaan 23. kesäkuuta Helsingin uuden luonnonsuojelualueohjelman vuosille 2025–2040. Lisäksi kaupunginhallitus hyväksyi osaltaan Ressun peruskoulun perusparannuksen hankesuunnitelman.

"Tulipalot | Vimpelin Veto menetti tulipalossa koripallonsa, seuralta hätähuuto"
https://some.henkkalaukka.fi/@uutisbot/114755355364450421
Onneksi koripallot eivät ole kalliita. Biltemasta saavat uuden alle kympillä

Wecken durch Motorsägen und Laubbläser. Der Sommer in Heiligensee ist wie der Herbst. Und der Winter.

huh, i just found this. it is clearly llm generated, and thus riddled with inaccuracies and hallucinations. but on first glance, i am very much impressed by the stuff that doesn't immediately jump at me as wrong: https://deepwiki.com/stef/libopaque and

AggTruth: Contextual Hallucination Detection using Aggregated Attention Scores in LLMs
Piotr Matys, Jan Eliasz, Konrad Kie{\l}czy\'nski, Miko{\l}aj Langner, Teddy Ferdinan, Jan Koco\'n, Przemys{\l}aw Kazienko
https://arxiv.org/abs/2506.18628

AggTruth: Contextual Hallucination Detection using Aggregated Attention Scores in LLMs
In real-world applications, Large Language Models (LLMs) often hallucinate, even in Retrieval-Augmented Generation (RAG) settings, which poses a significant challenge to their deployment. In this paper, we introduce AggTruth, a method for online detection of contextual hallucinations by analyzing the distribution of internal attention scores in the provided context (passage). Specifically, we propose four different variants of the method, each varying in the aggregation technique used to calcul…

Halogen abundance evidence for the formation and metasomatism of the primary lunar crust: #moon –> new clues about formation.

Recent hype is that o3 solved AI web search.
But so far I've tried using it for two "hard" search problems (meaning I first tried searching manually and didn't find much). Both times it hallucinated a ton, making claims that weren't actually in the sources.
On the other hand, both times, I did ultimately find nuggets of useful information after picking through all the hallucinations. So I'm glad I used it?
[1/3]

KI-Drohnen: Helsing erhält weitere 600 Millionen Euro Investorengelder
Die Staaten Europas wollen ihre Streitkräfte verteidigungsbereit und möglichst unabhängig machen. Damit fließen auch enorme Summen in die Rüstungsbranche.

COIN: Uncertainty-Guarding Selective Question Answering for Foundation Models with Provable Risk Guarantees
Zhiyuan Wang, Jinhao Duan, Qingni Wang, Xiaofeng Zhu, Tianlong Chen, Xiaoshuang Shi, Kaidi Xu
https://arxiv.org/abs/2506.20178

COIN: Uncertainty-Guarding Selective Question Answering for Foundation Models with Provable Risk Guarantees
Uncertainty quantification (UQ) for foundation models is essential to identify and mitigate potential hallucinations in automatically generated text. However, heuristic UQ approaches lack formal guarantees for key metrics such as the false discovery rate (FDR) in selective prediction. Previous work adopts the split conformal prediction (SCP) framework to ensure desired coverage of admissible answers by constructing prediction sets, but these sets often contain incorrect candidates, limiting the…

BLOCKS: Blockchain-supported Cross-Silo Knowledge Sharing for Efficient LLM Services
Zhaojiacheng Zhou, Hongze Liu, Shijing Yuan, Hanning Zhang, Jiong Lou, Chentao Wu, Jie Li
https://arxiv.org/abs/2506.21033

BLOCKS: Blockchain-supported Cross-Silo Knowledge Sharing for Efficient LLM Services
The hallucination problem of Large Language Models (LLMs) has increasingly drawn attention. Augmenting LLMs with external knowledge is a promising solution to address this issue. However, due to privacy and security concerns, a vast amount of downstream task-related knowledge remains dispersed and isolated across various "silos," making it difficult to access. To bridge this knowledge gap, we propose a blockchain-based external knowledge framework that coordinates multiple knowledge silos to pr…