Tootfinder

Replaced article(s) found for econ.TH. https://arxiv.org/list/econ.TH/new
[1/1]:
- Reputational cheap talk: influentialness and welfare
Allen Vong
https://arxiv.org/abs/2505.11877 https://mastoxiv.page/@arXiv_econTH_bot/114538953311274731
- Local Strategy-proofness and Dictatorship
Abinash Panda, Anup Pramanik, Ragini Saxena
https://arxiv.org/abs/2507.00913 https://mastoxiv.page/@arXiv_econTH_bot/114782653634807033
- Endogenous Inequality Aversion: Decision criteria for triage and other ethical tradeoffs
Federico Echenique, Teddy Mekonnen, M. Bumin Yenmez
https://arxiv.org/abs/2601.22250 https://mastoxiv.page/@arXiv_econTH_bot/115999970982292698
- Generalized Multidimensional Contests with Asymmetric Players: Equilibrium and Optimal Prize Design
Siyuan Fan, Zhonghong Kuang, Jingfeng Lu
https://arxiv.org/abs/2602.21564 https://mastoxiv.page/@arXiv_econTH_bot/116136166100174153
- Stable Matchings with Choice Correspondences Under Acyclicity
Varun Bansal, Mihir Bhattacharya, Ojasvi Khare
https://arxiv.org/abs/2603.23038 https://mastoxiv.page/@arXiv_econTH_bot/116288671686912472
- Calibrated Forecasting and Persuasion
Atulya Jain, Vianney Perchet
https://arxiv.org/abs/2406.15680 https://mastoxiv.page/@arXiv_csGT_bot/112675926662453962
- Feedback-Coupled Memory Systems: A Dynamical Model for Adaptive Coordination
Stefano Grassi
https://arxiv.org/abs/2603.11560 https://mastoxiv.page/@arXiv_csMA_bot/116220713458383739
toXiv_bot_toot

A #PaintedLady came to visit yesterday (Vanessa cardui, #Distelfalter) - shot from a looong distance with maximum optical and some digital zoom. Hadn't known this #butterfly species but https://www.hr4.de/programm/themen/portraets-von-12-heimischen-schmetterlingsarten-v2,zwoelf-einheimische-schmetterlingsarten-100.html readily identified it.

Grounding Pages sind die AMP-Seiten der KI-Ära. 🤷
Eine parallele Fakten-Seite nur für Maschinen? Das hat schon bei AMP nicht funktioniert. KI-Systeme werden besser darin, normale Webseiten zu lesen – nicht schlechter.
Stattdessen: Über-uns-Seite als Entity-Dokument schreiben. Dritte Person, Hard Facts, keine Floskeln. Funktioniert für Google, ChatGPT und Kunden gleichzeitig.

Grounding Pages? So wird deine Über-uns-Seite zum Entity-Dokument 🏗️
Grounding Pages sind Overengineering. Wie du deine Über-uns-Seite so schreibst, dass Google und KI-Systeme deine Marke als Entität erkennen – ohne Parallelseiten.

Thousands in US to join 'no school, no work, no shopping' May Day protest in economic blackout (Lex McMenamin/The Guardian)
https://www.theguardian.com/us-news/2026/may/01/may-day-strong-economic-protests
http://www.memeorandum.com/260501/p67#a260501p67

Cybersecurity developments don't stop on the weekend, so check out today's Metacurity for the most important infosec news you might have missed since Friday, including
--DOGE-aligned White House web projects funnel citizen data to analytics firm,
--Microsoft's threat to security researcher draws criticism,
--Commerce IG says NIST has mismanaged NVD,
--Obama White House Instagram was hacked,
--Teen researcher flagged flaws in India's school exam bo…

DOGE-aligned White House web projects funnel citizen data to analytics firm
Microsoft's threat to security researcher draws criticism, Commerce IG says NIST has mismanaged NVD, Obama White House Instagram was hacked, Teen researcher flagged flaws in India's school exam board website, Gravity Bridge exploited for $5.4m, DxSale legacy site hacked for $7.3m, much more

Courtroom-Style Multi-Agent Debate with Progressive RAG and Role-Switching for Controversial Claim Verification
Masnun Nuha Chowdhury, Nusrat Jahan Beg, Umme Hunny Khan, Syed Rifat Raiyan, Md Kamrul Hasan, Hasan Mahmud
https://arxiv.org/abs/2603.28488 https://arxiv.org/pdf/2603.28488 https://arxiv.org/html/2603.28488
arXiv:2603.28488v1 Announce Type: new
Abstract: Large language models (LLMs) remain unreliable for high-stakes claim verification due to hallucinations and shallow reasoning. While retrieval-augmented generation (RAG) and multi-agent debate (MAD) address this, they are limited by one-pass retrieval and unstructured debate dynamics. We propose a courtroom-style multi-agent framework, PROClaim, that reformulates verification as a structured, adversarial deliberation. Our approach integrates specialized roles (e.g., Plaintiff, Defense, Judge) with Progressive RAG (P-RAG) to dynamically expand and refine the evidence pool during the debate. Furthermore, we employ evidence negotiation, self-reflection, and heterogeneous multi-judge aggregation to enforce calibration, robustness, and diversity. In zero-shot evaluations on the Check-COVID benchmark, PROClaim achieves 81.7% accuracy, outperforming standard multi-agent debate by 10.0 percentage points, with P-RAG driving the primary performance gains ( 7.5 pp). We ultimately demonstrate that structural deliberation and model heterogeneity effectively mitigate systematic biases, providing a robust foundation for reliable claim verification. Our code and data are publicly available at https://github.com/mnc13/PROClaim.
toXiv_bot_toot

„Deutschland denkt vom Produkt her. Der industrielle Kern – Maschinenbau, Automobil, Mittelstand, Hidden Champions – folgt einem tief verwurzelten Prinzip: Das Produkt trifft die Entscheidungen, die anderswo von der Marke getroffen werden.“
💯 Punktlandung von Kim Notz.
#marketing #newsletter

Happy outcome of the #SolarEclipse 2026 dress rehearsal at the #planetarium in #Bochum, Germany, today: the whole event from beginning 14° high through maximum eclipse in 6° elevation down to 3° can be folllowed without obstruction from a specific - and covenient - zone on the premises. This picture was taken at the very 'moment' of maximum eclipse with the Sun in my back: whereever it's shining here, it could be seen. Not ideal for telescopes or big tripods, but many people with eclipse glasses and handheld cameras could be served. Only the weather on 12 August has to be as gorgeous as today ...

Marco DeepResearch: Unlocking Efficient Deep Research Agents via Verification-Centric Design
Bin Zhu, Qianghuai Jia, Tian Lan, Junyang Ren, Feng Gu, Feihu Jiang, Longyue Wang, Zhao Xu, Weihua Luo
https://arxiv.org/abs/2603.28376 https://arxiv.org/pdf/2603.28376 https://arxiv.org/html/2603.28376
arXiv:2603.28376v1 Announce Type: new
Abstract: Deep research agents autonomously conduct open-ended investigations, integrating complex information retrieval with multi-step reasoning across diverse sources to solve real-world problems. To sustain this capability on long-horizon tasks, reliable verification is critical during both training and inference. A major bottleneck in existing paradigms stems from the lack of explicit verification mechanisms in QA data synthesis, trajectory construction, and test-time scaling. Errors introduced at each stage propagate downstream and degrade the overall agent performance. To address this, we present Marco DeepResearch, a deep research agent optimized with a verification-centric framework design at three levels: \textbf{(1)~QA Data Synthesis:} We introduce verification mechanisms to graph-based and agent-based QA synthesis to control question difficulty while ensuring answers are unique and correct; \textbf{(2)~Trajectory Construction:} We design a verification-driven trajectory synthesis method that injects explicit verification patterns into training trajectories; and \textbf{(3)~Test-time scaling:} We use Marco DeepResearch itself as a verifier at inference time and effectively improve performance on challenging questions. Extensive experimental results demonstrate that our proposed Marco DeepResearch agent significantly outperforms 8B-scale deep research agents on most challenging benchmarks, such as BrowseComp and BrowseComp-ZH. Crucially, under a maximum budget of 600 tool calls, Marco DeepResearch even surpasses or approaches several 30B-scale agents, like Tongyi DeepResearch-30B.
toXiv_bot_toot