Tootfinder

Attribution Gradients: Incrementally Unfolding Citations for Critical Examination of Attributed AI Answers
Hita Kambhamettu, Alyssa Hwang, Philippe Laban, Andrew Head
https://arxiv.org/abs/2510.00361

Attribution Gradients: Incrementally Unfolding Citations for Critical Examination of Attributed AI Answers
AI question answering systems increasingly generate responses with attributions to sources. However, the task of verifying the actual content of these attributions is in most cases impractical. In this paper, we present attribution gradients as a solution. Attribution gradients provide integrated, incremental affordances for diving into an attributed passage. A user can decompose a sentence of an answer into its claims. For each claim, the user can view supporting and contradictory excerpts min…

Trump is very, very angry about a New York Times piece
that carefully documents his obvious physical decline.
Trump raged over the piece,
letting out a stream of lies about his performance in office
and even attacking the reporter who wrote the piece as ugly.
This may seem typical of Trump,
but we think it gets at something deeper.
Trump’s entire political mystique is premised on the fiction that he’s a strong, virile, formidable figure
who …

Transcript: Trump Fury Erupts at NYT as Mental Decline Visibly Worsens
As Trump’s rage at the paper of record reveals too much, the author of piece on Trump as “lame duck dictator” explains how devastating it is for him that his carefully-cultivated illusion of strength is collapsing.

Heute hatte ich ein auf den ersten Blick echt wirkendes Schreiben von Vodafone im Briefkasten, das mir meine neue Service-PIN mitteilt.
Meine Anschrift ist korrekt, die Anrede ist persönlich, keine Grammatik- oder Tippfehler.
Und trotzdem Fake: Die Kundennummer ist bis auf die letzten 3 Stellen ausgeXt (und einen Vertrag mit diesen letzten drei Ziffern habe ich nicht), die Service-PIN ist nicht verdeckt, das Rot ist blasser als das VF-Rot, das Papier ist dünn.
Vorsicht, wenn I…

mR3: Multilingual Rubric-Agnostic Reward Reasoning Models
David Anugraha, Shou-Yi Hung, Zilu Tang, Annie En-Shiun Lee, Derry Tanti Wijaya, Genta Indra Winata
https://arxiv.org/abs/2510.01146

mR3: Multilingual Rubric-Agnostic Reward Reasoning Models
Evaluation using Large Language Model (LLM) judges has been widely adopted in English and shown to be effective for automatic evaluation. However, their performance does not generalize well to non-English settings, and it remains unclear what constitutes effective multilingual training for such judges. In this paper, we introduce mR3, a massively multilingual, rubric-agnostic reward reasoning model trained on 72 languages, achieving the broadest language coverage in reward modeling to date. We …

CodeGenLink: A Tool to Find the Likely Origin and License of Automatically Generated Code
Daniele Bifolco, Guido Annicchiarico, Pierluigi Barbiero, Massimiliano Di Penta, Fiorella Zampetti
https://arxiv.org/abs/2510.01077

CodeGenLink: A Tool to Find the Likely Origin and License of Automatically Generated Code
Large Language Models (LLMs) are widely used in software development tasks nowadays. Unlike reusing code taken from the Web, for LLMs' generated code, developers are concerned about its lack of trustworthiness and possible copyright or licensing violations, due to the lack of code provenance information. This paper proposes CodeGenLink, a GitHub CoPilot extension for Visual Studio Code aimed at (i) suggesting links containing code very similar to automatically generated code, and (ii) whenever …

Hearing the Order: Investigating Selection Bias in Large Audio-Language Models
Yu-Xiang Lin, Chen-An Li, Sheng-Lun Wei, Po-Chun Chen, Hsin-Hsi Chen, Hung-yi Lee
https://arxiv.org/abs/2510.00628

Hearing the Order: Investigating Selection Bias in Large Audio-Language Models
Large audio-language models (LALMs) are often used in tasks that involve reasoning over ordered options. An open question is whether their predictions are influenced by the order of answer choices, which would indicate a form of selection bias and undermine their reliability. In this paper, we identify and analyze this problem in LALMs. We demonstrate that no model is immune to this bias through extensive experiments on six LALMs across three widely used benchmarks and their spoken counterparts…

Pay-Per-Search Models are Abstention Models
Mustafa Omer Gul, Claire Cardie, Tanya Goyal
https://arxiv.org/abs/2510.01152 https://arxiv.org/pdf/2510.01152

Pay-Per-Search Models are Abstention Models
LLMs cannot reliably recognize their parametric knowledge boundaries and often hallucinate answers to outside-of-boundary questions. In contrast, humans recognize their limitations and can either seek external help for such questions or abstain. In this paper, we introduce MASH (Modeling Abstention via Selective Help-seeking), a training framework that readily extracts abstentions from LLMs. Our key idea is that any external help-seeking by an LLM, i.e. search tool use, can serve as a proxy for…

Can't find pencil & paper, can't I just text the answer in? #TOTP

I recently debugged sound staying mute. Yet the speakers, bluetooth, pipewire, gnome all claimed everything is fine and connected.
That day I learned you can mute a tab group in a Sideberry.
via @…
https…

Since Ghislaine Maxwell’scontroversial transfer to a low-security prison camp this summer,
her time at Texas’s FPC Bryan has prompted uproar over alleged favorable treatment
– including claims this week that she was provided custom-made meals and access to a puppy
Some of the recent accusations were in a 9 November letter that Jamie Raskin sent to Donald Trump.
The Democratic representative’s letter, which cited whistleblower information, demanded answers about Maxwel…