Tootfinder

When Punctuation Matters: A Large-Scale Comparison of Prompt Robustness Methods for LLMs
Mikhail Seleznyov, Mikhail Chaichuk, Gleb Ershov, Alexander Panchenko, Elena Tutubalina, Oleg Somov
https://arxiv.org/abs/2508.11383

When Punctuation Matters: A Large-Scale Comparison of Prompt Robustness Methods for LLMs
Large Language Models (LLMs) are highly sensitive to subtle, non-semantic variations in prompt phrasing and formatting. In this work, we present the first systematic evaluation of 5 methods for improving prompt robustness within a unified experimental framework. We benchmark these techniques on 8 models from Llama, Qwen and Gemma families across 52 tasks from Natural Instructions dataset. Our evaluation covers robustness methods from both fine-tuned and in-context learning paradigms, and tests …

[2025-07-17 Thu (UTC), 5 new articles found for cs.LO Logic in Computer Science]
toXiv_bot_toot

I scored 5/21 on https://e-mail.wtf and all I got was this lousy text to share on social media.
Sadly, this means I'm not allowed to Fediverse anymore.
For today.

Email is Easy
Everyone knows what an email address is, right?

Die fossile Rechte des #Staddrat freundschaftlich vereint.
Foto von Holger #Zastrow|s Instagramprofil.

"Es gibt zwei Quellen für Vertrauen in den Staat. Und die erste Quelle für Vertrauen in den Staat ist, wenn der Staat ein Papa ist, der sich gut kümmert. [...] Dieses Vertrauen währt genau solange, bis der Wohlstand sich mindert. Bis ich Zumutungen erfahre. In diesem Modell kann ich meinen Bürgern nichts zumuten. [...]" (1/5)
@… im Gespräch mit

I've got a 3am start with derpmaster Mike coming up in the next few days. He normally gets up at 5.30am so I've got a feeling there's going to be no settling him after I arrive and his regular humans all leave.
Still, hot weather will mean an early walk and breakfast and then we can both snooze!
My fourth year of looking after Michael.
#Greyhounds

Performance of GPT-5 Frontier Models in Ophthalmology Question Answering
Fares Antaki, David Mikhail, Daniel Milad, Danny A Mammo, Sumit Sharma, Sunil K Srivastava, Bing Yu Chen, Samir Touma, Mertcan Sevgi, Jonathan El-Khoury, Pearse A Keane, Qingyu Chen, Yih Chung Tham, Renaud Duval
https://arxiv.org/abs/2508.09956

Performance of GPT-5 Frontier Models in Ophthalmology Question Answering
Large language models (LLMs) such as GPT-5 integrate advanced reasoning capabilities that may improve performance on complex medical question-answering tasks. For this latest generation of reasoning models, the configurations that maximize both accuracy and cost-efficiency have yet to be established. We evaluated 12 configurations of OpenAI's GPT-5 series (three model tiers across four reasoning effort settings) alongside o1-high, o3-high, and GPT-4o, using 260 closed-access multiple-choice que…

Improving Generative Cross-lingual Aspect-Based Sentiment Analysis with Constrained Decoding
Jakub \v{S}m\'id, Pavel P\v{r}ib\'a\v{n}, Pavel Kr\'al
https://arxiv.org/abs/2508.10369

Improving Generative Cross-lingual Aspect-Based Sentiment Analysis with Constrained Decoding
While aspect-based sentiment analysis (ABSA) has made substantial progress, challenges remain for low-resource languages, which are often overlooked in favour of English. Current cross-lingual ABSA approaches focus on limited, less complex tasks and often rely on external translation tools. This paper introduces a novel approach using constrained decoding with sequence-to-sequence models, eliminating the need for unreliable translation tools and improving cross-lingual performance by 5\% on ave…

Regardless of the weather, the cyber world is always hot, so check out today's Metacurity for the crucial infosec developments you should know, including
--M&S and Qantas leaders remain mum on ransomware payments,
--M&S chairman says two other British companies' ransomware attacks have gone unreported,
--5.7m customers impacted by Qantas attack,
--US sanctions DPRK man for IT worker scheme,
--DoJ seeks to bring COVID hacker to US,
--Rubio im…

M&S and Qantas leaders remain mum on ransomware payments
M&S chairman says two other British companies' ransomware attacks have gone unreported, 5.7m customers impacted by Qantas attack, US sanctions DPRK man for IT worker scheme, DoJ seeks to bring COVID hacker to US, Rubio impersonated in AI voice effort, MSFT patches 137+ flaws, much more

Woman running at the Thom B Classic in Hammond Hill State Forest -- This Sunday I'm going to be at the Ithaca 5&10 put on by the Fingerlakes Runners Club
https://fingerlakesrunners.org/race/ithaca-5-10-2025/

twitter_events: Twitter, 5 events (2013-2014)
Various multiplex networks of retweets, mentions, and replies among Twitter users during specific events or occasions in 2013 and 2014.
This network has 327707 nodes and 396671 edges.
Tags: Social, Online, Multilayer, Unweighted
https://networks.skew…

🤡
https://mobil.osnabrueck.de/de/aktuelles/bedeutender-schritt-fuer-oepnv-in-stadt-und-landkreis-beteiligungsverfahren-zum-5-nahverkehrsplan-startet-in-kuerze/…

aktuelles #Wetter #Behrendorf #Wische 11-07-2025 15:35
Temperatur: (min: 10,7C) 24,6C (max: 25,2C)
Feuchte: 52%
Luftdruck: 1013,5 hPa (langsam fallend)
Windrichtung: 304 Grad
W…

Reporte de visita m\'edica sobre la aplicaci\'on de braquiterapia de baja tasa con Cs-137 en una paciente con c\'ancer de cuello uterino, estadio IIB
Miguel Angel Mencomo Cede\~no
https://arxiv.org/abs/2509.00306

Reporte de visita médica sobre la aplicación de braquiterapia de baja tasa con Cs-137 en una paciente con cáncer de cuello uterino, estadio IIB
During an academic visit to the National Oncology Institute (ION), the low-dose-rate brachytherapy procedure with Cs-137 was observed in a patient with stage IIB cervical cancer. The experience provided insight into both the clinical preparation of the patient and the dose planning process through imaging studies. Despite certain logistical limitations, strict adherence to safety protocols, disinfection procedures, and radioactive source handling was emphasized. The coordinated participation of…

Probing Nucleon-$\Omega_{\rm ccc}$Interaction via Lattice QCD at Physical Quark Masses
Liang Zhang, Takumi Doi, Yan Lyu, Tetsuo Hatsuda, Yu-Gang Ma
https://arxiv.org/abs/2508.10388

Probing Nucleon-$Ω_{\rm ccc}$Interaction via Lattice QCD at Physical Quark Masses
We study the S-wave interactions between the nucleon ($N$) and the triply charmed Omega baryon ($Ω_{\mathrm{ccc}}$) using (2+1)-flavor lattice QCD with a physical pion mass ($m_π\simeq 137.1$ MeV) on a lattice volume $\simeq (8.1~\mathrm{fm})^3$. The charm quark is implemented with a relativistic heavy-quark action at its physical mass. Employing the time-dependent HAL QCD method, the $N$-$Ω_{\mathrm{ccc}}$ potentials in the spin-1 ($^3\mathrm{S}_1$) and spin-2 ($^5\mathrm{S}_2$) channels ar…

Ihr erinnert Euch, dass #BadenWürttemberg im Sommer rund 1440 unbenutzte originalverpackte Lehrkräftestellen aufm Dachboden gefunden hat? In unserer Berichterstattung zum Schuljahresstart kam das Thema wieder vor (Überraschung!) und es kam eine Hörerfrage dazu, warum das denn niemand gemerkt hat im Haushalt. Ich hab versucht, es zu erklären: 1/5

Achtung Enkeltrick!
#merz

№ 23 Carter Meshanic (whose brother Hunter is also on the team) of the William Smith Senators tries to get the ball past № 14 Alex Penders 6' 5'' of the IC Bombers
#photo #photography #sports

Have I missed something or is there no way with openssh to exempt certain netblocks from the MaxStartups setting?
There's PerSourcePenaltyExemptList but that's spoecifically for the "penalties" things which are separate from MaxStartups, right?
https://

The NN$\Omega_{ccc}$ Bound-States with the HAL QCD interaction
Faisal Etminan
https://arxiv.org/abs/2508.12276 https://arxiv.org/pdf/2508.12276

The NN$Ω_{ccc}$ Bound-States with the HAL QCD interaction
Newly, first-principles lattice QCD results at the physical pion mass, $ m_π\backsimeq 137.1 $ MeV, have been reported by the HAL QCD Collaboration for the S-wave interaction between the nucleon (N) and the triply charmed Omega baryon ($Ω_{ccc}$). The N$Ω_{ccc}$ potentials in the spin-1 $ \left(^{3}S_{1}\right) $ and spin-2 $ \left(^{5}S_{2}\right) $ channels were derived and found to be attractive, though no bound state was supported in these channels. It is shown that, when the spin-1 and …

Wärmekarte für #Stuttgart.
https://www.meteoblue.com/de/products/cityclimate/heatmaps/stuttgart#13.5/48.78028/9.18017
Viel trinken,…

Wärmekarte für Stuttgart - meteoblue
Identifizierung städtischer Wärmeinseln mit hyperlokalen (10 m) Temperaturkarten für Städte weltweit ✓ Management von Hitzewellen ✓ Stadtplanung ✓ Lokale Klimazonen

Seeing is Believing: Emotion-Aware Audio-Visual Language Modeling for Expressive Speech Generation
Weiting Tan, Jiachen Lian, Hirofumi Inaguma, Paden Tomasello, Philipp Koehn, Xutai Ma
https://arxiv.org/abs/2508.16188

Seeing is Believing: Emotion-Aware Audio-Visual Language Modeling for Expressive Speech Generation
We present an Audio-Visual Language Model (AVLM) for expressive speech generation by integrating full-face visual cues into a pre-trained expressive speech model. We explore multiple visual encoders and multimodal fusion strategies during pre-training to identify the most effective integration approach. Subsequent fine-tuning on emotion recognition and expressive dialogue tasks yields substantial gains over speech-only baselines (e.g., +5 F1 in emotion recognition). AVLM highlights the value of…

Replaced article(s) found for cs.CL. https://arxiv.org/list/cs.CL/new
[2/3]:
- Annif at SemEval-2025 Task 5: Traditional XMTC augmented by LLMs
Osma Suominen, Juho Inkinen, Mona Lehtinen

HebID: Detecting Social Identities in Hebrew-language Political Text
Guy Mor-Lan, Naama Rivlin-Angert, Yael R. Kaplan, Tamir Sheafer, Shaul R. Shenhav
https://arxiv.org/abs/2508.15483

HebID: Detecting Social Identities in Hebrew-language Political Text
Political language is deeply intertwined with social identities. While social identities are often shaped by specific cultural contexts and expressed through particular uses of language, existing datasets for group and identity detection are predominantly English-centric, single-label and focus on coarse identity categories. We introduce HebID, the first multilabel Hebrew corpus for social identity detection: 5,536 sentences from Israeli politicians' Facebook posts (Dec 2018-Apr 2021), manually…

Filling the Gap for Uzbek: Creating Translation Resources for Southern Uzbek
Mukhammadsaid Mamasaidov, Azizullah Aral, Abror Shopulatov, Mironshoh Inomjonov
https://arxiv.org/abs/2508.14586

Filling the Gap for Uzbek: Creating Translation Resources for Southern Uzbek
Southern Uzbek (uzs) is a Turkic language variety spoken by around 5 million people in Afghanistan and differs significantly from Northern Uzbek (uzn) in phonology, lexicon, and orthography. Despite the large number of speakers, Southern Uzbek is underrepresented in natural language processing. We present new resources for Southern Uzbek machine translation, including a 997-sentence FLORES+ dev set, 39,994 parallel sentences from dictionary, literary, and web sources, and a fine-tuned NLLB-200 …

Long Chain-of-Thought Reasoning Across Languages
Josh Barua, Seun Eisape, Kayo Yin, Alane Suhr
https://arxiv.org/abs/2508.14828 https://arxiv.org/pdf/2508.…

Long Chain-of-Thought Reasoning Across Languages
Scaling inference through long chains-of-thought (CoTs) has unlocked impressive reasoning capabilities in large language models (LLMs), yet the reasoning process remains almost exclusively English-centric. We construct translated versions of two popular English reasoning datasets, fine-tune Qwen 2.5 (7B) and Qwen 3 (8B) models, and present a systematic study of long CoT generation across French, Japanese, Latvian, and Swahili. Our experiments reveal three key findings. First, the efficacy of us…

#Stuttgart21, oder wie man einen Bahnhof unter die Erde verfrachtet, um (unter anderem) obendrauf Bauflächen zu haben, wie dann aber das #Eisenbahngesetz geändert wurde und die schönen Bauflächen plötzlich nicht mehr hätten bebaut werden dürfen, und wie das Eisenbahngesetz jetzt wieder g…

Bundestag beschließt Eisenbahngesetz-Änderung: 5.700 Wohnungen in Stuttgart könnten entstehen
Die Bundesregierung hat in der Nacht eine Reform des Eisenbahngesetzes verabschiedet. Damit könnten Wohnprojekte leichter auf alten Bahnflächen gebaut werden - wie in Stuttgart.

My macOS laptop has felt... less than responsive of late.
I'm thinking about hitting the restore-to-factory-defaults type button and using Time Machine to bring stuff back.
I haven't done this procedure in probably 5-10 years.
Any hiccups or pitfalls I should know about?
Running Sequoia 15.6.1
#macOS #Apple #TimeMachine #maintenance

The Digital Sous Chef -- A Comparative Study on Fine-Tuning Language Models for Recipe Generation
Shubham Pundhir, Ganesh Bagler
https://arxiv.org/abs/2508.14718 https://…

The Digital Sous Chef -- A Comparative Study on Fine-Tuning Language Models for Recipe Generation
We established a rigorous benchmark for text-based recipe generation, a fundamental task in natural language generation. We present a comprehensive comparative study contrasting a fine-tuned GPT-2 large (774M) model against the GPT-2 small (124M) model and traditional LSTM/RNN baselines on the 5-cuisine corpus from RecipeDB. Our key contribution is a targeted tokenization strategy that augments the vocabulary with 23 common fraction tokens and custom structural markers. This approach addresses …

Datarus-R1: An Adaptive Multi-Step Reasoning LLM for Automated Data Analysis
Ayoub Ben Chaliah, Hela Dellagi
https://arxiv.org/abs/2508.13382 https://arxiv…

Datarus-R1: An Adaptive Multi-Step Reasoning LLM for Automated Data Analysis
We present Datarus-R1-14B, a 14 B-parameter open-weights language model fine-tuned from Qwen 2.5-14B-Instruct to act as a virtual data analyst and graduate-level problem solver. Datarus is trained not on isolated question-answer pairs but on full analytical trajectories including reasoning steps, code execution, error traces, self-corrections, and final conclusions, all captured in a ReAct-style notebook format spanning finance, medicine, numerical analysis, and other quantitative domains. Our …

Saudi-Dialect-ALLaM: LoRA Fine-Tuning for Dialectal Arabic Generation
Hassan Barmandah
https://arxiv.org/abs/2508.13525 https://arxiv.org/pdf/2508.13525

Saudi-Dialect-ALLaM: LoRA Fine-Tuning for Dialectal Arabic Generation
Large language models (LLMs) for Arabic are still dominated by Modern Standard Arabic (MSA), with limited support for Saudi dialects such as Najdi and Hijazi. This underrepresentation hinders their ability to capture authentic dialectal variation. Using a privately curated Saudi Dialect Instruction dataset (Hijazi and Najdi; 5,466 synthetic instruction-response pairs; 50/50 split), we LoRA-tune ALLaM-7B-Instruct-preview, the first foundation model developed in Saudi Arabia, for Saudi dialect ge…

Präsenz im Funkhaus
#shoesandfloorchallenge 5/7