Tootfinder

A Taxonomy of Prompt Defects in LLM Systems
Haoye Tian, Chong Wang, BoYang Yang, Lyuye Zhang, Yang Liu
https://arxiv.org/abs/2509.14404 https://arxiv.org/p…

A Taxonomy of Prompt Defects in LLM Systems
Large Language Models (LLMs) have become key components of modern software, with prompts acting as their de-facto programming interface. However, prompt design remains largely empirical and small mistakes can cascade into unreliable, insecure, or inefficient behavior. This paper presents the first systematic survey and taxonomy of prompt defects, recurring ways that prompts fail to elicit their intended behavior from LLMs. We organize defects along six dimensions: (1) Specification and Intent, …

Too Easily Fooled? Prompt Injection Breaks LLMs on Frustratingly Simple Multiple-Choice Questions
Xuyang Guo, Zekai Huang, Zhao Song, Jiahao Zhang
https://arxiv.org/abs/2508.13214

Too Easily Fooled? Prompt Injection Breaks LLMs on Frustratingly Simple Multiple-Choice Questions
Large Language Models (LLMs) have recently demonstrated strong emergent abilities in complex reasoning and zero-shot generalization, showing unprecedented potential for LLM-as-a-judge applications in education, peer review, and data quality evaluation. However, their robustness under prompt injection attacks, where malicious instructions are embedded into the content to manipulate outputs, remains a significant concern. In this work, we explore a frustratingly simple yet effective attack settin…

Google Photos will let users edit images using text or voice prompts, first on Pixel 10 devices in the US, and will add support for C2PA Content Credentials (Sarah Perez/TechCrunch)
https://techcrunch.com/2025/08/20/you-can-now-talk-to-google-photos-to…

You can now talk to Google Photos to make your edits | TechCrunch
Google Photos introduces an "ask to edit" feature that lets you speak or text your edit requests.

Calibration-Aware Prompt Learning for Medical Vision-Language Models
Abhishek Basu, Fahad Shamshad, Ashshak Sharifdeen, Karthik Nandakumar, Muhammad Haris Khan
https://arxiv.org/abs/2509.15226

Calibration-Aware Prompt Learning for Medical Vision-Language Models
Medical Vision-Language Models (Med-VLMs) have demonstrated remarkable performance across diverse medical imaging tasks by leveraging large-scale image-text pretraining. However, their confidence calibration is largely unexplored, and so remains a significant challenge. As such, miscalibrated predictions can lead to overconfident errors, undermining clinical trust and decision-making reliability. To address this, we introduce CalibPrompt, the first framework to calibrate Med-VLMs during prompt …

AI anatomy is fun. 😂
https://masto.ai/@vagina_museum/115231091370485890

Vagina Museum (@vagina_museum@masto.ai)
Attached: 1 image First, we wanted to test a very basic prompt, likely to be similar to something the original poster would have said. "Show me a diagram of the organs during pregnancy" returned this.

Something appears to have changed at ChatGPT to make deep research better. I ran it yesterday and one search took three hours! It also prompted me with a series of multiple questions first, not just the usual four.

Mass killings
Was looking through Wikipedia's list of mass killings in America (#guns #GunViolence #Shooting

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 …

okay, now I get it when people complain about #gemini free..... 😧
First prompt of the day with this file added as context: https://github.com/openai-php/client/blob/main/README.md…

TextMine: LLM-Powered Knowledge Extraction for Humanitarian Mine Action
Chenyue Zhou, G\"urkan Solmaz, Flavio Cirillo, Kiril Gashteovski, Jonathan F\"urst
https://arxiv.org/abs/2509.15098

TextMine: LLM-Powered Knowledge Extraction for Humanitarian Mine Action
Humanitarian Mine Action has generated extensive best-practice knowledge, but much remains locked in unstructured reports. We introduce TextMine, an ontology-guided pipeline that uses Large Language Models to extract knowledge triples from HMA texts. TextMine integrates document chunking, domain-aware prompting, triple extraction, and both reference-based and LLM-as-a-Judge evaluation. We also create the first HMA ontology and a curated dataset of real-world demining reports. Experiments show o…