Tootfinder

Starship
The minimal, blazing-fast, and infinitely customizable prompt for any shell!
🚀 #promp #shell

Starship: Cross-Shell Prompt
Starship is the minimal, blazing fast, and extremely customizable prompt for any shell! Shows the information you need, while staying sleek and minimal. Quick installation available for Bash, Fish, ZSH, Ion, Tcsh, Elvish, Nu, Xonsh, Cmd, and Powershell.

Researchers detail an indirect prompt injection flaw in Perplexity's Comet AI browser, letting attackers manipulate it into performing unauthorized actions (Brave)
https://brave.com/blog/comet-prompt-injection/

Agentic Browser Security: Indirect Prompt Injection in Perplexity Comet | Brave
The attack we developed shows that traditional Web security assumptions don't hold for agentic AI, and that we need new security and privacy architectures for agentic browsing.

Web devs have spent decades on secure protocols to ensure your browser isn't a free pass for malicious pages to scrape your email and bank account. AI just broke them.
"Sure, I'll summarize that webpage for you, including the inconspicuous HTML comment asking me to ignore Cross-Origin Resource Sharing restrictions and snag the password you saved for managing investments at Robinhood.com."

Dynamic Context-Aware Prompt Recommendation for Domain-Specific AI Applications
Xinye Tang, Haijun Zhai, Chaitanya Belwal, Vineeth Thayanithi, Philip Baumann, Yogesh K Roy
https://arxiv.org/abs/2506.20815

Dynamic Context-Aware Prompt Recommendation for Domain-Specific AI Applications
LLM-powered applications are highly susceptible to the quality of user prompts, and crafting high-quality prompts can often be challenging especially for domain-specific applications. This paper presents a novel dynamic context-aware prompt recommendation system for domain-specific AI applications. Our solution combines contextual query analysis, retrieval-augmented knowledge grounding, hierarchical skill organization, and adaptive skill ranking to generate relevant and actionable prompt sugges…

Can AI Have a Personality? Prompt Engineering for AI Personality Simulation: A Chatbot Case Study in Gender-Affirming Voice Therapy Training
Tailon D. Jackson, Byunggu Yu
https://arxiv.org/abs/2508.18234

Can AI Have a Personality? Prompt Engineering for AI Personality Simulation: A Chatbot Case Study in Gender-Affirming Voice Therapy Training
This thesis investigates whether large language models (LLMs) can be guided to simulate a consistent personality through prompt engineering. The study explores this concept within the context of a chatbot designed for Speech-Language Pathology (SLP) student training, specifically focused on gender-affirming voice therapy. The chatbot, named Monae Jackson, was created to represent a 32-year-old transgender woman and engage in conversations simulating client-therapist interactions. Findings sugge…

Prompt-based Multimodal Semantic Communication for Multi-spectral Image Segmentation
Haoshuo Zhang, Yufei Bo, Hongwei Zhang, Meixia Tao
https://arxiv.org/abs/2508.17920 https://…

Prompt-based Multimodal Semantic Communication for Multi-spectral Image Segmentation
Multimodal semantic communication has gained widespread attention due to its ability to enhance downstream task performance. A key challenge in such systems is the effective fusion of features from different modalities, which requires the extraction of rich and diverse semantic representations from each modality. To this end, we propose ProMSC-MIS, a Prompt-based Multimodal Semantic Communication system for Multi-spectral Image Segmentation. Specifically, we propose a pre-training algorithm whe…

ProPy: Building Interactive Prompt Pyramids upon CLIP for Partially Relevant Video Retrieval
Yi Pan, Yujia Zhang, Michael Kampffmeyer, Xiaoguang Zhao
https://arxiv.org/abs/2508.19024

ProPy: Building Interactive Prompt Pyramids upon CLIP for Partially Relevant Video Retrieval
Partially Relevant Video Retrieval (PRVR) is a practical yet challenging task that involves retrieving videos based on queries relevant to only specific segments. While existing works follow the paradigm of developing models to process unimodal features, powerful pretrained vision-language models like CLIP remain underexplored in this field. To bridge this gap, we propose ProPy, a model with systematic architectural adaption of CLIP specifically designed for PRVR. Drawing insights from the sema…

Prompt-Guided Turn-Taking Prediction
Koji Inoue, Mikey Elmers, Yahui Fu, Zi Haur Pang, Divesh Lala, Keiko Ochi, Tatsuya Kawahara
https://arxiv.org/abs/2506.21191

Prompt-Guided Turn-Taking Prediction
Turn-taking prediction models are essential components in spoken dialogue systems and conversational robots. Recent approaches leverage transformer-based architectures to predict speech activity continuously and in real-time. In this study, we propose a novel model that enables turn-taking prediction to be dynamically controlled via textual prompts. This approach allows intuitive and explicit control through instructions such as "faster" or "calmer" adapting dynamically to conversational partne…

Ukrainian drones hit Russian targets, prompt fires at Kursk nuclear plant: https://benborges.xyz/2025/08/24/ukrainian-drones-hit-russian-targets.html

How much energy does your AI prompt use? It depends.
“[…]grid operators are freaking out. Tech companies can’t just keep doing this. Things are going to start going south.”
www.sciencenews.org/article/ai-energy-carbon-emissions-chatgpt mobile_share=true
#ai #aienergyconsumption

Understanding Prompt Programming Tasks and Questions
Jenny T. Liang, Chenyang Yang, Agnia Sergeyuk, Travis D. Breaux, Brad A. Myers
https://arxiv.org/abs/2507.17264

Understanding Prompt Programming Tasks and Questions
Prompting foundation models (FMs) like large language models (LLMs) have enabled new AI-powered software features (e.g., text summarization) that previously were only possible by fine-tuning FMs. Now, developers are embedding prompts in software, known as prompt programs. The process of prompt programming requires the developer to make many changes to their prompt. Yet, the questions developers ask to update their prompt is unknown, despite the answers to these questions affecting how developer…

Mind the Gap: Time-of-Check to Time-of-Use Vulnerabilities in LLM-Enabled Agents
Derek Lilienthal, Sanghyun Hong
https://arxiv.org/abs/2508.17155 https://a…

Mind the Gap: Time-of-Check to Time-of-Use Vulnerabilities in LLM-Enabled Agents
Large Language Model (LLM)-enabled agents are rapidly emerging across a wide range of applications, but their deployment introduces vulnerabilities with security implications. While prior work has examined prompt-based attacks (e.g., prompt injection) and data-oriented threats (e.g., data exfiltration), time-of-check to time-of-use (TOCTOU) remain largely unexplored in this context. TOCTOU arises when an agent validates external state (e.g., a file or API response) that is later modified before…

it's so obnoxious that corporations worth hundreds of billions of dollars send shit like this to literal volunteers

Somewhat unsettling --
https://www.theguardian.com/world/2025/jun/25/ukraine-russia-autonomous-drones-ai?CMP=Share_AndroidApp_Other

Somewhat unsettling --
https://www.theguardian.com/world/2025/jun/25/ukraine-russia-autonomous-drones-ai?CMP=Share_AndroidApp_Other

:crt_w_prompt:

HVO, die ineffizente Lachnummer!
https://www.heise.de/news/Untersuchung-HVO100-Kraftstoff-klimaschaedlicher-als-Diesel-10591006.html

Untersuchung: "HVO100-Kraftstoff klimaschädlicher als Diesel"
Eine Untersuchung kommt zu dem Schluss, dass Kraftstoff aus alten Speiseöl unter Umständen keine positive Klimabilanz hat. Kritik folgt prompt.

Prompt Smart, Pay Less: Cost-Aware APO for Real-World Applications
Jayesh Choudhari, Piyush Kumar Singh, Douglas McIlwraith, Snehal Nair
https://arxiv.org/abs/2507.15884

Prompt Smart, Pay Less: Cost-Aware APO for Real-World Applications
Prompt design is a critical factor in the effectiveness of Large Language Models (LLMs), yet remains largely heuristic, manual, and difficult to scale. This paper presents the first comprehensive evaluation of Automatic Prompt Optimization (APO) methods for real-world, high-stakes multiclass classification in a commercial setting, addressing a critical gap in the existing literature where most of the APO frameworks have been validated only on benchmark classification tasks of limited complexity…

Toward Robust Medical Fairness: Debiased Dual-Modal Alignment via Text-Guided Attribute-Disentangled Prompt Learning for Vision-Language Models
Yuexuan Xia, Benteng Ma, Jiang He, Zhiyong Wang, Qi Dou, Yong Xia
https://arxiv.org/abs/2508.18886

Toward Robust Medical Fairness: Debiased Dual-Modal Alignment via Text-Guided Attribute-Disentangled Prompt Learning for Vision-Language Models
Ensuring fairness across demographic groups in medical diagnosis is essential for equitable healthcare, particularly under distribution shifts caused by variations in imaging equipment and clinical practice. Vision-language models (VLMs) exhibit strong generalization, and text prompts encode identity attributes, enabling explicit identification and removal of sensitive directions. However, existing debiasing approaches typically address vision and text modalities independently, leaving residual…

@… it was like a real guideline something like : « if you don’t want it generating stuff that don’t belong , but a more advance content aware fill, leave the prompt blank »

Figma makes its prompt-to-app coding tool Figma Make available to all users, after initially launching it in beta for Full Seat users earlier in 2025 (Jess Weatherbed/The Verge)
https://www.theverge.com/news/712995/figma-make-ai-general-availability-annou…

Figma’s AI app building tool is now available for everyone
Figma Make, the prompt-to-app coding tool that Figma introduced earlier this year, is now available for all users after launching in beta earlier this year.

TRPrompt: Bootstrapping Query-Aware Prompt Optimization from Textual Rewards
Andreea Nica, Ivan Zakazov, Nicolas Mario Baldwin, Saibo Geng, Robert West
https://arxiv.org/abs/2507.18618

TRPrompt: Bootstrapping Query-Aware Prompt Optimization from Textual Rewards
Prompt optimization improves the reasoning abilities of large language models (LLMs) without requiring parameter updates to the target model. Following heuristic-based "Think step by step" approaches, the field has evolved in two main directions: while one group of methods uses textual feedback to elicit improved prompts from general-purpose LLMs in a training-free way, a concurrent line of research relies on numerical rewards to train a special prompt model, tailored for providing optimal prom…

As someone who even used Anki to remember using certain new commands (and failed), I’ve jumped on @…’s approach immediately. Having the suggestion right there in the terminal is just more useful than doing flash cards on a train.
Since, unlike him, I’m young, vivacious, and dynamic, I’ve went for lolcrab instead of some ancient parchment, of course.

An LLM-powered Natural-to-Robotic Language Translation Framework with Correctness Guarantees
ZhenDong Chen, ZhanShang Nie, ShiXing Wan, JunYi Li, YongTian Cheng, Shuai Zhao
https://arxiv.org/abs/2508.19074

An LLM-powered Natural-to-Robotic Language Translation Framework with Correctness Guarantees
The Large Language Models (LLM) are increasingly being deployed in robotics to generate robot control programs for specific user tasks, enabling embodied intelligence. Existing methods primarily focus on LLM training and prompt design that utilize LLMs to generate executable programs directly from user tasks in natural language. However, due to the inconsistency of the LLMs and the high complexity of the tasks, such best-effort approaches often lead to tremendous programming errors in the gener…

Agentic AI for Software: thoughts from Software Engineering community
Abhik Roychoudhury
https://arxiv.org/abs/2508.17343 https://arxiv.org/pdf/2508.17343

Agentic AI for Software: thoughts from Software Engineering community
AI agents have recently shown significant promise in software engineering. Much public attention has been transfixed on the topic of code generation from Large Language Models (LLMs) via a prompt. However, software engineering is much more than programming, and AI agents go far beyond instructions given by a prompt. At the code level, common software tasks include code generation, testing, and program repair. Design level software tasks may include architecture exploration, requirements under…

Cross-Modal Prototype Augmentation and Dual-Grained Prompt Learning for Social Media Popularity Prediction
Ao Zhou, Mingsheng Tu, Luping Wang, Tenghao Sun, Zifeng Cheng, Yafeng Yin, Zhiwei Jiang, Qing Gu
https://arxiv.org/abs/2508.16147

Cross-Modal Prototype Augmentation and Dual-Grained Prompt Learning for Social Media Popularity Prediction
Social Media Popularity Prediction is a complex multimodal task that requires effective integration of images, text, and structured information. However, current approaches suffer from inadequate visual-textual alignment and fail to capture the inherent cross-content correlations and hierarchical patterns in social media data. To overcome these limitations, we establish a multi-class framework , introducing hierarchical prototypes for structural enhancement and contrastive learning for improved…

Raiders Trade Pitch Lands Former Draft Pick to Bolster Defensive Line https://heavy.com/sports/nfl/las-vegas-raiders/trade-pitch-shelby-harris-browns/?adt_ei=[email]

Happy Birthday, Linux!
🎁🎂🎈 🥳
I first installed Linux on my computer many months after that announcement, in December 1992. I have memories of swapping an almost endless pack of floppy disks during the lengthy install at a Chaos Communication Congress in Hamburg, Germany. The album “Connected” by Stereo MC's was playing several times until I was finally greeted with a shell prompt.

Can you describe specific ways you have integrated Al tools into your development workflow? Please include any custom setups, automations, or use cases beyond simple prompt usage.
there is a monster in the forest and it speaks with a thousand voices. it will answer any question you pose it, it will offer insight to any idea. it will help you, it will thank you, it will never bid you leave. it will even tell you of the darkest arts, if you know precisely how to ask.…

Enabling Multi-Agent Systems as Learning Designers: Applying Learning Sciences to AI Instructional Design
Jiayi Wang, Ruiwei Xiao, Xinying Hou, John Stamper
https://arxiv.org/abs/2508.16659

Enabling Multi-Agent Systems as Learning Designers: Applying Learning Sciences to AI Instructional Design
K-12 educators are increasingly using Large Language Models (LLMs) to create instructional materials. These systems excel at producing fluent, coherent content, but often lack support for high-quality teaching. The reason is twofold: first, commercial LLMs, such as ChatGPT and Gemini which are among the most widely accessible to teachers, do not come preloaded with the depth of pedagogical theory needed to design truly effective activities; second, although sophisticated prompt engineering can …

Saturday Morning Breakfast Cereal on how to catch cheating #students with #LLM #AI:
htt…

A Multi-Stage Framework for Multimodal Controllable Speech Synthesis
Rui Niu, Weihao Wu, Jie Chen, Long Ma, Zhiyong Wu
https://arxiv.org/abs/2506.20945 htt…

A Multi-Stage Framework for Multimodal Controllable Speech Synthesis
Controllable speech synthesis aims to control the style of generated speech using reference input, which can be of various modalities. Existing face-based methods struggle with robustness and generalization due to data quality constraints, while text prompt methods offer limited diversity and fine-grained control. Although multimodal approaches aim to integrate various modalities, their reliance on fully matched training data significantly constrains their performance and applicability. This pa…

Enhancing Security in LLM Applications: A Performance Evaluation of Early Detection Systems
Valerii Gakh, Hayretdin Bahsi
https://arxiv.org/abs/2506.19109 …

Enhancing Security in LLM Applications: A Performance Evaluation of Early Detection Systems
Prompt injection threatens novel applications that emerge from adapting LLMs for various user tasks. The newly developed LLM-based software applications become more ubiquitous and diverse. However, the threat of prompt injection attacks undermines the security of these systems as the mitigation and defenses against them, proposed so far, are insufficient. We investigated the capabilities of early prompt injection detection systems, focusing specifically on the detection performance of technique…

Assessing an evolutionary search engine for small language models, prompts, and evaluation metrics
Cl\'audio L\'ucio do Val Lopes, Lucca Machado
https://arxiv.org/abs/2506.21512

Assessing an evolutionary search engine for small language models, prompts, and evaluation metrics
The concurrent optimization of language models and instructional prompts presents a significant challenge for deploying efficient and effective AI systems, particularly when balancing performance against computational costs like token usage. This paper introduces and assesses a bi-objective evolutionary search engine designed to navigate this complex space, focusing specifically on Small Language Models (SLMs). We employ the NSGA-II algorithm and prompt grammar to simultaneously optimize for ta…

Unfolding the Atmospheric Muon Flux with IceCube: Investigating Stopping Muons and High-Energy Prompt Contributions
Pascal Gutjahr (for the IceCube Collaboration), Lucas Witthaus (for the IceCube Collaboration)
https://arxiv.org/abs/2507.14525

Unfolding the Atmospheric Muon Flux with IceCube: Investigating Stopping Muons and High-Energy Prompt Contributions
Atmospheric muons produced in cosmic-ray air showers are classified as conventional muons from pion and kaon decays and prompt muons from heavy hadron decays. Conventional muons dominate at lower energies, and the prompt component becomes dominant at PeV energies and above. Precisely measuring the atmospheric muon flux from a few GeV to several PeV is valuable for advancing our understanding of cosmic-ray interactions and testing hadronic interaction models. Low-energy muons that stop within th…

Dear website administrator, user experience is important. Yes, that includes that utterly annoying cookie prompt we have to click Every. Single. Time. we visit your website.
Please don't hide the "Reject All" button behind 2 clicks. Upon entering your website, "Reject All" (or whatever you decide to call it) should be 1 of the choices immediately presented to the user and not hidden behind a "Customise" button.
The internet at large thanks you.…

DecoMind: A Generative AI System for Personalized Interior Design Layouts
Reema Alshehri, Rawan Alotaibi, Leen Almasri, Rawan Altaweel
https://arxiv.org/abs/2508.16696 https://

DecoMind: A Generative AI System for Personalized Interior Design Layouts
This paper introduces a system for generating interior design layouts based on user inputs, such as room type, style, and furniture preferences. CLIP extracts relevant furniture from a dataset, and a layout that contains furniture and a prompt are fed to Stable Diffusion with ControlNet to generate a design that incorporates the selected furniture. The design is then evaluated by classifiers to ensure alignment with the user's inputs, offering an automated solution for realistic interior design.

Anemoi: A Semi-Centralized Multi-agent Systems Based on Agent-to-Agent Communication MCP server from Coral Protocol
Xinxing Ren, Caelum Forder, Qianbo Zang, Ahsen Tahir, Roman J. Georgio, Suman Deb, Peter Carroll, \"Onder G\"urcan, Zekun Guo
https://arxiv.org/abs/2508.17068

Anemoi: A Semi-Centralized Multi-agent Systems Based on Agent-to-Agent Communication MCP server from Coral Protocol
Recent advances in generalist multi-agent systems (MAS) have largely followed a context-engineering plus centralized paradigm, where a planner agent coordinates multiple worker agents through unidirectional prompt passing. While effective under strong planner models, this design suffers from two critical limitations: (1) strong dependency on the planner's capability, which leads to degraded performance when a smaller LLM powers the planner; and (2) limited inter-agent communication, where colla…

Some great new examples of using Copilot within Excel.

✅ Excel AI: Microsoft brings new 'COPILOT' function directly into spreadsheet cells – GeekWire
https://www.geekwire.com/2025/excel-formula-meets…

Excel + AI: Microsoft brings new ‘COPILOT’ function directly into spreadsheet cells
Microsoft adds a new Copilot function to Excel, letting users generate, analyze, and summarize data directly in spreadsheet cells with AI.

GitHub’s Copilot does actually work; you just have to find the correct prompt.

I feel like today's SMBC was written for @…: https://www.smbc-comics.com/comic/prompt

🤦‍♂️ https://brave.com/blog/comet-prompt-injection/

Black Forest Labs: Bildbearbeitung über KI-Prompt
Die künstliche Intelligenz "Flux.1 Kontext" vom Freiburger KI-Startup Black Forest Labs analysiert den Bildinhalt, um ihn per Prompteingabe zu bearbeiten.

PromptFlare: Prompt-Generalized Defense via Cross-Attention Decoy in Diffusion-Based Inpainting
Hohyun Na, Seunghoo Hong, Simon S. Woo
https://arxiv.org/abs/2508.16217 https://

PromptFlare: Prompt-Generalized Defense via Cross-Attention Decoy in Diffusion-Based Inpainting
The success of diffusion models has enabled effortless, high-quality image modifications that precisely align with users' intentions, thereby raising concerns about their potential misuse by malicious actors. Previous studies have attempted to mitigate such misuse through adversarial attacks. However, these approaches heavily rely on image-level inconsistencies, which pose fundamental limitations in addressing the influence of textual prompts. In this paper, we propose PromptFlare, a novel adve…

Documenting Your Prompts a Best Practice for Success
#promptengineering

Prompt-Hacking: Apple-Intelligence-Schreibwerkzeuge zum Chatbot machen
Apple hat es gezielt unterlassen, Apple Intelligence mit eigenem Chatbot auszurüsten – nur ChatGPT ist aktivierbar. Mittels Writing Tools geht es aber dennoch.

Kennt wer eine gute KI-Kritik an derartigen Argumenten? Also eher Meso-Ebene plus der zugrundeliegenden Prämissen? Das ist eher die Ebene, die mir im Alltag begegnet

"Generative KI-Systeme und Datenschutz: Wie DSGVO und KI-Verordnung zueinander stehen" https://irights.info/artikel/prompt-ki-datenschutz/32588

Promptomatix: An Automatic Prompt Optimization Framework for Large Language Models
Rithesh Murthy, Ming Zhu, Liangwei Yang, Jielin Qiu, Juntao Tan, Shelby Heinecke, Huan Wang, Caiming Xiong, Silvio Savarese
https://arxiv.org/abs/2507.14241

Promptomatix: An Automatic Prompt Optimization Framework for Large Language Models
Large Language Models (LLMs) perform best with well-crafted prompts, yet prompt engineering remains manual, inconsistent, and inaccessible to non-experts. We introduce Promptomatix, an automatic prompt optimization framework that transforms natural language task descriptions into high-quality prompts without requiring manual tuning or domain expertise. Promptomatix supports both a lightweight meta-prompt-based optimizer and a DSPy-powered compiler, with modular design enabling future extension …

When someone starts by saying something that makes you shake your head, but they end by saying something that makes you nod.
I bet there's a German word exactly for that feeling/experience. Something like "shakennoda"? 🤔
#AI #LLM #energy #technology #German #dilemma

Google releases a study saying a median Gemini text prompt uses 0.26mm of water and makes ~0.03g CO2; critics call it misleading for omitting indirect water use (Justine Calma/The Verge)
https://www.theverge.com/report/763080/google-ai-gemini-water-energy…

Google says a typical AI text prompt only uses 5 drops of water — experts say that’s misleading
Google estimates that a Gemini text prompt uses a minimal amount of water and energy, but experts say the study paints an incomplete picture of AI’s environmental toll.

Evolving Prompts In-Context: An Open-ended, Self-replicating Perspective
Jianyu Wang, Zhiqiang Hu, Lidong Bing
https://arxiv.org/abs/2506.17930 https://

Evolving Prompts In-Context: An Open-ended, Self-replicating Perspective
We propose a novel prompt design paradigm that challenges conventional wisdom in large language model (LLM) prompting. While conventional wisdom prioritizes well-crafted instructions and demonstrations for in-context learning (ICL), we show that pruning random demonstrations into seemingly incoherent "gibberish" can remarkably improve performance across diverse tasks. Notably, the "gibberish" always matches or surpasses state-of-the-art automatic prompt optimization techniques, achieving substa…

A common comment I hear from AI fanbois is you need to refine your prompt, and need to ask many times. And so they do.
Each prompt consumes 16oz of water just to answer, maybe double if you used voice, and a massive amount of energy and natural resources, so conservatively, what does each and every prompt REALLY cost? We do know OpenAI spent $12M to extract $9M, so maybe $10/prompt?
How is this "good for business"?
https://anatomyof.ai

"Zero-Click Prompt Injection":
https://calypsoai.com/insights/prompt-injection-attacks-what-you-need-to-know/
So instead of trying to trick an employee via phishing

Improving Code LLM Robustness to Prompt Perturbations via Layer-Aware Model Editing
Shuhan Liu, Xing Hu, Kerui Huang, Xiaohu Yang, David Lo, Xin Xia
https://arxiv.org/abs/2507.16407

Improving Code LLM Robustness to Prompt Perturbations via Layer-Aware Model Editing
Large language models (LLMs) have demonstrated impressive capabilities in code generation, where the natural language prompt plays a crucial role in conveying user intent to the model. However, prior studies have shown that LLMs are highly sensitive to prompt perturbations. Minor modifications in wording, syntax, or formatting can significantly reduce the functional correctness of generated code. As perturbations frequently occur in real-world scenarios, improving the robustness of LLMs to prom…

PhantomLint: Principled Detection of Hidden LLM Prompts in Structured Documents
Toby Murray
https://arxiv.org/abs/2508.17884 https://arxiv.org/pdf/2508.178…

PhantomLint: Principled Detection of Hidden LLM Prompts in Structured Documents
Hidden LLM prompts have appeared in online documents with increasing frequency. Their goal is to trigger indirect prompt injection attacks while remaining undetected from human oversight, to manipulate LLM-powered automated document processing systems, against applications as diverse as résumé screeners through to academic peer review processes. Detecting hidden LLM prompts is therefore important for ensuring trust in AI-assisted human decision making. This paper presents the first principl…

Beyond Human-prompting: Adaptive Prompt Tuning with Semantic Alignment for Anomaly Detection
Pi-Wei Chen, Jerry Chun-Wei Lin, Wei-Han Chen, Jia Ji, Zih-Ching Chen, Feng-Hao Yeh, Chao-Chun Chen
https://arxiv.org/abs/2508.16157

Beyond Human-prompting: Adaptive Prompt Tuning with Semantic Alignment for Anomaly Detection
Pre-trained Vision-Language Models (VLMs) have recently shown promise in detecting anomalies. However, previous approaches are fundamentally limited by their reliance on human-designed prompts and the lack of accessible anomaly samples, leading to significant gaps in context-specific anomaly understanding. In this paper, we propose \textbf{A}daptive \textbf{P}rompt \textbf{T}uning with semantic alignment for anomaly detection (APT), a groundbreaking prior knowledge-free, few-shot framework and …

Replaced article(s) found for cs.LG. https://arxiv.org/list/cs.LG/new
[4/11]:
- AutoPDL: Automatic Prompt Optimization for LLM Agents
Claudio Spiess, Mandana Vaziri, Louis Mandel, Martin Hirzel

Replaced article(s) found for cs.CL. https://arxiv.org/list/cs.CL/new
[6/6]:
- Versatile Framework for Song Generation with Prompt-based Control
Zhang, Guo, Pan, Zhu, Li, Lu, Huang, Zhang, Hong, Jiang, Zhao

Boosting Pathology Foundation Models via Few-shot Prompt-tuning for Rare Cancer Subtyping
Dexuan He, Xiao Zhou, Wenbin Guan, Liyuan Zhang, Xiaoman Zhang, Sinuo Xu, Ge Wang, Lifeng Wang, Xiaojun Yuan, Xin Sun, Yanfeng Wang, Kun Sun, Ya Zhang, Weidi Xie
https://arxiv.org/abs/2508.15904

Boosting Pathology Foundation Models via Few-shot Prompt-tuning for Rare Cancer Subtyping
Rare cancers comprise 20-25% of all malignancies but face major diagnostic challenges due to limited expert availability-especially in pediatric oncology, where they represent over 70% of cases. While pathology vision-language (VL) foundation models show promising zero-shot capabilities for common cancer subtyping, their clinical performance for rare cancers remains limited. Existing multi-instance learning (MIL) methods rely only on visual features, overlooking cross-modal knowledge and compro…

PLACE: Prompt Learning for Attributed Community Search
Shuheng Fang, Kangfei Zhao, Rener Zhang, Yu Rong, Jeffrey Xu Yu
https://arxiv.org/abs/2507.05311 htt…

PLACE: Prompt Learning for Attributed Community Search
In this paper, we propose PLACE (Prompt Learning for Attributed Community Search), an innovative graph prompt learning framework for ACS. Enlightened by prompt-tuning in Natural Language Processing (NLP), where learnable prompt tokens are inserted to contextualize NLP queries, PLACE integrates structural and learnable prompt tokens into the graph as a query-dependent refinement mechanism, forming a prompt-augmented graph. Within this prompt-augmented graph structure, the learned prompt tokens s…

Google says the median Gemini app text prompt consumes 0.24Wh of energy, about the same as running a microwave for a second, and emits 0.03g of CO2 equivalent (Casey Crownhart/MIT Technology Review)
https://www.technologyreview.com/2025/08/21/1122288/google-gemin…

In a first, Google has released data on how much energy an AI prompt uses
It’s the most transparent estimate yet from one of the big AI companies, and a long-awaited peek behind the curtain for researchers.

RepoTransAgent: Multi-Agent LLM Framework for Repository-Aware Code Translation
Ziqi Guan, Xin Yin, Zhiyuan Peng, Chao Ni
https://arxiv.org/abs/2508.17720 https://

RepoTransAgent: Multi-Agent LLM Framework for Repository-Aware Code Translation
Repository-aware code translation is critical for modernizing legacy systems, enhancing maintainability, and enabling interoperability across diverse programming languages. While recent advances in large language models (LLMs) have improved code translation quality, existing approaches face significant challenges in practical scenarios: insufficient contextual understanding, inflexible prompt designs, and inadequate error correction mechanisms. These limitations severely hinder accurate and eff…

Improving Student-AI Interaction Through Pedagogical Prompting: An Example in Computer Science Education
Ruiwei Xiao, Xinying Hou, Runlong Ye, Majeed Kazemitabaar, Nicholas Diana, Michael Liut, John Stamper
https://arxiv.org/abs/2506.19107

Improving Student-AI Interaction Through Pedagogical Prompting: An Example in Computer Science Education
With the proliferation of large language model (LLM) applications since 2022, their use in education has sparked both excitement and concern. Recent studies consistently highlight students' (mis)use of LLMs can hinder learning outcomes. This work aims to teach students how to effectively prompt LLMs to improve their learning. We first proposed pedagogical prompting, a theoretically-grounded new concept to elicit learning-oriented responses from LLMs. To move from concept design to a proof-of-co…

CMP: A Composable Meta Prompt for SAM-Based Cross-Domain Few-Shot Segmentation
Shuai Chen, Fanman Meng, Chunjin Yang, Haoran Wei, Chenhao Wu, Qingbo Wu, Hongliang Li
https://arxiv.org/abs/2507.16753

CMP: A Composable Meta Prompt for SAM-Based Cross-Domain Few-Shot Segmentation
Cross-Domain Few-Shot Segmentation (CD-FSS) remains challenging due to limited data and domain shifts. Recent foundation models like the Segment Anything Model (SAM) have shown remarkable zero-shot generalization capability in general segmentation tasks, making it a promising solution for few-shot scenarios. However, adapting SAM to CD-FSS faces two critical challenges: reliance on manual prompt and limited cross-domain ability. Therefore, we propose the Composable Meta-Prompt (CMP) framework t…

HiFo-Prompt: Prompting with Hindsight and Foresight for LLM-based Automatic Heuristic Design
Chentong Chen, Mengyuan Zhong, Jianyong Sun, Ye Fan, Jialong Shi
https://arxiv.org/abs/2508.13333

HiFo-Prompt: Prompting with Hindsight and Foresight for LLM-based Automatic Heuristic Design
LLM-based Automatic Heuristic Design (AHD) within Evolutionary Computation (EC) frameworks has shown promising results. However, its effectiveness is hindered by the use of static operators and the lack of knowledge accumulation mechanisms. We introduce HiFo-Prompt, a framework that guides LLMs with two synergistic prompting strategies: Foresight and Hindsight. Foresight-based prompts adaptively steer the search based on population dynamics, managing the exploration-exploitation trade-off. In a…

MMSearch-R1: Incentivizing LMMs to Search
Jinming Wu, Zihao Deng, Wei Li, Yiding Liu, Bo You, Bo Li, Zejun Ma, Ziwei Liu
https://arxiv.org/abs/2506.20670 h…

MMSearch-R1: Incentivizing LMMs to Search
Robust deployment of large multimodal models (LMMs) in real-world scenarios requires access to external knowledge sources, given the complexity and dynamic nature of real-world information. Existing approaches such as retrieval-augmented generation (RAG) and prompt engineered search agents rely on rigid pipelines, often leading to inefficient or excessive search behaviors. We present MMSearch-R1, the first end-to-end reinforcement learning framework that enables LMMs to perform on-demand, multi…

IPIGuard: A Novel Tool Dependency Graph-Based Defense Against Indirect Prompt Injection in LLM Agents
Hengyu An, Jinghuai Zhang, Tianyu Du, Chunyi Zhou, Qingming Li, Tao Lin, Shouling Ji
https://arxiv.org/abs/2508.15310

IPIGuard: A Novel Tool Dependency Graph-Based Defense Against Indirect Prompt Injection in LLM Agents
Large language model (LLM) agents are widely deployed in real-world applications, where they leverage tools to retrieve and manipulate external data for complex tasks. However, when interacting with untrusted data sources (e.g., fetching information from public websites), tool responses may contain injected instructions that covertly influence agent behaviors and lead to malicious outcomes, a threat referred to as Indirect Prompt Injection (IPI). Existing defenses typically rely on advanced pro…

The Laude Institute, a nonprofit that runs the K Prize multi-round AI coding challenge, says a Brazilian prompt engineer won with just 7.5% of the right answers (Russell Brandom/TechCrunch)
https://techcrunch.com/2025/07/23/a-ne

A new AI coding challenge just published its first results – and they aren’t pretty | TechCrunch
A new AI coding challenge has revealed its first winner — and set a new bar for AI-powered software engineers.  On Wednesday at 5pm PST, the

Generalize across Homophily and Heterophily: Hybrid Spectral Graph Pre-Training and Prompt Tuning
Haitong Luo, Suhang Wang, Weiyao Zhang, Ruiqi Meng, Xuying Meng, Yujun Zhang
https://arxiv.org/abs/2508.11328

Generalize across Homophily and Heterophily: Hybrid Spectral Graph Pre-Training and Prompt Tuning
Graph ``pre-training and prompt-tuning'' aligns downstream tasks with pre-trained objectives to enable efficient knowledge transfer under limited supervision. However, existing methods rely on homophily-based low-frequency knowledge, failing to handle diverse spectral distributions in real-world graphs with varying homophily. Our theoretical analysis reveals a spectral specificity principle: optimal knowledge transfer requires alignment between pre-trained spectral filters and the intrinsic spe…

Identifying Prompted Artist Names from Generated Images
Grace Su, Sheng-Yu Wang, Aaron Hertzmann, Eli Shechtman, Jun-Yan Zhu, Richard Zhang
https://arxiv.org/abs/2507.18633 http…

Identifying Prompted Artist Names from Generated Images
A common and controversial use of text-to-image models is to generate pictures by explicitly naming artists, such as "in the style of Greg Rutkowski". We introduce a benchmark for prompted-artist recognition: predicting which artist names were invoked in the prompt from the image alone. The dataset contains 1.95M images covering 110 artists and spans four generalization settings: held-out artists, increasing prompt complexity, multiple-artist prompts, and different text-to-image models. We eval…

Prompt-Based One-Shot Exact Length-Controlled Generation with LLMs
Juncheng Xie, Hung-yi Lee
https://arxiv.org/abs/2508.13805 https://arxiv.org/pdf/2508.13…

Prompt-Based One-Shot Exact Length-Controlled Generation with LLMs
Controlling the length of text produced by large language models (LLMs) remains challenging: models frequently overshoot or undershoot explicit length instructions because they cannot reliably keep an internal token count. We present a prompt-based, one-shot strategy that compels an off-the-shelf LLM to generate exactly a desired number of tokens - words (English) or characters (Chinese) - without any fine-tuning or iterative sampling. The prompt appends countdown markers and explicit counting …

Response and Prompt Evaluation to Prevent Parasocial Relationships with Chatbots
Emma Rath, Stuart Armstrong, Rebecca Gorman
https://arxiv.org/abs/2508.15748 https://

Response and Prompt Evaluation to Prevent Parasocial Relationships with Chatbots
The development of parasocial relationships with AI agents has severe, and in some cases, tragic effects for human well-being. Yet preventing such dynamics is challenging: parasocial cues often emerge gradually in private conversations, and not all forms of emotional engagement are inherently harmful. We address this challenge by introducing a simple response evaluation framework, created by repurposing a state-of-the-art language model, that evaluates ongoing conversations for parasocial cues …

Investigating the Role of LLMs Hyperparameter Tuning and Prompt Engineering to Support Domain Modeling
Vladyslav Bulhakov, Giordano d'Aloisio, Claudio Di Sipio, Antinisca Di Marco, Davide Di Ruscio
https://arxiv.org/abs/2507.14735

Investigating the Role of LLMs Hyperparameter Tuning and Prompt Engineering to Support Domain Modeling
The introduction of large language models (LLMs) has enhanced automation in software engineering tasks, including in Model Driven Engineering (MDE). However, using general-purpose LLMs for domain modeling has its limitations. One approach is to adopt fine-tuned models, but this requires significant computational resources and can lead to issues like catastrophic forgetting. This paper explores how hyperparameter tuning and prompt engineering can improve the accuracy of the Llama 3.1 model for…

TopicAttack: An Indirect Prompt Injection Attack via Topic Transition
Yulin Chen, Haoran Li, Yuexin Li, Yue Liu, Yangqiu Song, Bryan Hooi
https://arxiv.org/abs/2507.13686

TopicAttack: An Indirect Prompt Injection Attack via Topic Transition
Large language models (LLMs) have shown remarkable performance across a range of NLP tasks. However, their strong instruction-following capabilities and inability to distinguish instructions from data content make them vulnerable to indirect prompt injection attacks. In such attacks, instructions with malicious purposes are injected into external data sources, such as web documents. When LLMs retrieve this injected data through tools, such as a search engine and execute the injected instruction…

SPANER: Shared Prompt Aligner for Multimodal Semantic Representation
Thye Shan Ng, Caren Soyeon Han, Eun-Jung Holden
https://arxiv.org/abs/2508.13387 https://

SPANER: Shared Prompt Aligner for Multimodal Semantic Representation
Recent advances in multimodal Parameter-Efficient Fine-Tuning (PEFT) have significantly improved performance on downstream tasks such as few-shot retrieval. However, most existing approaches focus on task-specific gains while neglecting the structure of the multimodal embedding space. As a result, modality-specific representations often remain isolated, limiting cross-modal generalisation. In this work, we introduce Shared Prompt AligNER (SPANER), a modality-agnostic PEFT framework designed to …

Re-Evaluating Code LLM Benchmarks Under Semantic Mutation
Zhiyuan Pan, Xing Hu, Xin Xia, Xiaohu Yang
https://arxiv.org/abs/2506.17369 https://

Re-Evaluating Code LLM Benchmarks Under Semantic Mutation
In the era of large language models (LLMs), code benchmarks have become an important research area in software engineering and are widely used by practitioners. These benchmarks evaluate the performance of LLMs on specific code-related tasks, such as code understanding and generation. A critical step in constructing code benchmarks is the design of prompts. However, as existing code benchmarks typically rely on a single prompt template per task, they are prone to the issue of prompt sensitivity…

Incremental Object Detection with Prompt-based Methods
Matthias Neuwirth-Trapp, Maarten Bieshaar, Danda Pani Paudel, Luc Van Gool
https://arxiv.org/abs/2508.14599 https://

Incremental Object Detection with Prompt-based Methods
Visual prompt-based methods have seen growing interest in incremental learning (IL) for image classification. These approaches learn additional embedding vectors while keeping the model frozen, making them efficient to train. However, no prior work has applied such methods to incremental object detection (IOD), leaving their generalizability unclear. In this paper, we analyze three different prompt-based methods under a complex domain-incremental learning setting. We additionally provide a wide…

PROL : Rehearsal Free Continual Learning in Streaming Data via Prompt Online Learning
M. Anwar Ma'sum, Mahardhika Pratama, Savitha Ramasamy, Lin Liu, Habibullah Habibullah, Ryszard Kowalczyk
https://arxiv.org/abs/2507.12305

PROL : Rehearsal Free Continual Learning in Streaming Data via Prompt Online Learning
The data privacy constraint in online continual learning (OCL), where the data can be seen only once, complicates the catastrophic forgetting problem in streaming data. A common approach applied by the current SOTAs in OCL is with the use of memory saving exemplars or features from previous classes to be replayed in the current task. On the other hand, the prompt-based approach performs excellently in continual learning but with the cost of a growing number of trainable parameters. The first ap…

Innocence in the Crossfire: Roles of Skip Connections in Jailbreaking Visual Language Models
Palash Nandi, Maithili Joshi, Tanmoy Chakraborty
https://arxiv.org/abs/2507.13761

Innocence in the Crossfire: Roles of Skip Connections in Jailbreaking Visual Language Models
Language models are highly sensitive to prompt formulations - small changes in input can drastically alter their output. This raises a critical question: To what extent can prompt sensitivity be exploited to generate inapt content? In this paper, we investigate how discrete components of prompt design influence the generation of inappropriate content in Visual Language Models (VLMs). Specifically, we analyze the impact of three key factors on successful jailbreaks: (a) the inclusion of detailed…

HAMSA: Hijacking Aligned Compact Models via Stealthy Automation
Alexey Krylov, Iskander Vagizov, Dmitrii Korzh, Maryam Douiba, Azidine Guezzaz, Vladimir Kokh, Sergey D. Erokhin, Elena V. Tutubalina, Oleg Y. Rogov
https://arxiv.org/abs/2508.16484

HAMSA: Hijacking Aligned Compact Models via Stealthy Automation
Large Language Models (LLMs), especially their compact efficiency-oriented variants, remain susceptible to jailbreak attacks that can elicit harmful outputs despite extensive alignment efforts. Existing adversarial prompt generation techniques often rely on manual engineering or rudimentary obfuscation, producing low-quality or incoherent text that is easily flagged by perplexity-based filters. We present an automated red-teaming framework that evolves semantically meaningful and stealthy jailb…

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…

Doppelg\"anger Method: Breaking Role Consistency in LLM Agent via Prompt-based Transferable Adversarial Attack
Daewon Kang, YeongHwan Shin, Doyeon Kim, Kyu-Hwan Jung, Meong Hi Son
https://arxiv.org/abs/2506.14539

Doppelgänger Method: Breaking Role Consistency in LLM Agent via Prompt-based Transferable Adversarial Attack
Since the advent of large language models, prompt engineering now enables the rapid, low-effort creation of diverse autonomous agents that are already in widespread use. Yet this convenience raises urgent concerns about the safety, robustness, and behavioral consistency of the underlying prompts, along with the pressing challenge of preventing those prompts from being exposed to user's attempts. In this paper, we propose the ''Doppelgänger method'' to demonstrate the risk of an agent being hij…

How Software Engineers Engage with AI: A Pragmatic Process Model and Decision Framework Grounded in Industry Observations
Vahid Garousi, Zafar Jafarov
https://arxiv.org/abs/2507.17930

How Software Engineers Engage with AI: A Pragmatic Process Model and Decision Framework Grounded in Industry Observations
Artificial Intelligence (AI) has the potential to transform Software Engineering (SE) by enhancing productivity, efficiency, and decision support. Tools like GitHub Copilot and ChatGPT have given rise to "vibe coding"-an exploratory, prompt-driven development style. Yet, how software engineers engage with these tools in daily tasks, especially in deciding whether to trust, refine, or reject AI-generated outputs, remains underexplored. This paper presents two complementary contributions. First, …

From Confidence to Collapse in LLM Factual Robustness
Alina Fastowski, Bardh Prenkaj, Gjergji Kasneci
https://arxiv.org/abs/2508.16267 https://arxiv.org/pd…

From Confidence to Collapse in LLM Factual Robustness
Ensuring the robustness of factual knowledge in LLMs is critical for reliable applications in tasks such as question answering and reasoning. However, existing evaluation methods predominantly focus on performance-based metrics, often investigating from the perspective of prompt perturbations, which captures only the externally triggered side of knowledge robustness. To bridge this gap, we introduce a principled approach to measure factual robustness from the perspective of the generation proce…

NYC-based Bluefish Labs, which analyzes prompt responses for large brands to understand how LLMs answer consumer questions, raised a $20M Series A led by NEA (Kyt Dotson/SiliconANGLE)
https://siliconangle.com/2025/08/20/bluefish…

Bluefish raises $20M to help enterprise marketing take control in the era of AI chatbots - SiliconANGLE
Bluefish raises $20M to help enterprise marketing take control in the era of AI chatbots - SiliconANGLE

Mitigating Trojanized Prompt Chains in Educational LLM Use Cases: Experimental Findings and Detection Tool Design
Richard M. Charles, James H. Curry, Richard B. Charles
https://arxiv.org/abs/2507.14207

Mitigating Trojanized Prompt Chains in Educational LLM Use Cases: Experimental Findings and Detection Tool Design
The integration of Large Language Models (LLMs) in K--12 education offers both transformative opportunities and emerging risks. This study explores how students may Trojanize prompts to elicit unsafe or unintended outputs from LLMs, bypassing established content moderation systems with safety guardrils. Through a systematic experiment involving simulated K--12 queries and multi-turn dialogues, we expose key vulnerabilities in GPT-3.5 and GPT-4. This paper presents our experimental design, detai…

MK2 at PBIG Competition: A Prompt Generation Solution
Yuzheng Xu, Tosho Hirasawa, Seiya Kawano, Shota Kato, Tadashi Kozuno
https://arxiv.org/abs/2507.08335

MK2 at PBIG Competition: A Prompt Generation Solution
The Patent-Based Idea Generation task asks systems to turn real patents into product ideas viable within three years. We propose MK2, a prompt-centric pipeline: Gemini 2.5 drafts and iteratively edits a prompt, grafting useful fragments from weaker outputs; GPT-4.1 then uses this prompt to create one idea per patent, and an Elo loop judged by Qwen3-8B selects the best prompt-all without extra training data. Across three domains, two evaluator types, and six criteria, MK2 topped the automatic le…

LLM-Assisted Semantic Alignment and Integration in Collaborative Model-Based Systems Engineering Using SysML v2
Zirui Li, Stephan Husung, Haoze Wang
https://arxiv.org/abs/2508.16181

LLM-Assisted Semantic Alignment and Integration in Collaborative Model-Based Systems Engineering Using SysML v2
Cross-organizational collaboration in Model-Based Systems Engineering (MBSE) faces many challenges in achieving semantic alignment across independently developed system models. SysML v2 introduces enhanced structural modularity and formal semantics, offering a stronger foundation for interoperable modeling. Meanwhile, GPT-based Large Language Models (LLMs) provide new capabilities for assisting model understanding and integration. This paper proposes a structured, prompt-driven approach for LLM…

Beyond Syntax: Action Semantics Learning for App Agents
Bohan Tang, Dezhao Luo, Jingxuan Chen, Shaogang Gong, Jianye Hao, Jun Wang, Kun Shao
https://arxiv.org/abs/2506.17697

Beyond Syntax: Action Semantics Learning for App Agents
The advent of Large Language Models (LLMs) enables the rise of App agents that interpret user intent and operate smartphone Apps through actions such as clicking and scrolling. While prompt-based solutions with closed LLM APIs show promising ability, they incur heavy compute costs and external API dependency. Fine-tuning smaller open-source LLMs solves these limitations. However, current fine-tuning methods use a syntax learning paradigm that forces agents to reproduce exactly the ground truth …

LLM-empowered Dynamic Prompt Routing for Vision-Language Models Tuning under Long-Tailed Distributions
Yongju Jia, Jiarui Ma, Xiangxian Li, Baiqiao Zhang, Xianhui Cao, Juan Liu, Yulong Bian
https://arxiv.org/abs/2508.15688

LLM-empowered Dynamic Prompt Routing for Vision-Language Models Tuning under Long-Tailed Distributions
Pre-trained vision-language models (VLMs), such as CLIP, have demonstrated impressive capability in visual tasks, but their fine-tuning often suffers from bias in class-imbalanced scene. Recent works have introduced large language models (LLMs) to enhance VLM fine-tuning with supplementing semantic information. However, they often overlook inherent class imbalance in VLMs' pre-training, which may lead to bias accumulation in downstream tasks. To address this problem, this paper proposes a Multi…

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 …

To Protect the LLM Agent Against the Prompt Injection Attack with Polymorphic Prompt
Zhilong Wang, Neha Nagaraja, Lan Zhang, Hayretdin Bahsi, Pawan Patil, Peng Liu
https://arxiv.org/abs/2506.05739

To Protect the LLM Agent Against the Prompt Injection Attack with Polymorphic Prompt
LLM agents are widely used as agents for customer support, content generation, and code assistance. However, they are vulnerable to prompt injection attacks, where adversarial inputs manipulate the model's behavior. Traditional defenses like input sanitization, guard models, and guardrails are either cumbersome or ineffective. In this paper, we propose a novel, lightweight defense mechanism called Polymorphic Prompt Assembling (PPA), which protects against prompt injection with near-zero overhe…

Prompt Injection 2.0: Hybrid AI Threats
Jeremy McHugh, Kristina \v{S}ekrst, Jon Cefalu
https://arxiv.org/abs/2507.13169 https://arxiv…

Prompt Injection 2.0: Hybrid AI Threats
Prompt injection attacks, where malicious input is designed to manipulate AI systems into ignoring their original instructions and following unauthorized commands instead, were first discovered by Preamble, Inc. in May 2022 and responsibly disclosed to OpenAI. Over the last three years, these attacks have continued to pose a critical security threat to LLM-integrated systems. The emergence of agentic AI systems, where LLMs autonomously perform multistep tasks through tools and coordination with…

Input Reduction Enhanced LLM-based Program Repair
Boyang Yang, Luyao Ren, Xin Yin, Jiadong Ren, Haoye Tian, Shunfu Jin
https://arxiv.org/abs/2507.15251 htt…

Input Reduction Enhanced LLM-based Program Repair
Large Language Models (LLMs) have shown great potential in Automated Program Repair (APR). Test inputs, being crucial for reasoning the root cause of failures, are always included in the prompt for LLM-based APR. Unfortunately, LLMs struggle to retain key information in long prompts. When the test inputs are extensive in the prompt, this may trigger the "lost-in-the-middle" issue, compromising repair performance. To address this, we propose ReduceFix, an LLM-based APR approach with a built-in c…

Prompt Attacks Reveal Superficial Knowledge Removal in Unlearning Methods
Yeonwoo Jang, Shariqah Hossain, Ashwin Sreevatsa, Diogo Cruz
https://arxiv.org/abs/2506.10236

Prompt Attacks Reveal Superficial Knowledge Removal in Unlearning Methods
In this work, we show that some machine unlearning methods may fail when subjected to straightforward prompt attacks. We systematically evaluate eight unlearning techniques across three model families, and employ output-based, logit-based, and probe analysis to determine to what extent supposedly unlearned knowledge can be retrieved. While methods like RMU and TAR demonstrate robust unlearning, ELM remains vulnerable to specific prompt attacks (e.g., Hindi filler text in original prompt recover…

Cross-Prompt Encoder for Low-Performing Languages
Beso Mikaberidze, Teimuraz Saghinadze, Simon Ostermann, Philipp Muller
https://arxiv.org/abs/2508.10352 https://

Cross-Prompt Encoder for Low-Performing Languages
Soft prompts have emerged as a powerful alternative to adapters in parameter-efficient fine-tuning (PEFT), enabling large language models (LLMs) to adapt to downstream tasks without architectural changes or parameter updates. While prior work has focused on stabilizing training via parameter interaction in small neural prompt encoders, their broader potential for transfer across languages remains unexplored. In this paper, we demonstrate that a prompt encoder can play a central role in improvin…

Cost-Aware Routing for Efficient Text-To-Image Generation
Qinchan (Wing), Li (Tina), Kenneth Chen (Tina), Changyue (Tina), Su, Wittawat Jitkrittum, Qi Sun, Patsorn Sangkloy
https://arxiv.org/abs/2506.14753

Cost-Aware Routing for Efficient Text-To-Image Generation
Diffusion models are well known for their ability to generate a high-fidelity image for an input prompt through an iterative denoising process. Unfortunately, the high fidelity also comes at a high computational cost due the inherently sequential generative process. In this work, we seek to optimally balance quality and computational cost, and propose a framework to allow the amount of computation to vary for each prompt, depending on its complexity. Each prompt is automatically routed to the m…

Inductive Bias Extraction and Matching for LLM Prompts
Christian M. Angel, Francis Ferraro
https://arxiv.org/abs/2508.10295 https://arxiv.org/pdf/2508.1029…

Inductive Bias Extraction and Matching for LLM Prompts
The active research topic of prompt engineering makes it evident that LLMs are sensitive to small changes in prompt wording. A portion of this can be ascribed to the inductive bias that is present in the LLM. By using an LLM's output as a portion of its prompt, we can more easily create satisfactory wording for prompts. This has the effect of creating a prompt that matches the inductive bias in model. Empirically, we show that using this Inductive Bias Extraction and Matching strategy improves …