50 articles mentioning Prompt-Guard-86M
arXiv:2609.04197v1 Announce Type: cross Abstract: Evolutionary prompt optimizers such as GEPA suffer from prompt bloat: each iteration appends rules and caveats, producing prompts up to 3$\times$ longer yet no more accurate. We trace this to three deficiencies - incomplete error observation, limited
arXiv:2609.03402v1 Announce Type: new Abstract: Artificial intelligence (AI) teaching assistants powered by large language models (LLMs) offer scalable educational support but often provide limited personalization. This study presents a prompt-engineering-based framework for personalizing general-pu
arXiv:2606.21657v2 Announce Type: replace-cross Abstract: Do people perceive the same facial expression in the same way? Should we expect vision models to be flexible in how they perceive facial expressions? Facial expressions are nonverbal social signals used in human interaction, but facial expres
arXiv:2412.08414v2 Announce Type: replace Abstract: Mental manipulation severely undermines mental wellness by covertly and negatively distorting decision-making. While there is an increasing interest in mental health care within the natural language processing community, progress in tackling manipu
arXiv:2609.03662v1 Announce Type: new Abstract: Recent unlearning methods (e.g. NPO, DPO, LUNAR) make use of refusal alignment to suppress forgotten data. However, it has been shown that refusal responses might leave traces of unlearning, and recent attacks have been able to successfully recover som
arXiv:2607.15565v2 Announce Type: replace-cross Abstract: Where should the question go in a vision-language model (VLM) prompt: before the image or after it? Intuition says before: knowing what is asked should tell the model where to look. Yet across visual question answering benchmarks, question-fi
arXiv:2609.02998v1 Announce Type: cross Abstract: On-policy distillation (OPD) accelerates post-training by providing dense token-level supervision from a frozen teacher on the student's own rollouts. Vanilla OPD applies this supervision uniformly across prompts, without checking whether the teacher
arXiv:2609.03218v1 Announce Type: new Abstract: Large Language Models (LLMs) increasingly use user context such as memory, profiles, and role prompts to personalize their responses. This personalization can affect evidence-based judgment: the same evidence may lead to different conclusions under dif
arXiv:2609.03635v1 Announce Type: new Abstract: Clinical trials in the UK can cost up to {\pounds}1.3 million, with approximately 90% drug failure rate. Toxicity is a major contributing factor in drug failure. Testing is time and cost intensive. In recent years, the use of artificial intelligence ha
arXiv:2609.02093v1 Announce Type: new Abstract: To address the sequential and evolving nature of time series, the Online Time Series Forecasting (OTSF) task has been extensively studied in multiple domains. Existing research focuses on adapting to non-stationary environments by employing memory buff
arXiv:2609.01802v1 Announce Type: new Abstract: Prompt tuning provides a parameter-efficient way to adapt foundation models (FMs) by freezing the pretrained backbone and updating only a small set of learnable prompts. This property makes prompt tuning especially suitable for decentralized federated
arXiv:2606.17372v3 Announce Type: replace-cross Abstract: Two recent studies \citep{jones2026llms, zeng2026lvlms} reach apparently contradictory conclusions about whether large vision-language models (LVLMs) can coordinate similarly to humans on efficient referring expressions. We control for task d
arXiv:2603.18593v2 Announce Type: replace Abstract: We propose a method that represents language models by log-likelihood vectors over prompt-response pairs and constructs model maps for comparing their conditional distributions. In this space, squared Euclidean distances between models are approxim
arXiv:2609.01615v1 Announce Type: new Abstract: Personalizing a frozen large language model (LLM) to individual users is often framed as a meta-learning problem in prompt space: each user is a task, and one seeks a shared natural-language adaptation policy that, given a handful of the user's labeled
arXiv:2603.25184v3 Announce Type: replace-cross Abstract: Reinforcement learning (RL) has become essential for post-training large language models (LLMs) in reasoning tasks. While scaling rollouts can stabilize training and enhance performance, the computational overhead is a critical issue. In algo
arXiv:2511.03913v3 Announce Type: replace-cross Abstract: Deep diffusion models have revolutionized image generation by producing high-quality outputs. However, achieving specific objectives with these models often requires costly adaptations such as fine-tuning, which can be resource-intensive and
arXiv:2606.03557v2 Announce Type: replace Abstract: As generative AI capabilities expand, AI-driven virtual worlds face a growing architectural challenge. Users interact through in-world interfaces in multimodal ways, yet their requests demand fundamentally different AI backend models and computatio
arXiv:2605.03096v2 Announce Type: replace-cross Abstract: In classification tasks, models may rely on confounding variables to achieve strong in-distribution performance, capturing spurious features that fail under distribution shift. This shortcut behavior leads to substantial degradation in out-of
arXiv:2407.14845v4 Announce Type: replace-cross Abstract: Large language models (LLMs) are widely used in decision-making across diverse domains. Ensuring the generation of safe and reliable responses is critical for the effective deployment of LLM-based applications, particularly in high-stakes dom
arXiv:2601.12983v4 Announce Type: replace Abstract: Multimodal large language models (MLLMs) are increasingly used to automate chart generation from data tables, improving efficiency but introducing new misuse risks. We present ChartAttack, a framework for evaluating how MLLMs use design misleaders
arXiv:2606.18216v2 Announce Type: replace Abstract: Knowledge distillation transfers a teacher's competence to a small student but is brittle in the small-student regime: forcing the student to imitate logits from a much larger teacher concentrates it on the teacher's sharpest modes, hurting general
arXiv:2607.00684v4 Announce Type: replace Abstract: The classification accuracy of pretrained Vision-Language Models (VLMs) relies on the quality of the text prompts. Handcrafted templates and Large Language Model (LLM)-generated descriptions not only make predictions more interpretable, but also en
arXiv:2609.01798v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly deployed on mobile devices, making energy efficiency a key deployment constraint, yet the energy impact of prompt design remains underexplored. This paper aims to understand how two prompt properties, cogni
arXiv:2609.00047v1 Announce Type: cross Abstract: Graph prompt learning is an effective paradigm to adapt pre-trained graph models to downstream tasks in low-resource scenarios. However, existing multi-task graph pre-training frameworks generally use randomly initialized prompts, leading to poor ali
arXiv:2609.00621v1 Announce Type: new Abstract: Prompt optimization can improve multi-agent LLM systems, but the prompts being optimized often serve two entangled roles: generating task-relevant content and specifying execution-critical protocols, such as message routing, output formatting, and term
arXiv:2609.00222v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used as judges for subjective tasks, where annotators disagree and the relevant question is not only how accurate a judge is, but whose judgments it reproduces. Sociodemographic prompting conditions the jud
arXiv:2609.01046v1 Announce Type: cross Abstract: Production LLMs must handle inputs that attempt to override system instructions, bypass safety policies or elicit harmful responses. A common mitigation is a separate guardrail model. Existing reports, however, provide little evidence on Russian prom
arXiv:2609.01217v1 Announce Type: new Abstract: Despite their strong performance, large language models remain highly sensitive to prompt formulation. Prior work addresses this through refined data construction or through dedicated robustness objectives. We reproduce and compare these strategies und
arXiv:2606.06698v4 Announce Type: replace-cross Abstract: Production agentic systems routinely face evolving constraints and must comply from the very next interaction. Scenarios like a tool-call notification changing a compliance threshold or a policy update adding disclosure requirements fit this
arXiv:2605.21776v2 Announce Type: replace Abstract: Estimating mutual information from text usually requires training a task-specific critic, which limits its use in low-data settings. We ask whether large language models can instead estimate pointwise mutual information zero-shot, using only prompt
arXiv:2609.01310v1 Announce Type: cross Abstract: Medical image segmentation remains difficult to scale because high-performing methods typically rely on dense expert annotations and task-specific training. We introduce GazeRefine, a training-free framework that uses gaze as an inference-time prompt
arXiv:2609.00709v1 Announce Type: cross Abstract: Large Vision-Language Models produce fluent image descriptions but offer limited semantic control: users cannot reliably specify whether captions should emphasize attributes, relations, or particular image regions. We present Fine-grained Captioning
arXiv:2603.09493v3 Announce Type: replace-cross Abstract: The adaptation of large-scale vision-language models (VLMs) to downstream tasks with limited labeled data remains a significant challenge. While parameter-efficient prompt learning methods offer a promising path, they often suffer from catast
arXiv:2609.01249v1 Announce Type: cross Abstract: Invisible watermarks are typically evaluated against predefined perturbations such as compression, blur, noise, cropping, and denoising. Public foundation image models expose a distinct threat: an attacker can submit a watermarked image with a single
arXiv:2608.30362v2 Announce Type: replace Abstract: As LLM agents take real-world actions through tools, indirect prompt injection (IPI) has emerged as a serious threat. The standard metric, Attack Success Rate (ASR), counts whether an injection succeeds but ignores what the user notices in the agen
arXiv:2608.29790v2 Announce Type: replace Abstract: As large language models (LLMs) continue to scale, parameter-efficient fine-tuning (PEFT) has become a practical alternative to full-parameter adaptation. Prompt tuning is effective, but existing approaches either use flat prompt structures or hier
With Google Pics, Google is pushing deeper into the creative software market dominated by Canva and Adobe, but with a distinctly AI-first approach.
arXiv:2608.18265v2 Announce Type: replace-cross Abstract: We introduce a general, easy-to-implement AI-based method for modeling and analyzing the structure and complexity of human behavior. We assign a large language model a "type vector" and then prompt it to choose actions across settings in whic
arXiv:2607.23991v2 Announce Type: replace-cross Abstract: Large Language Models (LLMs) are increasingly controlled through system prompts that specify roles, formats, and safety requirements. However, models follow these prompts only implicitly through in-context learning, which can be insufficient
arXiv:2608.06714v2 Announce Type: replace Abstract: Recent systems for optimizing prompts, programs, and ML workflows typically rely on explicit outer-loop controllers such as evolutionary search, bandits, or textual-gradient methods. We ask a fundamentally different question: how much of this searc
arXiv:2412.15957v2 Announce Type: replace-cross Abstract: The rapid development of large language models (LLMs) has transformed many industries, including healthcare. In practice, hospitals and patients increasingly seek LLM-based systems capable of interpreting personal health records and providing
arXiv:2608.30362v1 Announce Type: new Abstract: As LLM agents take real-world actions through tools, indirect prompt injection (IPI) has emerged as a serious threat. The standard metric, Attack Success Rate (ASR), counts whether an injection succeeds but ignores what the user notices in the agent's
arXiv:2608.30086v1 Announce Type: new Abstract: Credit-default prediction is an important task in financial decision making. Traditional methods use fitted classifiers such as logistic regression and random forests on tabular features. Large language models (LLMs) have recently been applied to this
arXiv:2608.30230v1 Announce Type: new Abstract: Test-time prompt tuning (TPT) has emerged as a powerful paradigm, refining prompts for each test sample via entropy minimization (EM) over multiple augmented views. However, we identify a limitation in the standard EM-based adaptation: it inherently dr
arXiv:2608.02665v2 Announce Type: replace-cross Abstract: A benchmark score is a measurement instrument, yet most benchmarks read each item at a single canonical surface form. We ask whether that reading is faithful: when an item's intent is held fixed and only its meaning-preserving surface form va
arXiv:2608.28640v1 Announce Type: cross Abstract: In this paper, we present PromptKWS, a novel Prompt-guided keyword spotting (KWS) framework to improve the accuracy of open vocabulary KWS systems. In specific terms, we introduce the Prompt Phrases Prediction Network (PPN), an encoder-decoder archit
arXiv:2608.28875v1 Announce Type: cross Abstract: Supplying context at inference time to a large multimodal model is an inexpensive lever for adapting speech transcription to a domain, and earlier results on smaller models reported large gains. This work tested that mechanism where it ships, in the
arXiv:2608.30041v1 Announce Type: cross Abstract: Large language model agents place outputs from external skills into their execution context, allowing attacker-controlled data to influence later privileged actions. Existing defenses mainly classify untrusted content or authorize proposed operations
arXiv:2608.29054v1 Announce Type: new Abstract: Graph Neural Networks (GNNs) have emerged as a cornerstone for representing complex relational dependencies in diverse multimedia tasks, particularly in cross-platform user interest modeling and cross-modal semantic alignment. In the real world, a prac
arXiv:2511.19811v2 Announce Type: replace-cross Abstract: Image diversity remains a fundamental challenge for text-to-image diffusion models. Low-diversity generation often leads to repetitive outputs, increasing sampling redundancy and hindering both creative exploration and downstream applications