Prompt Injection as Role Confusion
arXiv:2603.12277v3 Announce Type: replace-cross Abstract: Language models remain vulnerable to prompt injection attacks despite extensive safety training. We trace this failure to role confusion: mode
Knowledge catalogue
arXiv:2603.12277v3 Announce Type: replace-cross Abstract: Language models remain vulnerable to prompt injection attacks despite extensive safety training. We trace this failure to role confusion: mode
arXiv:2508.17784v2 Announce Type: replace-cross Abstract: Supervised fine-tuning (SFT) of foundation models often leads to poor generalization, where prior capabilities deteriorate after tuning on new
arXiv:2602.19509v3 Announce Type: replace-cross Abstract: We observe that LLM cascading and routing implicitly solves an anytime computation problem -- a class of algorithms, well-studied in classical
arXiv:2603.16068v3 Announce Type: replace-cross Abstract: Given limited and costly computational infrastructure, resource efficiency is a key requirement for large language models (LLMs). Efficient LL
arXiv:2604.11082v1 Announce Type: new Abstract: Visual glitches in video games degrade player experience and perceived quality, yet manual quality assurance cannot scale to the growing test surface of
arXiv:2604.11759v1 Announce Type: new Abstract: Organizational knowledge used by AI agents typically lacks epistemic structure: retrieval systems surface semantically relevant content without distingu
arXiv:2604.10916v1 Announce Type: cross Abstract: Ultrasound acquisition requires skilled probe manipulation and real-time adjustments. Vision-language models (VLMs) could enable autonomous ultrasound
arXiv:2604.11037v1 Announce Type: cross Abstract: Multi-step agentic reinforcement learning benefits from fine-grained credit assignment, yet existing approaches offer limited options: critic-free met
arXiv:2503.05534v3 Announce Type: replace Abstract: Purpose: The Segment Anything Model (SAM) promises to ease the annotation bottleneck in medical segmentation, but overlapping anatomy and blurred bo
arXiv:2604.11402v1 Announce Type: new Abstract: Scene change detection (SCD) is crucial for urban monitoring and navigation but remains challenging in real-world environments due to lighting variation
arXiv:2511.18329v3 Announce Type: replace Abstract: Scientific posters play a vital role in academic communication by presenting ideas through visual summaries. Analyzing reading order and parent-chil
arXiv:2604.10299v1 Announce Type: cross Abstract: Large Vision-Language Models (LVLMs) rely on attention-based retrieval of safety instructions to maintain alignment during generation. Existing attack
arXiv:2505.15489v4 Announce Type: replace-cross Abstract: The impact of multimodal misinformation arises not only from factual inaccuracies but also from the misleading narratives that creators delibe
arXiv:2604.10027v1 Announce Type: new Abstract: Large language models (LLMs) suffer from hallucination and context forgetting. Prior studies suggest that attention drift is a primary cause of these pr
arXiv:2604.11140v1 Announce Type: new Abstract: Integrating frame-based RGB cameras with event streams offers a promising solution for robust object detection under challenging dynamic conditions. How
arXiv:2604.10527v1 Announce Type: cross Abstract: Referring multi-object tracking (RMOT) is a task of associating all the objects in a video that semantically match with given textual queries or refer
arXiv:2604.09562v1 Announce Type: cross Abstract: Efficient LLM serving must balance throughput and latency across diverse, bursty workloads. We introduce StreamServe, a disaggregated prefill decode s
arXiv:2511.06101v3 Announce Type: replace-cross Abstract: Web agents struggle to adapt to new websites due to the scarcity of environment specific tasks and demonstrations. Recent works have explored
As “deploy fast” meets enterprise reality, data governance has quickly emerged as the bottleneck between AI ambition and outcomes. The disconnect is measurable. In their Agentic AI Study, Qlik Technol
arXiv:2604.11724v1 Announce Type: new Abstract: The age of artificial intelligence has brought many new possibilities and pitfalls in many fields and tasks. The devil is in the details, and those come
arXiv:2411.06376v3 Announce Type: replace-cross Abstract: Peripheral Component Interconnect Express (PCIe) is the de facto interconnect standard for high-speed peripherals and CPUs. The development of
arXiv:2604.09911v1 Announce Type: cross Abstract: In the psychological literature the term `general intelligence' describes correlations between abilities and not simply the number of abilities. This
arXiv:2604.10999v1 Announce Type: new Abstract: Vision-language models (VLMs) perform strongly on many multimodal benchmarks. However, the ability to follow complex visual paths -- a task that human o
arXiv:2604.11522v1 Announce Type: new Abstract: Reinforcement learning for open-ended text generation is constrained by the lack of verifiable rewards, necessitating reliance on judge models that requ
arXiv:2604.11036v1 Announce Type: cross Abstract: Scientific fact-checking is vital for assessing claims in specialized domains such as biomedicine and materials science, yet existing systems often ha
arXiv:2604.11611v1 Announce Type: new Abstract: To overcome the sparse reward challenge in reinforcement learning (RL) for agents based on large language models (LLMs), we propose Mutual Information S
arXiv:2604.10800v1 Announce Type: cross Abstract: Learned classifiers deployed in agentic pipelines face a fundamental reliability problem: predictions are probabilistic inferences, not verified concl
arXiv:2604.10341v1 Announce Type: new Abstract: extbf{VeriTrans} is a reliability-first ML system that compiles natural-language requirements into solver-ready logic with validator-gated reliability.
arXiv:2604.09943v1 Announce Type: new Abstract: Reservoir computing (RC) is a computational framework known for its training efficiency, making it ideal for physical hardware implementations. However,
arXiv:2604.10167v1 Announce Type: cross Abstract: Multi-vector models dominate Visual Document Retrieval (VDR) due to their fine-grained matching capabilities, but their high storage and computational
arXiv:2506.02387v3 Announce Type: replace Abstract: Recent advancements in Vision Language Models (VLMs) have expanded their capabilities to interactive agent tasks, yet existing benchmarks remain lim
Cursor AI shared observations on X validating that multi-agent architectures demonstrate superior performance when tackling novel problems that fall outside the distribution of training data. The post
arXiv:2604.11374v1 Announce Type: cross Abstract: Personalized image aesthetics assessment (PIAA) is an important research problem with practical real-world applications. While methods based on vision
arXiv:2604.09885v1 Announce Type: new Abstract: Recent work has shown that probing model internals can reveal a wealth of information not apparent from the model generations. This poses the risk of un
arXiv:2407.08101v4 Announce Type: replace Abstract: Vision-language models have shown impressive progress in recent years. However, existing models are largely limited to turn-based interactions, wher
arXiv:2604.02458v2 Announce Type: replace-cross Abstract: Behavioral simulation is increasingly used to anticipate responses to interventions. Large language models (LLMs) enable researchers to specif
arXiv:2604.10996v1 Announce Type: cross Abstract: Can large language models (LLMs) generate continuous numerical features that improve reinforcement learning (RL) trading agents? We build a modular pi
arXiv:2604.11663v1 Announce Type: new Abstract: Large Language Models (LLMs) have been shown to generate harmful content. However, the underlying causes of such behavior remain under explored. We prop
arXiv:2604.10495v1 Announce Type: new Abstract: As Large Language Models (LLMs) are increasingly deployed in real-world applications, reliable uncertainty quantification (UQ) becomes critical for safe
arXiv:2604.10589v1 Announce Type: new Abstract: We introduce a hierarchical categorical framework for Schema-Based Learning (SBL) structured across four interconnected levels. At the schema level, a f
arXiv:2604.09244v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models have emerged as the mainstream of embodied intelligence. Recent VLA models have expanded their input modalities fr
arXiv:2604.09294v1 Announce Type: new Abstract: Dexterity is a central yet ambiguously defined concept in the design and evaluation of anthropomorphic robotic hands. In practice, the term is often use
arXiv:2604.08588v1 Announce Type: cross Abstract: Effective automation hinges on deciding when to act and when to escalate. We model this as a decision under uncertainty: an LLM forms a prediction, es
arXiv:2604.05529v2 Announce Type: replace Abstract: Human mobility modeling is indispensable for diverse urban applications. However, existing data-driven methods often suffer from data scarcity, limi
arXiv:2604.04161v2 Announce Type: replace Abstract: In Vision-Language-Action (VLA) models, action chunking (i.e., executing a sequence of actions without intermediate replanning) is a key technique t
arXiv:2604.09035v1 Announce Type: new Abstract: Model-based reinforcement learning (MBRL) with autoregressive world models suffers from compounding errors, whereas diffusion world models mitigate this
arXiv:2604.09470v1 Announce Type: new Abstract: Translating natural language into Jira Query Language (JQL) requires resolving ambiguous field references, instance-specific categorical values, and com
arXiv:2604.08722v1 Announce Type: cross Abstract: Sport analysis is crucial for team performance since it provides actionable data that can inform coaching decisions, improve player performance, and e
arXiv:2510.26641v4 Announce Type: replace Abstract: Autonomous Vehicles (AVs) are transforming the future of transportation through advances in intelligent perception, decision-making, and control sys
arXiv:2604.09418v1 Announce Type: new Abstract: This paper studies Automated Instruction Revision (AIR), a rule-induction-based method for adapting large language models (LLMs) to downstream tasks usi
arXiv:2604.08552v1 Announce Type: cross Abstract: Scientific metadata are often incomplete and noncompliant with community standards, limiting dataset findability, interoperability, and reuse. When re
Benchmarked @DJLougen ’s Ornstein-27B-v2 Q6_K on my RTX 3090 using hermes-bench, my new open-source benchmarking UI for local LLMs and Hermes agents. Ornstein is a Qwen 3.5 27B fine-tune trained on re
arXiv:2507.04736v2 Announce Type: replace Abstract: Large Language Models have emerged as powerful tools for automating Register-Transfer Level (RTL) code generation, yet they face critical limitation
arXiv:2503.00035v2 Announce Type: replace-cross Abstract: Large language models (LLMs) struggle with hallucinations due to false or outdated knowledge. Given the high resource demands of retraining th
arXiv:2510.21588v2 Announce Type: replace-cross Abstract: Biological and artificial learners are inherently exposed to a stream of data and experience throughout their lifetimes and must constantly ad
arXiv:2604.08577v1 Announce Type: cross Abstract: Large Language Models (LLMs) tend to respond correctly to prompts that align to the data they were trained and fine-tuned on. Yet, small shifts in wor
arXiv:2604.09425v1 Announce Type: new Abstract: Vision Language Models (VLMs) have achieved remarkable success by integrating visual encoders with large language models (LLMs). While VLMs process dens
arXiv:2604.08565v1 Announce Type: cross Abstract: At typical context lengths, the feed-forward MLP block accounts for a large share of a transformer's compute budget, motivating sparse alternatives to
arXiv:2604.08644v1 Announce Type: new Abstract: This technical report introduces EXAONE 4.5, the first open-weight vision language model released by LG AI Research. EXAONE 4.5 is architected by integr
arXiv:2604.02598v2 Announce Type: replace-cross Abstract: LLM-generated explanations can make technical content more accessible, but there is a ceiling on what they can support interactively. Because