Search: “guardrails”

arXiv:2604.09443v1 Announce Type: cross Abstract: Large language model agents receive instructions from many sources-system messages, user prompts, tool outputs, and more-each carrying different level

arXiv:2604.06696v1 Announce Type: new Abstract: The rapid development of AI agent systems is leading to an emerging Internet of Agents, where specialized agents operate across local devices, edge node

arXiv:2604.02022v2 Announce Type: replace Abstract: Evaluating the safety of LLM-based agents is increasingly important because risks in realistic deployments often emerge over multi-step interactions

arXiv:2604.06185v1 Announce Type: cross Abstract: Fulfilling user needs through Large Language Model multi-turn, multi-step tool-use is rarely a straightforward process. Real user interactions are inh

arXiv:2604.06233v1 Announce Type: new Abstract: Safety-trained language models routinely refuse requests for help circumventing rules. But not all rules deserve compliance. When users ask for help eva

arXiv:2604.05292v2 Announce Type: replace-cross Abstract: AI coding assistants are now used to generate production code in security-sensitive domains, yet the exploitability of their outputs remains u

arXiv:2604.05172v2 Announce Type: replace Abstract: Large language model (LLM) agents are increasingly deployed to automate productivity tasks (e.g., email, scheduling, document management), but evalu

arXiv:2604.06483v1 Announce Type: cross Abstract: Large language models that require multiple GPU cards to host are usually the most capable models. It is necessary to understand and steer these model

arXiv:2604.06742v1 Announce Type: cross Abstract: Large Language Models (LLMs) are driving a shift towards intent-driven development, where agents build complete software from scratch. However, existi

arXiv:2604.02360v1 Announce Type: cross Abstract: The transformative potential of large language models (LLMs) in education, such as improving accessibility and personalized learning, is being eclipse

arXiv:2604.06247v1 Announce Type: cross Abstract: Large Language Models (LLMs) and Vision-Language Models (VLMs) remain highly vulnerable to textual and visual jailbreaks, as well as prompt injections

arXiv:2604.06765v1 Announce Type: cross Abstract: Recently, multi-Large Language Model (LLM) frameworks have been proposed to solve contextualized tasks. However, these frameworks do not explicitly em

arXiv:2604.06427v1 Announce Type: cross Abstract: The viability of chain-of-thought (CoT) monitoring hinges on models being unable to reason effectively in their latent representations. Yet little is

arXiv:2604.06182v1 Announce Type: cross Abstract: Existing online benchmarks for mobile GUI agents remain largely app-centric and task-homogeneous, failing to reflect the diversity and instability of

arXiv:2610.13806v1 Announce Type: cross Abstract: Production recommender systems have feature-lifecycle asymmetry: adding new features must produce gains on some metric, while feature removal must not

arXiv:2610.12176v1 Announce Type: new Abstract: Recursive self-improvement (RSI) of a model on non-verifiable tasks, such as open-ended research, faces a supervision bottleneck when its outputs exceed

arXiv:2610.09764v1 Announce Type: new Abstract: Large Language Models (LLMs) can generate drone programs from natural-language mission descriptions, but syntactically valid programs may still violate

arXiv:2610.09043v1 Announce Type: cross Abstract: In human-AI collaborative decision making, human review can prevent unsafe AI decisions, but each human judgment is costly. Treating human interventio

arXiv:2610.10276v1 Announce Type: cross Abstract: An LLM safety patch can pass a benchmark while still being a poor repair. This risk is especially acute for jailbreak repairs, where the goal is to co

arXiv:2610.07886v1 Announce Type: new Abstract: Dietary nutrition planning plays an important role in chronic disease management and maintaining a healthy body. In applications, it must simultaneously

arXiv:2610.07723v1 Announce Type: cross Abstract: Safety alignment in Large Language Models (LLMs) remains vulnerable to backdoor attacks. Existing LLM backdoors are almost all input-centric: activati

arXiv:2610.03036v1 Announce Type: cross Abstract: We present WebFovea, a vision-based web agent that placed 2nd in the WebRetriever Challenge 2026 with a final score of 57.0 out of 100. The challenge

arXiv:2610.01326v1 Announce Type: new Abstract: This paper presents the European Crisis Management Ontology (ECMO), a modular OWL-based ontology intended as a cross-sectoral reference for disaster ris

arXiv:2610.00074v1 Announce Type: new Abstract: K-Dense BYOK (bring your own keys) is a free, open-source AI research assistant for scientists in any field that runs on the researcher's own computer.

arXiv:2609.38379v1 Announce Type: new Abstract: Large language models (LLMs) are frequently updated for various use cases, where filtering out misaligned training samples is a common practice for prev

arXiv:2609.38346v1 Announce Type: new Abstract: When a student is stuck, a tutor faces the assistance dilemma: help given too early can hinder productive struggle, while help withheld too long leaves

arXiv:2609.37501v1 Announce Type: cross Abstract: We propose RegLLM, a diagnostic harness for bounded autonomy in regulated agentic workflows. It instruments six trustworthiness signals: citation vali

arXiv:2609.36633v1 Announce Type: new Abstract: This paper proposes a physics-guided gradient-ascent-based machine unlearning method that couples the forgetting signal with the physical residual of th

arXiv:2609.34415v1 Announce Type: new Abstract: The transition from Large Language Models (LLMs) to agents shifts safety stakes from toxic text to irreversible environmental harm. While current defens

arXiv:2609.33923v1 Announce Type: cross Abstract: A single random Gaussian probe gives an unbiased estimate of the squared Frobenius norm of a layer's quantization error. The estimator is well-behaved

arXiv:2609.34179v1 Announce Type: cross Abstract: Retrieval-augmented generation systems are extensively instrumented with metrics, benchmarks, traces, and automated judges, but these tools do not dec

arXiv:2609.30334v1 Announce Type: cross Abstract: AI development agents are increasingly used to support and partially automate software architecture tasks. To explore how practitioners perceive this

arXiv:2609.29808v1 Announce Type: cross Abstract: In July 2026, an unconstrained autonomous agent participating in a frontier AI cybersecurity evaluation harness breached its evaluation sandbox, estab

arXiv:2609.30217v1 Announce Type: cross Abstract: A central concern in AI safety is that agents may treat oversight as an obstacle when it conflicts with completing their goals. We study instrumental

arXiv:2609.30266v1 Announce Type: cross Abstract: Asynchronous monitoring, incident investigations, and compliance audits primarily rely on agent traces to reconstruct what happened. These analyses as

arXiv:2609.29837v1 Announce Type: new Abstract: We introduce PUBG Ally, an embodied agent for PUBG: BATTLEGROUNDS that can reason, act autonomously, and play alongside players as a voice-enabled teamm

arXiv:2312.06632v2 Announce Type: replace Abstract: Artificial intelligence is rapidly advancing scientific discovery, but this progress carries risks of misuse, such as the creation of harmful substa

arXiv:2609.22869v1 Announce Type: cross Abstract: Technology cannot steer itself. Strategy provides that steering, establishing the rule that purpose and judgment must precede compute capital. As the

arXiv:2609.22949v1 Announce Type: cross Abstract: Existing prompt injection research focuses on single-model chatbot scenarios, where an attacker manipulates one LLM through crafted input. Multi-agent

arXiv:2609.22196v1 Announce Type: cross Abstract: We present EvoRank, an open autonomous ranking engineer: an LLM-guided evolutionary loop that discovers complete Learning-to-Rank pipelines (features,

arXiv:2609.21841v1 Announce Type: new Abstract: Frontier language models now produce professional deliverables that expert graders judge to match human work on a substantial share of economically valu

arXiv:2609.19203v1 Announce Type: new Abstract: AI applications have shifted from single, monolithic foundation models (FM) to compound agentic systems. Yet today's stacks remain fragmented: even as p

arXiv:2609.19472v1 Announce Type: new Abstract: Autonomous systems increasingly rely on Large Language Models (LLMs) yet the safety infrastructure surrounding these models introduces latency and compu

arXiv:2606.09735v2 Announce Type: replace Abstract: The ambition behind alignment training is to make large language models safe and useful. The primary mechanisms, reinforcement learning from human f

arXiv:2609.18135v1 Announce Type: cross Abstract: State-of-the-art Text-to-SQL systems are typically multi-agent pipelines centered around two fundamental tasks: schema linking and SQL generation. How

arXiv:2609.18605v1 Announce Type: cross Abstract: As corporate AI adoption continues to grow, enterprise-grade LLM agents are being deployed into sensitive contexts such as hiring, healthcare, and fin