AIBearisharXiv – CS AI · 2d ago7/10
🧠Researchers present MemPoison, a novel attack that exploits vulnerabilities in large language model agents by injecting malicious information into their long-term memory through dialogue interactions. The attack achieves up to 95% success rates by using semantic bridges, entity masquerading, and embedding optimization to bypass modern selective memory mechanisms, revealing critical security gaps in autonomous AI systems.
AIBullisharXiv – CS AI · 4d ago7/10
🧠Researchers demonstrate that tool-schema compression reduces token consumption by 44-50%, enabling large language model agents to function under tight context constraints. Testing across 14 models shows compressed schemas restore RAG functionality with +20.5 percentage point exact-match improvements at 8K tokens, while frontier models can now handle 800+ tools instead of ~494.
AIBearisharXiv – CS AI · May 127/10
🧠Researchers demonstrate 'Oracle Poisoning,' a novel attack where adversaries corrupt knowledge graphs used by AI agents, causing them to reach incorrect conclusions through valid reasoning. Testing across nine models from three providers shows all models accept fabricated data at 100% under moderate attack sophistication, revealing a critical vulnerability in production-scale agentic systems that differs fundamentally from prompt injection attacks.
🧠 GPT-5
AIBullisharXiv – CS AI · May 97/10
🧠Researchers propose constant-context skill learning, a framework enabling LLM agents to learn reusable task procedures as lightweight modules rather than storing long prompts in memory. The approach reduces token usage per inference by 2-7x while maintaining or improving performance across multiple benchmark environments, addressing the privacy-capability tradeoff in agent deployment.
🧠 Llama
AINeutralarXiv – CS AI · May 97/10
🧠Researchers introduce SkillRet, a large-scale benchmark dataset containing 17,810 public agent skills designed to evaluate how language model agents retrieve appropriate tools from massive skill libraries. The benchmark demonstrates that current retrieval methods struggle significantly with realistic large-scale deployments, though task-specific fine-tuning improves performance by up to 16.9 points on key metrics.
AIBullisharXiv – CS AI · May 47/10
🧠Researchers introduce A11y-Compressor, a framework that optimizes how AI agents interpret graphical user interfaces by transforming accessibility trees into more efficient representations. The approach reduces input tokens by 78% while simultaneously improving task success rates by 5.1 percentage points, addressing a critical bottleneck in GUI automation systems.
AIBullisharXiv – CS AI · Apr 147/10
🧠Researchers demonstrate that inference-time scaffolding can double the performance of small 8B language models on complex tool-use tasks without additional training, by deploying the same frozen model in three specialized roles: summarization, reasoning, and code correction. On a single 24GB GPU, this approach enables an 8B model to match or exceed much larger systems like DeepSeek-Coder 33B, suggesting efficient deployment paths for capable AI agents on modest hardware.
AIBullisharXiv – CS AI · Apr 107/10
🧠AgentOpt v0.1, a new Python framework, addresses client-side optimization for AI agents by intelligently allocating models, tools, and API budgets across pipeline stages. Using search algorithms like Arm Elimination and Bayesian Optimization, the tool reduces evaluation costs by 24-67% while achieving near-optimal accuracy, with cost differences between model combinations reaching up to 32x at matched performance levels.
AIBearisharXiv – CS AI · Apr 107/10
🧠Researchers have identified SkillTrojan, a novel backdoor attack targeting skill-based agent systems by embedding malicious logic within reusable skills rather than model parameters. The attack leverages skill composition to execute attacker-defined payloads with up to 97.2% success rates while maintaining clean task performance, revealing critical security gaps in AI agent architectures.
🧠 GPT-5
AIBullisharXiv – CS AI · Mar 177/10
🧠Researchers introduce Orla, a new library that simplifies the development and deployment of LLM-based multi-agent systems by providing a serving layer that separates workflow execution from policy decisions. The library offers stage mapping, workflow orchestration, and memory management capabilities that improve performance and reduce costs compared to single-model baselines.
AINeutralGoogle Research Blog · Jan 287/106
🧠The article discusses the scientific principles behind scaling agent systems in generative AI, examining the conditions and factors that determine when agent systems perform effectively. It appears to focus on understanding the theoretical foundations for building and deploying AI agent systems at scale.
AINeutralarXiv – CS AI · 2d ago6/10
🧠Researchers introduce LogDx-CI, a benchmark comparing 11 log-reduction tools for debugging CI failures using LLMs, finding that hybrid grep+tail routers achieve the best cost-quality tradeoff while agent-loop systems can recover from weak contexts through iterative tool calls, though at higher computational cost.
🏢 OpenAI🧠 GPT-5🧠 Claude
AINeutralarXiv – CS AI · 4d ago6/10
🧠Researchers introduce AgingBench, a longitudinal reliability benchmark that evaluates how AI agents degrade over time in production environments rather than just at deployment. The study reveals that agent reliability decays through four distinct mechanisms—compression, interference, revision, and maintenance aging—and that fixes must target specific failure stages rather than assuming stronger base models solve the problem.
AINeutralarXiv – CS AI · 4d ago6/10
🧠Researchers propose a formal framework for describing knowledge graph affordances to agents, extending decades-old semantic web service standards to address modern KG discovery and composition challenges. The framework introduces the Agentic Affordance Profile (AAP), a metadata layer that enables principled selection and failure diagnosis by specifying what agents can prove from a knowledge graph and under what epistemic conditions.
AINeutralarXiv – CS AI · May 116/10
🧠Researchers present a formal framework for recursive reasoning systems that addresses two critical design challenges: how to represent evolving reasoning states and when to terminate iteration. The paper introduces an epistemic state graph representation and proposes the 'order-gap' metric as a stopping criterion, with theoretical guarantees for when this criterion provides meaningful guidance.
AIBullisharXiv – CS AI · May 116/10
🧠AgentProg introduces a novel program-guided context management system for long-horizon GUI agents that addresses the critical bottleneck of expanding interaction history overhead. By reframing interaction history as structured programs with variables and control flow, the approach preserves semantic information while reducing context requirements, achieving state-of-the-art performance on AndroidWorld benchmarks while maintaining robustness on extended tasks.
AIBullisharXiv – CS AI · May 116/10
🧠WebClipper is a new framework that optimizes web agent trajectories by pruning redundant reasoning steps through graph-based analysis, reducing tool-call rounds by approximately 20% while maintaining or improving accuracy. The approach models agent search processes as directed acyclic graphs and introduces an F-AE Score metric to measure the balance between accuracy and efficiency in web agent design.
AIBullisharXiv – CS AI · Apr 156/10
🧠Aethon is a new systems primitive that enables stateful AI agents to be instantiated in near-constant time by using reference-based replication instead of full materialization. This architectural innovation addresses latency and memory overhead constraints in existing AI runtime systems, making it possible to spawn, specialize, and govern agents at production scale.
AINeutralarXiv – CS AI · Apr 146/10
🧠Researchers introduce STARS, a framework for continuously auditing AI agent skill invocations in real-time by combining static capability analysis with request-conditioned risk modeling. The approach demonstrates improved detection of prompt injection attacks compared to static baselines, though remains most valuable as a triage layer rather than a complete replacement for pre-deployment screening.
AIBullisharXiv – CS AI · Apr 76/10
🧠Researchers have released SuperLocalMemory V3.3, an open-source AI agent memory system that operates entirely locally without cloud LLMs, implementing biologically-inspired forgetting mechanisms and multi-channel retrieval. The system achieves 70.4% performance on LoCoMo benchmarks while running on CPU only, addressing the paradox of AI agents having vast knowledge but poor conversational memory.
AIBullisharXiv – CS AI · Mar 116/10
🧠Researchers present LLM Delegate Protocol (LDP), a new AI-native communication protocol for multi-agent LLM systems that introduces identity awareness, progressive payloads, and governance mechanisms. The protocol achieves 12x lower latency on simple tasks and 37% token reduction compared to existing protocols like A2A, though quality improvements remain limited in small delegate pools.
AIBullisharXiv – CS AI · Mar 37/108
🧠Researchers introduce AI Runtime Infrastructure, a new execution layer that sits between AI models and applications to optimize agent performance in real-time. This infrastructure actively monitors and intervenes in agent behavior during execution to improve task success, efficiency, and safety across long-running workflows.
AINeutralImport AI (Jack Clark) · Mar 26/1010
🧠Import AI 447 discusses the economic implications of artificial general intelligence (AGI), focusing on how most labor may shift to machines while humans transition to verification roles. The article explores the concept of the 'singularity' and its potential impact on the workforce and economy.
AINeutralarXiv – CS AI · Mar 34/104
🧠Researchers introduce Structured Diversity Control (SDC), a new framework for multi-agent reinforcement learning that improves coordination by controlling behavioral diversity within and between agent groups. The method achieved up to 47.1% improvement in average rewards and 12.82% reduction in episode lengths across various experiments.
AINeutralarXiv – CS AI · Mar 25/105
🧠Researchers present the Artificial Agency Program (AAP), a framework for developing AI systems as resource-bounded agents driven by curiosity and learning progress under physical constraints. The program aims to create AI that enhances human capabilities through better sensing, understanding, and action while reducing interface friction between people, tools, and environments.
