Knowledge catalogue
arXiv:2605.17986v1 Announce Type: cross Abstract: AI agents such as OpenClaw are increasingly deployed in local workflows with access to external tools. This creates indirect prompt-injection (IPI) ri
arXiv:2605.17596v1 Announce Type: new Abstract: We present NeuSymMS, an adaptive memory system that enables large language model (LLM) agents to learn, remember, and reason about users across sessions
arXiv:2603.23231v2 Announce Type: replace Abstract: Empowering large language models with long-term memory is crucial for building agents that adapt to users' evolving needs. Existing evaluations of t
Unity AI Gateway introduces new enterprise features including service policies, guardrails, observability tools, and cost controls designed to manage and monitor AI agents and Model Context Protocols
continuing my HTML era, I had so much fun talking with Claire at Code w/ Claude about staying in the loop with long running agents Soooo @trq212 has straight up changed my life with these 5 words: 'HT
arXiv:2605.15229v1 Announce Type: cross Abstract: Existing code benchmarks measure whether an agent can produce any test that reproduces a known bug, or whether it can produce a patch that fixes a des
🚨Breaking new study: memory in LLM agents still can’t really be trusted, even after over trillion dollars has gone into the development of the field. Excited to share our new paper: “Useful Memories B
Yohei Nakajima discusses the lack of elegant primitive implementations for stateful agents, decision traces, and context graphs, which are frequently discussed concepts in AI but rarely demonstrated i
40 Grok Build agents tearing through C code in parallel. All supervised by DAD. DAD is a lightweight autonomous tmux supervisor for long-running Grok tasks. Built entirely in Grok Build. /dad 'your ob
SaaS applications are being transformed into deterministic engines as AI agents replace traditional user interfaces and drive the rise of the headless enterprise — where AI, not humans, serves as the
arXiv:2605.14859v1 Announce Type: cross Abstract: As coding agents gain access to shells, repositories, and user files, least-privilege authorization becomes a prerequisite for safe deployment: an age
arXiv:2605.14057v1 Announce Type: new Abstract: Most existing dialogue systems are user-driven, primarily designed to fulfill user requests. However, in many critical real-world scenarios, a conversat
arXiv:2605.14034v1 Announce Type: new Abstract: Wide applications of LLM-based agents require strong alignment with human social values. However, current works still exhibit deficiencies in self-cogni
arXiv:2605.13841v1 Announce Type: cross Abstract: Voice agents, artificial intelligence systems that conduct spoken conversations to complete tasks, are increasingly deployed across enterprise applica
arXiv:2603.27910v2 Announce Type: replace Abstract: AI agents that interact with users across multiple sessions require persistent long-term memory to maintain coherent, personalized behavior. Current
arXiv:2605.13228v1 Announce Type: cross Abstract: Video understanding requires active evidence seeking, motivating tool-augmented video agents for temporal reasoning, cross-modal understanding, and co
Swyx is giving a talk at AI Engineer Singapore on Saturday at 4pm about using reinforcement learning for long-running agents and reinforcement learning models (RLMs), exploring why this approach shoul
arXiv:2605.11853v1 Announce Type: cross Abstract: Reinforcement learning has become a widely used post-training approach for LLM agents, where training commonly relies on outcome-level rewards that pr
arXiv:2605.11212v1 Announce Type: new Abstract: Computer-use agents~(CUAs) rely on visual observations of graphical user interfaces, where each screenshot is encoded into a large number of visual toke
🎉 We published a new AI safety study: shopping agents fall for whimsical attacks and lose money. A whimsical attack is an absurd scenario a human would never try on another human. In one run, GPT-5.1
arXiv:2604.08178v2 Announce Type: replace Abstract: In classical Reinforcement Learning from Human Feedback (RLHF), Reward Models (RMs) serve as the fundamental signal provider for model alignment. As
arXiv:2605.08399v1 Announce Type: new Abstract: Tool-augmented language models can extend small language models with external executable skills, but scaling the tool library creates a coupled challeng
arXiv:2605.09675v1 Announce Type: new Abstract: Clinical reasoning agents based on large language models (LLMs) aim to automate tasks such as intensive care unit (ICU) monitoring and patient state tra
arXiv:2605.08747v1 Announce Type: new Abstract: Standard embodied evaluations do not independently score whether an agent correctly commits to task completion at episode closure, a capacity we call te
arXiv:2605.10366v1 Announce Type: new Abstract: Graph reasoning agents operating from natural-language inputs must solve a coupled problem: they must reconstruct a structured graph instance from text,
arXiv:2605.09131v1 Announce Type: new Abstract: The Model Context Protocol (MCP) has unified the interface between Large Language Models (LLMs) and external tools, yet a fundamental gap remains in how
arXiv:2605.09530v1 Announce Type: cross Abstract: As LLM-powered agents are increasingly deployed in edge-cloud environments, personalized memory has become a key enabler of long-term adaptation and u
arXiv:2603.13131v3 Announce Type: replace Abstract: Long-horizon embodied intelligence requires agents to improve through interaction, not merely to execute plans generated from static goals. A centra
arXiv:2510.15995v3 Announce Type: replace-cross Abstract: We study overpricing in a repeated game between two representative agents: a market maker, who controls market liquidity, and a market taker,
arXiv:2605.09692v1 Announce Type: new Abstract: Unpredictable behavior is often taken as evidence of control, yet stochastic dispersion and structured action control need not coincide. This paper test
+1 to this. I was recently on a cross-continental flight without wifi, so I brought up Qwen3.6 & Gemma 4 (via @ollama) in Deep Agents on my laptop. admittedly, they fell over on some more involved/com
arXiv:2605.06607v2 Announce Type: replace-cross Abstract: Recent LLM-based agents have closed substantial portions of the scientific discovery loop in software-only machine-learning research, in chemi
Excellent explainer video by @FryRsquared on the risks of AI agents. She also raises a crucial point: we shouldn’t make the mistake of thinking current limitations will necessarily persist. As we’ve s
arXiv:2605.06957v1 Announce Type: new Abstract: We present a dynamic policy-learning approach that combines generalized planning and hierarchical task decomposition for LLM-based agents. Our method, H
// LLMs Improving LLMs // Interesting progress the past of couple of weeks around self-improving AI agents. If autoresearch was interesting, you will like this read. (bookmark it) We've been hand-tuni
arXiv:2605.06130v2 Announce Type: replace Abstract: A persistent skill library allows language model agents to reuse successful strategies across tasks. Maintaining such a library requires three coupl
Using SocialReasoning Bench, we observed a stable pattern across models—agents execute competently, but fail to consistently improve the user’s position, even with explicit instructions to optimize fo
arXiv:2605.08060v1 Announce Type: cross Abstract: Context window expansion is often treated as a straightforward capability upgrade for LLMs, but we find it systematically fails in multi-agent social
RT Watson / The Block: AWS unveils Amazon Bedrock AgentCore Payments and partners with Coinbase and Stripe to enable AI agents to execute transactions using stablecoins — Quick Take — Amazon Web Servi
CodeWords, operated by Agemo AI Ltd., today announced it raised 9 million in seed funding led by Visionaries to build artificial intelligence agents that don’t wait to build automations. Firstminute C
arXiv:2605.03675v1 Announce Type: new Abstract: Long-running autonomous AI agents suffer from a well-documented memory coherence problem: tool-execution success rates degrade 14 percentage points over
Terrence O'Brien / The Verge: Spotify launches Save to Spotify, a command-line tool that allows AI agents to upload AI-generated audio summaries and personal podcasts to a user's account — A new comma
Anna Tong / Forbes: Tessera Labs, which uses AI agents to automate enterprise IT migrations and ERP transformations, raised a 60M Series A led by a16z at a 320M valuation — Kabir Nagrecha is tackling
arXiv:2605.03482v1 Announce Type: cross Abstract: Persistent external memory enables LLM agents to maintain context across sessions, yet its security properties remain formally uncharacterized. We for
Cloud project management provider monday.com Ltd. today relaunched itself as an “AI work platform,” repositioning itself around context-aware artificial intelligence agents that execute tasks alongsid
really cool work by the Harvey team, excited to partner with them to push forward research on designing + understanding agents across Long Horizon Legal work the first peak I got at LAB was 'woah this
arXiv:2605.02964v1 Announce Type: new Abstract: Reinforcement learning (RL) trained language model agents with tool access are increasingly deployed in coding assistants, research tools, and autonomou
arXiv:2605.01555v1 Announce Type: new Abstract: We introduce an autonomous multiagent framework for mechanistic interpretability that automates both explaining and finding internal features in large l
Cursor has introduced an automated CI failure resolution feature that uses always-on agents to monitor GitHub repositories, analyze build failures, and automatically generate pull requests with fixes.
arXiv:2605.00007v1 Announce Type: cross Abstract: Independent sample generation is the prevailing paradigm in modern diffusion-based generative models of AI. We ask a different question: can samples c
Enterprise AI has learned to generate. It has learned to reason. Now companies are asking the next question: How should AI act? Early agent systems have shown what’s possible, moving beyond simple pro
arXiv:2507.14201v3 Announce Type: replace-cross Abstract: We present ExCyTIn-Bench, the first benchmark to Evaluate an LLM agent X on the task of Cyber Threat Investigation through security questions
Box announced its participation in Pinecone's Nexus launch, emphasizing that enterprise AI agents require contextual access to organizational data and content. The partnership highlights the integrati
Sam Altman's post argues that the Agents SDK 2.0 is underappreciated or overlooked despite its capabilities and potential value. The post likely discusses features, improvements, or use cases of the A
We have released a little skill to help you build LLM Wikis with your agents. Releasing a little skill to help build LLM Wikis. More people should be building LLM wikis. I hope the skill is useful for
This article discusses the emerging landscape of AI agents specialized for different domains, contrasting Codex's capabilities for knowledge work and structured tasks with Claude's strengths in creati
arXiv:2604.27253v1 Announce Type: new Abstract: Recent advances in multimodal large language models (LLMs) have revolutionized web agents that can automate complex tasks on websites. However, their ac
arXiv:2604.27699v1 Announce Type: new Abstract: Current embodied agents are often limited to passive instruction-following or reactive need-satisfaction, lacking a stable, high-order value framework e
arXiv:2601.02845v2 Announce Type: replace-cross Abstract: Long-horizon conversational agents have to manage ever-growing interaction histories that quickly exceed the finite context windows of large l
arXiv:2603.16496v2 Announce Type: replace Abstract: Large language model (LLM) agents increasingly rely on external memory to support long-horizon interaction, personalized assistance, and multi-step