Knowledge catalogue
excited to be speaking at @aiDotEngineer World Fair next week on Improving Agents, Continual Learning, and why we think a large part of it is...Data Mining! Trace Mining is how we understand agent beh
arXiv:2606.21811v1 Announce Type: cross Abstract: End-to-end code agent training is resource-intensive and plateaus on the strategy-level reasoning needed to resolve code issues, since jointly optimiz
arXiv:2606.11447v1 Announce Type: new Abstract: Recent anecdotal evidence suggests that AI coding agents can reproduce published findings when provided with original data and code; yet systematic eval
arXiv:2606.11349v1 Announce Type: new Abstract: In hierarchical reasoning, failures often originate at intermediate decision points where the agent commits to a wrong branch without recognizing that i
arXiv:2606.11680v1 Announce Type: new Abstract: Large language model (LLM) agents struggle with long-horizon tasks due to their inherent statelessness, requiring all task-relevant information to be en
arXiv:2606.08106v1 Announce Type: new Abstract: Self-evolving agents improve by repeatedly proposing changes to their own prompts, skills, or workflows and keeping those that score higher on a small h
arXiv:2606.07682v1 Announce Type: cross Abstract: AI agents are increasingly expected to complete long-horizon workflows that require sustained progress over hours, millions of tokens, and complex env
arXiv:2606.06787v1 Announce Type: new Abstract: Large Language Models (LLMs) show promise as tool-using agents but remain limited in long-horizon tasks that require remembering, organizing, and reusin
At @tryramp, engineers use Devin Desktop to bring their favorite agents into one place. With Devin Desktop they can dispatch, monitor, and jump between agents from a single surface with shared context
arXiv:2606.06923v1 Announce Type: new Abstract: We study orchestration mechanisms for tool-using AI agents in realistic customer-service workflows over an unstructured knowledge base. We argue that de
arXiv:2606.07017v1 Announce Type: new Abstract: Foundation model agents are increasingly deployed for real-world decision-making, but suffer from the sim-to-real gap. While robotics and classical cont
arXiv:2606.05684v1 Announce Type: new Abstract: A central challenge for language agents is utilizing past experience to adapt to dynamic test-time conditions. While recent work demonstrates the promis
arXiv:2606.06090v1 Announce Type: new Abstract: LLM-based agents increasingly tackle long-horizon tasks with interdependent decisions, where each action reshapes future constraints and intermediate er
This post explores the concept of using ActiveGraph with event sourcing for AI agents operating under EU regulations, suggesting a potential architecture where detailed logging and event histories ser
arXiv:2606.05922v1 Announce Type: cross Abstract: AI agents rely on a harness of skills, tools, and workflows to solve complex problems. Continually improving this harness is essential for adapting to
arXiv:2606.04197v1 Announce Type: cross Abstract: How much should an LLM agent remember, and how should multi-agent systems be connected when trying to reach consensus? We show these two design choice
arXiv:2606.04627v1 Announce Type: new Abstract: Mobile agents are increasingly expected to operate everyday applications from screenshots and language goals, where reliable control requires reasoning
arXiv:2606.03054v1 Announce Type: new Abstract: Tool-augmented vision-language agents can acquire external perceptual evidence through OCR, detection, segmentation, and other tools, but executing ever
arXiv:2606.02908v1 Announce Type: cross Abstract: Multi-turn user-facing agents must infer user intent from incomplete requests, collect missing information through dialogue and tools, and execute val
arXiv:2606.01567v1 Announce Type: cross Abstract: Large language model (LLM) agents increasingly rely on reusable skills i.e. documents describing task-specific procedures. However, this introduces a
arXiv:2606.02357v1 Announce Type: cross Abstract: Tool-augmented multimodal agents show strong benchmark gains, often taken as evidence that agents have learned to use tools. We argue that this interp
Introducing Devin Desktop: the next generation of Windsurf Manage fleets of local and cloud agents from one surface Support for any ACP-compatible agent With a full IDE for when you need to jump into
arXiv:2606.01993v1 Announce Type: cross Abstract: Abundant procedural knowledge on the Web holds great potential for helping agents solve long-horizon tasks. However, such knowledge is often multimoda
Connecting agents directly to Snowflake, Slack, or BigQuery via MCP/CLIs is a multi-token disaster. They default to brute-force exploration, firing 20-30 tool calls just to rediscover context on every
🧑⚖️Evaluating Deep Agents with LangSmith on AWS Great deep dive blog with our friends at AWS on evaluating DeepAgents with LangSmith Covers datapoint and evaluator design for longer horizon agents ht
Learn more about the latest from @james_y_zou and our Frontier Agents Research team! To evaluate frontier AI agents, we need more complex tasks. But such tasks are also more prone to have design mista
arXiv:2605.30144v1 Announce Type: new Abstract: Despite the rapid deployment of LLMs into classrooms, validating educational AI remains uniquely intractable: interventions act on developing learners w
arXiv:2605.29927v1 Announce Type: cross Abstract: Despite recent advances, LLM-based web agents still struggle with limited exploration, omission of critical steps, and sensitivity to task constraints
make something agents want studying ActiveGraph by @yoheinakajima and apart from the concept the implementation itself is brilliant on the site it shares a prompt that will make my agent study docs, i
arXiv:2605.22166v2 Announce Type: replace Abstract: LLM agents are shaped not only by their language models, but also by the runtime harness that mediates observation, tool use, action execution, feed
arXiv:2605.28108v1 Announce Type: new Abstract: A long-lived LLM agent, such as OpenClaw, earns its value by acting on a user's preferences and constraints across sessions, not just the current reques
arXiv:2605.27820v1 Announce Type: new Abstract: As AI agents increasingly operate in open, real-world environments, they require a deep synergy of multimodal perception, tool invocation with multi-hop
arXiv:2605.28534v1 Announce Type: new Abstract: Despite the rapid progress of multimodal large language models in building Graphical User Interface (GUI) agents, their real-world task completion is fu
arXiv:2509.04310v4 Announce Type: replace Abstract: Recent research on Chain-of-Thought (CoT) reasoning in Large Language Models (LLMs) has demonstrated that agents can engage in extit{complex}, extit
oh yeah first* automations - instead of dumb crons kicking of agents, devin has made them smart. try it - it is the first non annoying proactive agent impl ive seen https://x.com/russelljkaplan/status
arXiv:2605.27068v1 Announce Type: cross Abstract: Social deduction games have become a popular testbed for probing reasoning, deception, coordination, and belief modeling in Large Language Model (LLM)
Zero-trust cybersecurity company Xage Security Inc. today unveiled new capabilities in its platform designed to give enterprises deterministic visibility into autonomous artificial intelligence agents
arXiv:2605.23590v1 Announce Type: new Abstract: ReAct-style agents for search-intensive, multi-step reasoning tasks rely largely on their own internal judgment to decide what evidence to seek, which r
/goal is really insane! It's how you can get the most out of coding agents today. For efficiency, I find it works best when you do planning before /goal. This ensures the agent has the right context a
Agents shouldn’t have direct visibility into env vars or credentials that can expose sensitive systems and data. Keeping secrets outside the agent’s context helps secure the env while still allowing a
Agents are too good at sounding right. And terrible at proving what actually happened, especially on complex, long-running tasks. So I built a fully traceable and forkable research agent with @yoheina
arXiv:2605.20616v1 Announce Type: new Abstract: Language agents increasingly operate over streams of related tasks, yet existing memory systems struggle to convert accumulated experience into reusable
This post references a previous article discussing stateful agents broadly, which served as inspiration for Yohei Nakajima's project. The post likely documents how earlier discourse on agent architect
Juicebox App Inc. today introduced a suite of autonomous recruiting agents that help hiring teams proactively identify and engage candidates across multiple open roles simultaneously, recommending pro
ICYMI: LangSmith Sandboxes are GA ✅ Agents get a real filesystem, shell, and package manager. Isolated from your infra. ✅ Works with Deep Agents, Open SWE, or your own code. ✅ Auth with the same API k
we've been building agents around the llm, starting with conversations, adding tools, giving rules, logging everything, and storing a form of it as retrievable state Active Graphs flips this. what if
arXiv:2605.16265v1 Announce Type: new Abstract: The safety of autonomous AI agents is increasingly recognized as a critical open problem. As agents transition from passive text generators to active ac
arXiv:2605.18048v1 Announce Type: new Abstract: While Graphical User Interface (GUI) agents have shown promising performance in automated device interaction, they primarily depend on static parametric
arXiv:2605.18684v1 Announce Type: cross Abstract: Legacy systems concentrate business rules, architectural decisions, and operational exceptions that often remain implicit in code, data, configuration
Chris Metinko / Axios: Dust, which helps enterprises design and deploy specialized AI agents that work alongside humans, raised a 40M Series B led by Abstract and Sequoia — Dust, a human-agent collabo
arXiv:2605.16045v1 Announce Type: cross Abstract: Memory systems often organize user-agent interactions as retrievable external memory and are crucial for long-running agents by overcoming the limited
LangSmith Engine feels like the missing CI/CD loop for AI agents , automatically detecting failures, clustering issues, proposing fixes, and generating evals from production traces. Agent engineering
This quote from our friends at @cogent_security says a lot: “At Cogent, our background agents can produce a huge volume of traces all at once. We need live observability into those systems, and SmithD
arXiv:2605.12887v1 Announce Type: cross Abstract: Web-enabled LLM agents are changing how online information influences search outcomes. Existing Generative Engine Optimization (GEO) studies mainly fo
arXiv:2605.12913v1 Announce Type: new Abstract: Long-horizon LM agents learn from multi-turn interaction, where a single early mistake can alter the subsequent state distribution and derail the whole
arXiv:2605.11732v1 Announce Type: cross Abstract: In this paper, we present AgentDisCo, a novel Disentangled and Collaborative agentic architecture that formulates deep research as an adversarial opti
Gyana Swain / CSO: Microsoft unveils MDASH, a security system that orchestrates 100+ AI agents to find vulnerabilities, and says it identified 16 previously unknown Windows flaws — The agentic tool, c
arXiv:2605.10286v1 Announce Type: new Abstract: Building effective clinical decision support systems requires the synthesis of complex heterogeneous multimodal data. Such modalities include temporal e
arXiv:2605.08463v1 Announce Type: new Abstract: Autonomous AI agents are increasingly deployed in open social environments, yet the relationship between their configuration specifications and their em
Excited to have @Box in the Interrupt lineup! CEO Aaron @Levie will take the stage with @hwchase17 to discuss agents in the enterprise. If you’re attending our agent conference, don’t miss their sessi