Knowledge catalogue
arXiv:2609.10221v2 Announce Type: new Abstract: Reinforcement learning over a frozen reasoner has become a common recipe for teaching a policy which external tools to invoke. We show that this recipe
arXiv:2609.06783v1 Announce Type: cross Abstract: LLM agents operate in workflows where unsafe actions can have real consequences. Existing safety evaluations often reduce behavior to a single score,
arXiv:2609.05576v1 Announce Type: new Abstract: The paradigm of LLMs has rapidly shifted from passive language interfaces to autonomous Claw-like agents that execute long-horizon tasks across stateful
arXiv:2609.06792v1 Announce Type: new Abstract: Android agents using a hybrid action space that combines GUI actions and tool actions (e.g., accessing application data via APIs) remain largely underex
arXiv:2609.10266v1 Announce Type: new Abstract: LLM serving systems already reuse KV caches, but only when the reused text sits at the very start of the prompt. Two growing workloads break this condit
arXiv:2508.05427v2 Announce Type: replace Abstract: Large language models (LLMs) are beginning to reshape how organic-synthesis workflows are represented, queried, planned, and connected to experiment
arXiv:2609.05531v1 Announce Type: cross Abstract: No specification says how a governed autotelic AI agent organization, where agents pursue self-generated goals inside guardrails, should be designed a
arXiv:2609.08162v1 Announce Type: new Abstract: How can vision-language-action (VLA) models adapt to new environments where world dynamics shift? While recent research has combined world modeling and
arXiv:2609.07154v1 Announce Type: new Abstract: We describe our entry to the EgoLongQA track of the Wearable-AI Challenge in ECCV 2026, which placed first in the <=2B parameter division with 0.8279 on
arXiv:2609.07529v1 Announce Type: new Abstract: Tool-augmented language agents are vulnerable to indirect prompt injection (IPI). Unlike direct prompt injection, IPI hides adversarial instructions in
arXiv:2609.07747v1 Announce Type: new Abstract: Human videos are an abundant source of dexterous manipulation behaviors, but they lack tactile information that is crucial for contact-rich interaction.
arXiv:2609.06006v1 Announce Type: new Abstract: Deep learning for time series has progressed through successive architectural paradigms, from recurrent networks and transformers to structured state-sp
arXiv:2609.06100v1 Announce Type: new Abstract: Verifiable outcome rewards guide language-model post-training, but sequence-level advantages do not identify which token-level decisions should be prese
arXiv:2609.04286v1 Announce Type: new Abstract: Artificial intelligence in recruitment has shifted the object being automated from profile pairs and ranked lists to multi-stage workflows that retrieve
arXiv:2609.04495v1 Announce Type: new Abstract: We formulate indirect prompt injection as a test-time search over a task-dependent attack surface induced by the environment, user task, and injection t
arXiv:2609.04748v1 Announce Type: cross Abstract: Prefix caching, in which a serving engine reuses the key and value tensors of a shared prompt prefix across requests, is enabled by default in the maj
arXiv:2609.04034v1 Announce Type: cross Abstract: While diffusion base models such as GPT-Image-2 and Nano-Banana exhibit remarkable visual expressiveness, their end-to-end generation inherently yield
arXiv:2609.03960v1 Announce Type: new Abstract: Banks need conversational systems that can answer product questions, assist customers with account-related requests, and operate safely within strict op
arXiv:2609.03727v1 Announce Type: new Abstract: Large language model agents can plan, invoke tools, and modify external states, yet most systems still take an explicit user instruction as a fixed star
arXiv:2609.03620v1 Announce Type: cross Abstract: Audio deepfake detection is commonly formulated as clip-level binary classification of single-domain audio. However, real-world manipulated audio can
arXiv:2609.01861v1 Announce Type: new Abstract: The performance of an LLM agent depends on the scaffold around a frozen model. A common way to improve that scaffold is to use a coding agent as an opti
arXiv:2609.02783v1 Announce Type: new Abstract: Evaluating LLM agents is essential for guiding their development, yet it has grown prohibitively expensive: a single pass of a frontier model over an ag
arXiv:2609.01660v1 Announce Type: cross Abstract: Production deployments of large language model (LLM) agents remain unreliable on long, multi-step workflows even as benchmark success rates climb stea
arXiv:2609.02548v1 Announce Type: cross Abstract: Modern large language models (LLMs) rely on reinforcement learning to build strong capabilities in individual domains, but integrating those capabilit
arXiv:2609.02095v1 Announce Type: new Abstract: An AI agent can perform well on benchmarks and still be unsuitable for deployment. Existing AI-agent benchmarks measure whether an agent can complete re
arXiv:2609.02421v1 Announce Type: new Abstract: Large Language Models (LLMs) have recently demonstrated strong capabilities in automated travel itinerary generation. However, real- world travel planni
arXiv:2609.01289v1 Announce Type: new Abstract: Hyperspectral unmixing is a key task in remote sensing that aims to decompose mixed pixels in hyperspectral images into their constituent material signa
arXiv:2609.01286v1 Announce Type: new Abstract: Sharing analog integrated circuit designs remains difficult: foundry non-disclosure agreements restrict the process details a design depends on, and the
arXiv:2609.01195v1 Announce Type: new Abstract: Entity matching (EM) requires fine-grained contextual understanding and domain knowledge. Recent work shows that large language models (LLMs) can serve
arXiv:2609.01281v1 Announce Type: cross Abstract: Vision-language-action (VLA) models map visual observations and language instructions directly to robot actions, but long-horizon tasks require more t
arXiv:2609.01245v1 Announce Type: cross Abstract: Reinforcement learning is a natural way to post-train LLM agents for long-horizon interactive tasks judged only by end-of-task verification, yet a sha
arXiv:2510.15144v4 Announce Type: replace Abstract: Simulating human reasoning in open-ended tasks has long been a central aspiration in AI and cognitive science. While large language models now appro
arXiv:2609.00865v1 Announce Type: cross Abstract: Production agent harnesses such as Claude Code and Qwen-Agent compress context during rollout, but training under compression creates a conditioning p
arXiv:2609.00384v1 Announce Type: new Abstract: Large Language Model (LLM) agents increasingly automate multi-step engineering workflows through tool use, interpretation of intermediate results, and i
arXiv:2609.00549v1 Announce Type: new Abstract: Large Language Model (LLM) agents increasingly rely on external skills, yet standard evaluations obscure whether retrieving these skills actually helps.
arXiv:2609.00038v1 Announce Type: cross Abstract: Outcome-only evaluation is the production default for LLM agents: show a judge the request and the final reply and ask whether it was handled well. Th
arXiv:2609.00028v1 Announce Type: new Abstract: Multimodal GUI agents have emerged as a promising paradigm for digital task automation, yet transitioning from benchmark-oriented models to dependable r
arXiv:2608.29606v1 Announce Type: new Abstract: Large language model (LLM) agents need durable, faithful memory of everything a user or organization has said and stored, yet most memory systems commit
arXiv:2601.12449v2 Announce Type: replace-cross Abstract: AI agents are autonomous systems that combine LLMs with external tools to solve complex tasks. While such tools extend capability, improper to
arXiv:2608.30092v1 Announce Type: cross Abstract: We present Arkios, a 1.04B-parameter dense transformer pretrained from scratch on 150B tokens of bilingual English-Nepali text, using a custom single-
arXiv:2608.30022v1 Announce Type: new Abstract: Introduction: NICE guidelines provide evidence-based recommendations for clinical care but remain largely in unstructured natural language. Existing app
arXiv:2608.28725v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used as graders, verifiers, and process auditors, but most mathematical evaluations still emphasize final-
arXiv:2608.29309v1 Announce Type: cross Abstract: This work provides an overview of the different strategies that can be used to evaluate the performance of AI models and agents based on large languag
arXiv:2509.23263v3 Announce Type: replace Abstract: Long-horizon GUI automation remains challenging due to error accumulation over extended interaction sequences. Process Reward Models (PRMs) provide
arXiv:2608.30426v1 Announce Type: new Abstract: Current dialogue systems struggle with dynamic information retrieval, often leading to hallucinations and lower response accuracy. We address this by ad
arXiv:2608.31057v1 Announce Type: new Abstract: Agent working memory is heterogeneous. Objects such as instructions, artifacts, tool outputs, and agent-generated state play different semantic roles an
arXiv:2608.31077v1 Announce Type: new Abstract: Outcome-based reinforcement learning provides verified feedback for language-model agents, but assigns trajectory-level advantage uniformly to all decis
arXiv:2606.15872v2 Announce Type: replace Abstract: Frontier scientific reasoning remains a major challenge for large language models (LLMs), where even the strongest commercial systems fall short of
arXiv:2608.29363v1 Announce Type: new Abstract: Enterprise data agents answer business queries by chaining many tool calls over multiple reasoning steps, routinely accumulating hundreds of thousands o
arXiv:2608.30705v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) struggle with fine-grained Visual Search, the task of locating small or rare objects in high-resolution images.
arXiv:2608.27924v1 Announce Type: new Abstract: Reliable handling of unanswerable questions (UAQs) is critical for trustworthy LLM-based agents. Although memory is widely used in agent systems, its ro
arXiv:2608.27768v1 Announce Type: cross Abstract: A language model with access to tools can commit to a final claim unsupported by the evidence it has seen, even when a single available tool call woul
arXiv:2608.26198v1 Announce Type: new Abstract: Operating a scientific instrument such as an atomic force microscope (AFM) requires continuous expert decision-making. A trained user defines the experi
arXiv:2608.26623v1 Announce Type: new Abstract: LLM judges are widely used to evaluate agentic tool-calling systems, yet their reliability on structured, dependency-driven workflows remains largely un
arXiv:2608.26329v1 Announce Type: new Abstract: While tool-augmented Large Language Models have significantly improved multi-step reasoning in quantitative STEM tasks, a critical residual failure mode
arXiv:2606.07867v2 Announce Type: replace Abstract: Are tool-calling LLM agents equally safe throughout a conversation? We discover they are not: agents are most vulnerable at the very start of a sess
arXiv:2601.13029v4 Announce Type: replace Abstract: While Vision-Language Models (VLMs) excel at 2D visual understanding, they remain constrained by 2D-centric paradigm that severely limits genuine 3D
arXiv:2608.24076v1 Announce Type: new Abstract: Evaluation of agentic information retrieval remains limited to scripted interactions with uniform users, missing both natural personality diversity and
arXiv:2608.24099v1 Announce Type: new Abstract: GUI agents often encounter dynamic anomalies when deployed on Android devices, from unexpected pop-ups to action misuse, yet existing benchmarks lack sy
arXiv:2608.24368v1 Announce Type: new Abstract: Reliable multi-turn tool use requires an agent to preserve an evolving task state and ensure that each action remains consistent with it. However, direc