Tool use, function calling, orchestration and the protocols that let models act rather than only answer.
arXiv:2610.34759v1 Announce Type: new Abstract: Effective collaboration requires language models to understand what others know and believe, and to act accordingly. Models that assume others share th...
arXiv:2605.11212v4 Announce Type: replace Abstract: Computer-use agents (CUAs) rely on visual observations of graphical user interfaces, where each screenshot is encoded into a large number of visual...
arXiv:2610.15910v2 Announce Type: replace Abstract: Deep Research agents rely on long-horizon reasoning and iterative tool use to solve complex information-seeking tasks. However, adaptive decision-m...
arXiv:2610.33019v1 Announce Type: new Abstract: As multimodal agents engage in sustained interactions, long-term memory serves two distinct but complementary functions: fine-grained evidence recall a...
arXiv:2610.33758v1 Announce Type: new Abstract: Road infrastructure perception is a fundamental component of autonomous driving, as it recovers the structured road elements that define roadway geomet...
arXiv:2610.34085v1 Announce Type: cross Abstract: Safe indoor operation requires autonomous tracking of objects that are not directly visible to the robot. Optical non-line-of-sight (NLoS) methods de...
arXiv:2605.13803v2 Announce Type: replace Abstract: Video temporal grounding (VTG) takes an untrimmed video and a natural-language query as input and localizes the temporal moment that best matches t...
The paper introduces Interaction-Effect Propagation (IE-Prop), a model that estimates error risk in large language model (LLM) based multi‑agent systems by conditioning on whether upstream agents are correct or incorrect. IE‑Prop learns separate effects for each sender‑receiver interaction, combines them using the sender’s estimated error risk, and recursively propagates these risks through the execution graph. Experiments on three architectures and benchmarks show that IE‑Prop improves final‑node AUROC by about 0.05 points over baselines and that multi‑step risk tracking adds an additional 0.03 AUROC in sequential systems.
FraudSociety is a novel LLM‑agent‑based simulator that jointly models informational and financial dynamics in the financial‑fraud economy. It features a dynamic social environment, persistent financial ledgers, and a Need‑Intent‑Action architecture, allowing fraudsters to adapt their actions based on interaction histories and ledger states. Validation shows macro‑level outcomes match real‑world patterns, micro‑level loss concentration aligns with Canadian statistics, and personalized interventions can dramatically reduce repeat‑payment rates.
The paper discusses how instruction hierarchy (IH) can be improved by training large language models on simulated instruction conflicts, particularly a newly identified type called goal‑goal conflicts. By fine‑tuning Qwen3.5‑9B and Gemma4‑12B‑it on examples of these conflicts, the authors demonstrate significant gains in IH performance across a variety of real‑world tasks, while maintaining negligible impact on non‑conflict tasks. The study suggests that incorporating goal‑goal conflicts into IH training sets can enhance model adherence to higher‑priority instructions.
Large Language Models: A Plain-Language Overview for Engineers explains that LLMs are neural networks trained on vast text collections to predict subsequent words, and outlines key concepts such as Transformers, pre‑training, fine‑tuning, reinforcement learning from human feedback, retrieval‑augmented generation, chain‑of‑thought prompting, and tool‑using agents. It also discusses efficiency techniques like FlashAttention and QLoRA, provides a life‑cycle figure and comparative tables, and offers a checklist for evaluating LLM claims, including checks for data contamination, model‑based grading, long‑context handling, and hallucination. The article aims to give engineers a clear picture of how training changes and prompts affect generative AI tools like ChatGPT.
The report evaluates the GPT-6-Astra multimodal large language model (MLLM) on the EBench platform, comparing two agent configurations: direct control, where the agent generates waypoint commands, and harness with policy, where it can invoke post‑trained vision‑language‑action (VLA) or world action model (WAM) policies. Direct control performs competitively overall, excelling on short‑horizon tasks and showing stable generalization, but falls behind on long‑horizon, tabletop, and high‑precision contact‑rich tasks. The harness with policy approach combines the strengths of both components, achieving success rates comparable to the best post‑trained policies while reducing token cost, yet still lags behind the standalone upper bounds due to suboptimal policy invocation and execution issues.
LiveOption is an evaluation framework designed to assess large language model (LLM) agents in option trading, a domain characterized by nonlinear payoffs and multi-leg strategy construction. The framework treats option trading as structured sequential decision-making under realistic execution and capital constraints, offering a reproducible environment with standardized interaction protocols. It includes three task suites—portfolio overlays, event-driven earnings trading, and 0DTE intraday trading—and introduces a hierarchical metric suite that evaluates action validity, decision quality, risk characteristics, and outcome-level performance, revealing that current agents often fail to achieve competitive returns.
AOS-Bench evaluates large language model agents as dynamic workload managers, testing their ability to handle multiple concurrent tasks under shared CPU and memory limits. The benchmark shows that isolated task success does not predict performance in a streaming workload, with significant drops in heavy and medium scenarios. Execution‑policy experiments reveal that simple concurrency controls can rival more complex management strategies, highlighting the importance of workload management as a distinct agent capability.
The paper introduces Iris, a framework that extends typed decision‑making models to multimodal settings where candidates can be images, text, or hybrids. Iris preserves key properties such as candidate‑order invariance, scalability, and cache reuse, and achieves high accuracy on a visual‑candidate benchmark while reducing latency through efficient encoding strategies. The authors also demonstrate significant speedups on GPU systems and outperform larger baseline models on MetaEncoder benchmarks.
arXiv:2609.22688v2 Announce Type: replace-cross Abstract: Parametric CAD modeling requires not only matching the target shape, but also preserving design intent after parameter editing. However, most...
arXiv:2610.34365v1 Announce Type: cross Abstract: Science agents increasingly package domain expertise as reusable skills that plan analyses, select tools, and execute workflows. This modularity crea...
arXiv:2610.34817v1 Announce Type: cross Abstract: Large language model (LLM) services process massive volumes of user requests, resulting in substantial inference costs. To reduce both cost and laten...
arXiv:2610.35817v1 Announce Type: cross Abstract: A robot policy needs to understand what a task is asking for in a scene, and how to move the robot to accomplish that task. To achieve these two skil...
arXiv:2610.34385v1 Announce Type: cross Abstract: Agents that act on real systems should predict what their actions will cause before acting. We study when causal structure helps a learned forward mo...