Tool use, function calling, orchestration and the protocols that let models act rather than only answer.
arXiv:2610.34163v1 Announce Type: new Abstract: Crowd barriers are widely used to regulate pedestrian movement during large public events such as pilgrimages. However, fixed barrier layouts often cre...
arXiv:2610.35044v1 Announce Type: new Abstract: World models are emerging as a fundamental component of policy learning for real-world agents, enabling agents to predict future world states and plan...
arXiv:2610.35850v1 Announce Type: new Abstract: Direct-policy World Action Models (WAM) have demonstrated the benefits of decoupling action generation from explicit future visual prediction for effic...
arXiv:2610.33966v1 Announce Type: cross Abstract: People have long dreamed of an assistant that is always there to guide them and remind what they forget, from new dishes to lost keys. Powered by mul...
arXiv:2610.34108v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models map a language instruction and the current observation directly to actions, without checking whether the instruct...
arXiv:2604.10383v4 Announce Type: replace Abstract: Authoring a multi-actor scenario for a living 3D world, where every action changes its state, and each action's validity depends on the state accum...
arXiv:2608.05803v2 Announce Type: replace Abstract: Recent advances in generative video modeling have enabled diverse generation, reference-based synthesis, extension, and editing, but existing appro...
arXiv:2610.14403v2 Announce Type: replace Abstract: Learning reusable skills can improve visual reasoning agents, yet applying a skill still requires the agent to implement or invoke a procedure and...
arXiv:2610.24016v2 Announce Type: replace-cross Abstract: Robots that follow natural-language instructions unlock a wide range of autonomous applications in households, warehouses, and labs. Real-wor...
arXiv:2610.33842v1 Announce Type: cross Abstract: Bayesian mixtures, and Solomonoff induction in particular, provide an idealized model of prediction learned by minimizing population log loss, and th...
The paper introduces the Preference-Behavior Model (PBM), which jointly infers Large Language Model (LLM) preferences and imperfect decision-making mechanisms from observed choices. By applying PBM to three social dilemma games—Public Goods, Collective-Risk, and Common-Pool Resource—the authors show that preferences derived from one dilemma can predict behavior in others, whereas traditional behavioral economic elicitation instruments fail to do so consistently. The study also evaluates how well inferred preferences align with persona instructions, revealing both broad agreement and specific divergences across dimensions, LLMs, and games.
The paper "Can LLMs Trust Appropriately? Evaluating Calibrated Reliance on Imperfect Partners" introduces a benchmark of 324 scenarios across 23 collaboration categories to test how well large language models translate partner evidence into calibrated trust decisions. Each scenario requires three sequential decisions with defined transition dynamics, feedback, and cost schedules, allowing precise analytical evaluation of choices against expected outcomes. The study evaluates seven models and 36 controlled-history probes, finding that while the best frontier model approaches near‑optimal decisions, other models display pathologies such as excessive safeguards, zero‑value verification purchases, prioritizing immediate savings over cooperative learning, and misallocating assistance.
MultiEvoSkill is a skill‑optimization framework that transforms skill documents into typed graphs, enabling explicit procedural topology and parallel branches. It employs a heterogeneous agent system to generate revision suggestions based on diverse experiences, and a structure‑aware debate mechanism to resolve edits while respecting dependencies and importance constraints. Experiments show that this approach outperforms state‑of‑the‑art methods and improves cross‑model generalization of skill artifacts.
The paper investigates how users should set the reasoning‑effort level in large language models, a parameter that trades off inference cost for potential performance gains. Experiments on proprietary and open‑weight models across four benchmarks show that higher effort does not consistently improve accuracy, can be up to 42× more expensive on easy tasks, and alters agent behavior by increasing verification and search steps. The authors propose a simple escalation strategy—starting at the lowest effort and only increasing after a verified failure—that matches or outperforms fixed high‑effort settings while reducing cost.
The paper presents the first dataset that captures how humans learn and use mechanisms in unfamiliar environments, using ARC‑AGI‑3 as a testbed. It includes 144 trajectories from 64 participants across 25 public games, each paired with time‑synchronized actions and think‑aloud audio, and 250 trajectories from large language model agents with and without harnesses. The dataset is annotated for mechanism triggers, action purposes, errors, and recovery, enabling a comparison of human and LLM cognition, highlighting differences in hypothesis testing, exploration, and use of qualitative representations.
InferPriv is a new benchmark that evaluates inferential privacy leakage in large language model (LLM) assistants. It measures whether an honest-but-curious recipient can deduce a user’s sensitive attributes from an LLM’s reply during routine tasks, a type of leakage not captured by existing direct‑leakage benchmarks. Testing five frontier models, the study finds that inferential leakage rates (16.2–25.2%) far exceed direct leakage, even when direct leakage is near zero under privacy‑aware prompts.
The paper introduces CEPO, a framework that separates coordination and execution in multi‑agent systems by using a shared coordinator prompt and agent‑specific executor prompts. CEPO guides the optimization of each prompt type according to its role, enabling consistent refinement of coordination rules while allowing execution instructions to remain adjustable. Experiments on 11 benchmarks across four domains show that CEPO consistently improves the performance of four existing prompt optimizers, and additional studies demonstrate better optimization budget efficiency and scaling with more agents.
AutoEnergyDesign is an LLM‑driven framework that optimizes building energy system designs by iteratively proposing configuration changes based on prior simulation results and automatically reverting unsuccessful changes via a git‑based mechanism. The agent combines a reasoning model with explicit memory (procedural, semantic, episodic, reflective), a constrained action interface, and a keep/revert signal, and evaluates designs using a 25‑year Net Present Cost model across eight technologies. In a case study with measured building loads, the agent (using Claude Haiku 5.5, Sonnet 5.5, and Opus 5.5) achieved an NPC within 1% of the best baseline after far fewer simulations, outperforming random search, TPE Bayesian optimization, and a $(1+1)$ evolution strategy.
RSIV-Harness is a framework that keeps language model weights fixed while recursively improving the surrounding harness based on failure traces, using a four-layer evaluation to ensure stability. It addresses challenges in open-domain conversational text-to-visualization, such as ambiguous user requests, data source discovery, and iterative chart refinement. The introduced ODAC-Vis benchmark evaluates these capabilities across 640 tasks, and experiments show significant gains in executability and chart structure after a single RSIV-Harness iteration.
The paper introduces HMOS‑Migrator, an agentic pipeline that converts Android or iOS applications into HarmonyOS NEXT ArkTS projects and validates them by replaying recorded user interactions on a real device. It evaluates migration quality using seven weighted metrics that rely solely on the source application's behavior and generated artifacts, avoiding reference implementations. Across twelve trials on six open‑source apps, the pipeline achieves high structural completeness (83.3–100%) but shows lower on‑device behavior (4.3–63.1%) and UI fidelity (21.1–63.2%) scores, with gradual performance improvements over time.