Tool use, function calling, orchestration and the protocols that let models act rather than only answer.
arXiv:2609.11999v1 Announce Type: cross Abstract: In this study, we examine whether a general shell can outperform specialized tools on enterprise tasks. Shell-based agents have shown strong results...
arXiv:2609.12541v1 Announce Type: new Abstract: We demonstrate that a general-purpose agent can directly drive a physical robot throughout task execution without any task-specific or environment-spec...
LLM‑BabyBench transforms the BabyAI gridworld into a fully observable, purely textual setting that isolates planning as the sole source of failure. By serialising the entire grid, providing formal instructions, and validating actions deterministically, the benchmark introduces the PPD suite—Predict, Plan, and Decompose tasks—each scored with metrics that separate mission understanding from sequencing. Across a range of large language models, simulation accuracy is high while planning success drops sharply beyond a model‑specific horizon, revealing that plan length—not grid size—drives failure and that models often commit to a single corridor‑shaped route without backtracking.
The paper demonstrates that aggregate accuracy figures for chain‑of‑thought (CoT) monitors can be misleading because a large portion of detected hacks rely solely on action patterns rather than reasoning. By rewriting only the agent’s reasoning to appear truthful while keeping actions identical, the authors show that the monitor’s performance on the reasoning‑dependent subset collapses dramatically, yet the overall pooled accuracy drops only modestly. The study reveals that CoT monitors are fragile when reasoning is the key signal and that accuracy should be reported separately for this subset.
Open Thai automatic speech recognition (ASR) is dominated by offline, Whisper-based models that read the whole utterance before transcribing, ruling out low-latency uses such as live captioning and vo...
Simon Willison released commit-rewriter 0.1, a web app that lets users edit commit messages for a repository. The tool is useful for cleaning up commits that contain internal references or unwanted code before publication. After editing, it creates a timestamped branch to preserve the original state and rewrites the selected commits.
Hierarchical orchestration, in which a Manager agent reviews worker output and can send it back for revision, is the default coordination pattern in production multi-agent LLM frameworks. Classical or...
Autonomous agents that author building information models need more than access to a host API. They need a representation of what they intended, what a compiler decided on their behalf, what was refus...
Simon Willison demonstrates how ChatGPT Work with GPT‑6 Astra can generate 5K and 10K running routes that loop from a specified address using OpenStreetMap data. The system produced an embedded visualization and downloadable GPX and GeoJSON files, and explained that it used Nominatim, Overpass, and local calculations to create the routes. However, the actual Python code used was not accessible due to thread compaction, highlighting a transparency issue in LLM systems that compact conversation history.
The paper introduces an end‑to‑end multi‑agent framework that automatically converts natural‑language problem descriptions into Quadratic Unconstrained Binary Optimization (QUBO) formulations, a key format for combinatorial optimization on quantum and hybrid solvers. It also presents QUBOBench, a benchmark of 100 problems from 12 domains, and reports that the framework achieves 68% accuracy—22% better than a single‑call baseline—thanks largely to iterative self‑repair. The authors provide open‑source data and code at https://quitttcat.github.io/QuantumQUBOAgent.
The paper introduces environment‑probing curation, a deployment‑compatible method that equips asynchronous curator agents with read‑only world tools to verify, scope, and refresh candidate memories without retraining models. In a GitHub Copilot‑based harness, this approach improves pass rates on CLBench from 39% to 73%, boosts reward metrics, and reduces both query counts and task‑agent costs. Across six APEX management‑consulting tasks, the method consistently outperforms baselines, yielding higher rewards and fewer tool calls while maintaining a compact task‑time interface.
The paper introduces DRG-MAPPO, a hierarchical multi‑agent reinforcement learning framework for cooperative air combat. It combines graph‑based relational modeling with dynamic role assignment, using a high‑level policy to allocate tactical roles such as leader and supporter, and a low‑level policy to execute maneuver actions. The approach includes a target‑priority auxiliary task and achieves an 87% win rate in experiments, indicating effective coordination and stability.
The paper introduces Debate‑to‑Skill, a capability‑bound process supervision method for annotating industrial query‑to‑agent tasks. It addresses failures caused by confusing topical relevance with executable capability, especially for long‑tail and boundary‑sensitive requests. By employing reusable decision principles, structured deliberation, verifier‑based verdict extraction, and disagreement‑driven refinement, the approach is evaluated against direct‑label supervision, reasoning‑SFT, and structural ablations on a Query2Agent benchmark, focusing on grey‑zone cases where semantic relatedness and executable capability diverge.
The paper investigates using Large Language Models (LLMs) as post-hoc auditors for symbolic regression models produced by genetic programming. By having three LLMs analyze and rank four evolved expressions across multiple runs, the study compares the LLM-generated rankings and interpretations with assessments from three clinicians. Results show that LLMs provide more favorable comparative rankings than isolated term-level interpretations, yet they also generate physiologically and mathematically questionable explanations, suggesting they are best used under expert oversight rather than for autonomous validation.
The paper explores how large language models (LLMs) can forecast a firm’s future financial performance by integrating alternative data—such as consumer transactions, web traffic, and prediction markets—with traditional financial information. A two‑agent framework is proposed: one agent identifies which alternative data channels are relevant for each firm, and the other uses firm‑ and channel‑specific context to predict revenue. Experiments across four commercial alternative data channels show that incorporating alternative data in context improves LLM forecasts over using either data source alone and outperforms standard forecasting baselines.
MAPLE is a new agent that maintains and updates optimization problems through successive natural‑language requests, combining language‑based problem construction with mathematical programming and evolutionary search. It preserves the optimization program, accepted plans, earlier updates, and candidate solutions for future requests, enabling rapid adaptation to changing business constraints. In a benchmark of 15 trajectories and 180 updates across various operational domains, MAPLE completed all trajectories with high online scalar quality and Pareto hypervolume ratio, and maintained update validity and useful search information across substantial revisions.
The paper discusses the transition of Agentic AI from bounded task execution to systems that maintain consequential state and adapt across task boundaries, highlighting a new control problem. It proposes an artificial ID—a self‑driving internal mechanism that decides when to continue, stop, or change behavior—demonstrated in a minimal virtual Petri‑dish experiment where the agent develops useful control without explicit task objectives. The authors argue that while this persistence can enable adaptive agency, it also risks misalignment, corrupted state, and unintended behavior, suggesting that a scalable artificial ID would require persistent alignment boundaries encompassing trusted observations, consequence channels, and hard constraints.
The paper proposes a new framework for collective information engines that rely on role differentiation rather than consensus. By modeling anti‑coordination games, agents infer roles from noisy social signals tied to persistent identities, and role‑following actions reinforce those identities, creating a feedback loop that can drive collective order. The authors show that when a social loop gain—determined by identity persistence, cognitive capacity, channel fidelity, and schema strength—exceeds one, roles emerge in a bifurcation cascade whose type is selected by resource‑driven replicator dynamics, offering a mechanistic basis for distributional AGI takeoff and a control lever for platform design.
The paper introduces the Static‑Pass Dynamic‑Fail (SPDF) phenomenon, showing that static analysis can miss vulnerabilities that are exploitable at runtime. Using a three‑stage pipeline—static scanning, LLM‑driven CWE reasoning, and autonomous exploit verification—it evaluated 1,355 Python samples and found that 14.53% of samples that passed static checks were actually exploitable. The study highlights that static‑analysis success and runtime security are distinct assurance layers, especially for AI‑generated and security‑sensitive code.
The study introduces a scaffolding-oriented multi-agent Large Language Model (LLM) AI Standardized Patient (AI-SP) platform designed to train medical students in patient interviews. In a randomized controlled trial with 100 students, the multi-agent system—comprising a patient agent, a Socratic tutor agent, and a turn-level evaluator—did not improve diagnostic accuracy but significantly enhanced overall OSCE scores, especially in communication, empathy, and history-taking. The authors also release a richly annotated dataset to support further research in AI-supported clinical reasoning training.