arXiv:2602. 13255v2 Announce Type: replace Abstract: We present DPBench, a benchmark for evaluating coordination in multi-agent systems built from large language models.
arXiv:2606. 07845v1 Announce Type: cross Abstract: We measure how well current large language models coordinate as multiple agents sharing a common resource, using the dining philosophers problem as a clean test bed.
arXiv:2606. 08340v1 Announce Type: new Abstract: As language models are increasingly deployed as autonomous agents, they must coordinate with others over long horizons in open-ended interactive tasks.
arXiv:2606. 17182v1 Announce Type: new Abstract: Multi-agent LLM systems share state through memory stores, vector indices, and tool registries.
The paper introduces a coroutine-bridge harness that lets a language model emit a Python program to manage tool calls in the CAR-bench evaluation. By decoupling model invocations from tool round-trips, the approach reduces model calls to a median of two per task while maintaining seven agent turns, achieving a median latency of 1.8 s on a Cerebras gpt‑oss‑120b. The harness achieved 60.0 % Pass³ on the official hidden evaluation, outperforming the baseline by 4.5× and matching frontier-model agents on GPT‑5.5, all while keeping the prompt largely cached and minimizing input compute.
arXiv:2606. 28514v1 Announce Type: new Abstract: Multimodal models are increasingly deployed to solve tasks collaboratively with humans or other artificial agents.
arXiv:2607. 05690v2 Announce Type: replace Abstract: Language agents run a loop - observe, reason, act - but the memory they reason over sits outside it: a store queried at most once per turn.
arXiv:2607. 05690v1 Announce Type: new Abstract: Language agents run a loop - observe, reason, act - but the memory they reason over sits outside it: a store queried at most once per turn.
arXiv:2607. 20531v1 Announce Type: new Abstract: Large language model (LLM) agents are increasingly deployed over Model Context Protocol (MCP) servers, yet the benchmarks used to evaluate them score the final answer or a fixed "ground-truth" list of tools, both of which are fragile once the underlying data is live and stateful.
arXiv:2606. 16478v1 Announce Type: new Abstract: Large language models (LLMs) remain limited in multi-agent planning because independently generated plans can create coordination failures such as spatial collisions, resource contention, and temporal deadlocks.
FinalityBench is an executable benchmark that tests how agents decide on shipping, re‑capturing, refunding, or waiting when a merchant’s payment processor, ledger, ERP, and bank feed receive delayed, duplicated, dropped, or reordered messages, causing contradictory beliefs about an order. The benchmark uses a hidden canonical event log and faulted delivery streams to generate system views, scoring each episode by the merchant’s terminal economic position relative to a privileged reference. It contains 321 tasks, including 45 twin pairs where all four views are identical yet the correct disposition differs, and evaluates nine programmatic policies, revealing that a ship‑on‑first‑sign policy performs best by accuracy but worst by paired loss, while a runtime‑gated irreversible‑action policy achieves 85.4% accuracy without losing money.
Writing Answer Set Programming (ASP) theories from scratch is a difficult and time-consuming task. We take a neurosymbolic approach to study whether a model can distill complete and correct theories, given a fixed agent harness with the solver in the loop.
MARS (Multi-Agent Relay of Specialized LLMs) is a prompt-only framework that assigns specialized LLM agents—each focused on a particular algorithmic domain such as dynamic programming, graphs, or geometry—to collaboratively solve competitive programming problems. Retrieval-augmented generation selects a small team of relevant specialists for each problem, and the agents iteratively refine a C++17 solution through sandboxed testing, passing structured packets between them until a final infrastructure-fixer normalizes the code. On the CodeContests benchmark, MARS achieves a pass rate of 0.624 with Gemma 4, improving over direct prompting by 14.4 percentage points while reducing wall‑clock cost and token‑spend variance compared to CodeSIM.