arXiv AI

An Exact Generate - Transform Decomposition of Small-LLM Team Scaling Across Orchestration Architectures

The paper investigates how scaling a team of small language‑model agents affects performance across different orchestration architectures. By testing eight architectures on five short‑answer benchmarks and an executable‑code benchmark, it finds that team scaling yields large gains on arithmetic word‑problem tasks but only modest improvements on multiple‑choice and code generation tasks, with no single architecture dominating all tasks. The authors explain these patterns using a generate‑transform decomposition that separates coverage and transformation effects, showing that arithmetic tasks benefit from both coverage and critic‑guided transformation, while other tasks are limited by saturation or poor conversion.

arXiv AI
Jul 9

Cost-Effective Agent Harnesses for Abstract Reasoning and Generalization on ARC-AGI-1

arXiv:2607. 06764v1 Announce Type: new Abstract: Recent progress on ARC-AGI-1 from disclosed architectures has come broadly from two regimes: heavy test-time compute over frontier models (evolutionary search, exhaustive sampling, extended chain-of-thought), or benchmark-specific training in which small models are fine-tuned on ARC data, often with task-specialized architectures.

By Kabir Moghe, Peter Chin
arXiv AI
Jul 1

ClawArena-Team: Benchmarking Subagent Orchestration and Dynamic Workflows in Language-Model Agents

arXiv:2606. 31174v1 Announce Type: new Abstract: Production large language-model (LLM) agents are increasingly deployed not as lone problem-solvers but as managers: a main model creates specialized subagents, delegates work, and orchestrates their parallel, asynchronous returns through dynamic workflows.

By Kaiwen Xiong, Haonian Ji, Shi Qiu, Zeyu Zheng, Cihang Xie, Xinyu Ye, Huaxiu Yao
arXiv AI
Sep 2

UniACE: A Unified Framework for Evaluating LLM Agentic Capabilities

UniACE is a unified framework that standardizes the evaluation of large language model (LLM) agents by representing each benchmark as an instruction–tool–environment triplet and running models through a shared, task‑agnostic harness in isolated runtimes. It preserves native success criteria, offers an offline mode for dynamic‑resource tasks, and standardizes efficiency metrics, execution records, and failure attribution. Applying UniACE to 7 benchmarks across 24 domains and 15 models revealed significant score shifts, ranking reversals, and sensitivity to evidence representation, highlighting the impact of evaluation configuration on reported agent performance.

By Pengyu Zhu, Lijun Li, Yaxing Lyu, Qianxin Luo, Jingyi Yang, Yi Liu, Tingfeng Hui, Xinyu Yuan, Li Sun, Sen Su, Jing Shao
arXiv AI
Oct 27

AgentOrch: How Close Can Global Orchestration Come to Step-by-Step Interaction?

AgentOrch is a benchmark that compares global orchestration—where a complete workflow is generated upfront—with step‑by‑step interaction on 650 enterprise standard operating procedures and 3,220 runtime variants. Across all evaluated models, step‑by‑step remains more accurate, but the performance gap has narrowed by about 79% from early to recent models. The study identifies three methods—incremental workflow construction, revising workflows with execution feedback, and targeted training on complete workflows—that can further close the gap, improving global orchestration accuracy by up to 12.18 percentage points.

By Zijie Liu, Qi Yang Huang, Jinhao Duan, Yueh Che Lee, Thomas A. Joseph, Zayne Turner, Adam Seligman, Tianlong Chen, Zhaozhuo Xu
arXiv Computation and Language
Sep 15

When Agents Slow Down: Understanding LLM Agents' Test-Time Strategies via Elo-per-token Analysis

arXiv:2609.15309v1 Announce Type: new Abstract: Large language model (LLM) agents allocate test-time compute adaptively as they revise solutions, use tools, explore alternatives, and decide when to s...

By Kaiyuan Liu, Qiuyang Mang, Bo Peng, Wenhao Chai, Hanchen Li, Shreyas Pimpalgaonkar, Luke Zettlemoyer, Alex Dimakis, Alvin Cheung
arXiv AI
Aug 28

Zero-Shot Self-Orchestration with Ledger-Based Control for Improved LLM Coding Performance

The paper evaluates a manager‑worker scaffold that uses a shared filesystem workspace to orchestrate multi‑agent large language model (LLM) coding tasks without training or tuning. Across nine models—including five open‑weight and four closed‑weight systems—the scaffold yields statistically significant accuracy gains for some models (e.g., Qwen3.8‑27B, GPT‑5.6‑Luna, GPT‑5.6‑Terra, Kimi‑K3, Minimax‑M3) while producing null or negative effects for others (e.g., Qwen3.6‑35B). The study shows that the manager can triple token usage but still achieves higher accuracy at a fraction of the cost compared to larger single‑pass models, with key mechanisms identified as context management and problem decomposition.

By Victor Gao (Sang Won), Vida Khosrowshahi (Sang Won), Ali Khosrowshahi (Sang Won), Xihao Sun (Sang Won), Juhyun Lee (Sang Won), Simon (Sang Won), Lee
arXiv AI
Oct 5

FastKernels: Benchmarking GPU Kernel Generation in Production

FastKernels is a benchmark comprising 384 GPU kernel tasks from 47 architectures across 8 categories, designed to evaluate kernel generation agents in realistic production settings. It covers 94.6% of HuggingFace Transformers architectures, scoring kernels on correctness, coverage, and speedup within their native execution paths. The benchmark reveals that kernel‑level speedups often collapse to modest end‑to‑end gains and that isolated performance rankings can misrepresent real‑world effectiveness.

By Gabriele Oliaro, Jaeseong Lee, Yichao Fu, Zhaoyuan Su, Owen Lu, Sreeram Vennam, May Jiang, Junli Wang, Hao Zhang, Zhihao Jia, Samyam Rajbhandari
arXiv Computation and Language
Oct 12

Proactive Agentic Orchestration

The paper introduces POrchestra, a proactive agentic orchestration framework that lets sub‑agents request harness adjustments during execution, enhancing adaptability to dynamic environments. It also proposes On‑Policy Orchestration Distillation (OPOD), which uses hindsight feedback from sub‑agents’ requests to provide targeted improvement signals for training the orchestration policy. Experiments on three long‑horizon benchmarks show that POrchestra boosts task success by about 9.1% over the best baseline, and with OPOD the improvement rises to 10.0%, while also improving cost‑performance trade‑offs.

By Xueyang Feng, Chaozhuo Li, Xiaohe Bo, Jiakai Tang, Chenxi Li, Qiwei Ye, Xu Chen
arXiv Machine Learning
Jun 2

ATLAS: Agentic Test-time Learning-to-Allocate Scaling

arXiv:2606. 01667v1 Announce Type: new Abstract: Test-time scaling has become a major way to improve large language model reasoning, but its orchestration has remained designer-engineered: a fixed sample budget, a fixed refinement loop, a fixed scoring rule, or a fixed search policy decides how compute is spent, leaving the model in charge of solving but not of orchestration.

By Peijia Qin, Qi Cao, Pengtao Xie