6 papers
MS-Eval: Benchmarking Sub-Agent Delegation and Orchestration in MCP-Intensive Working Scenarios
Bowen Ding, Qixin Wang, Lexing Ruan +4
LLM agents are increasingly integrated into real-world works, yet they must handle diverse domain-specific tools and complex multi-sub-task problems that sequential tool invocation…
Long-Horizon Scaling: How Model Capabilities Shape the Returns to Computation
Haoyu Zheng, Zhengyu Chen, Huaisheng Zhu +5
Long-horizon agents improve solutions through sustained interaction, execution, and task feedback. Scaling studies relate performance to resources and capabilities, yet how existin…
LLaDA-Image: Building Strong Image Generators with Fully Open Training Recipes
Chuyan Chen, Haoxing Chen, Kun Chen +27
We introduce LLaDA-Image, a unified framework that pairs a 6B Diffusion Transformer (DiT) trained from scratch with a frozen vision-language understanding module built on the LLaDA…
RODS: Reward-Driven Online Data Synthesis for Multi-Turn Tool-Use Agents
Ruishan Fang, Siyuan Lu, Chenyi Zhuang +1
Multi-turn tool-use RL is bottlenecked by the rapid depletion of informative samples in static datasets. We observe that the gradient signal in GRPO concentrates on tasks with the…
Do More Agents Help? Evaluating Accuracy and Cost in LLM Multi-Agent Workflows
Yuhang Fu, Ruishan Fang, Jiaqi Shao +5
Choosing whether to replace a single-agent controller requires evidence on both task performance and inference expenditure. We present BenchAgent, a shared evaluation framework, an…
Stop Wasting Your Tokens: Towards Efficient Runtime Multi-Agent Systems
Fulin Lin, Shaowen Chen, Ruishan Fang +2
While Multi-Agent Systems (MAS) excel at complex tasks, their growing autonomy with operational complexity often leads to critical inefficiencies, such as excessive token consumpti…