We present MARL-BattleGrounds (MARL-BGs), an easy-to-use JAX-native benchmark for heterogeneous and competitive multi-agent reinforcement learning (MARL), designed to be sample-efficient, computationally efficient, and researcher-centric. MARL benchmarks are typically simple and efficient, like SMAX, where winning redu...
Experience-based learning enables large language model (LLM) agents to improve over time by reusing knowledge from past executions. However, existing methods typically rely on final task outcomes as feedback for experience learning, overlooking the execution process and limiting their ability to evaluate and optimize e...
Bingjie Zhang, Yibo Yang, Qiantong Liu et al.· 0 citations
Data-driven weather forecasting models are commonly trained on ERA5, yet operational deployment must use timely initial states from physics-based numerical weather prediction (NWP) systems whose distributions differ substantially from ERA5. Direct initialization with these states can degrade autoregressive forecasts, m...
Mu-Tian Xi, Xinyu Su, Xiao-Hui Zhong et al.· 0 citations
The growing deployment of compact large language models on smartphones is enabling autonomous on-device decision-making, such as emerging mobile GUI agents. However, many mobile decisions, such as network configuration when users switch apps or connectivity conditions change, must satisfy stringent latency constraints...
Haowen Wan, Shangzhuo Xie, Syue-Ju Syue et al.· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
Claims of recursive self-improvement are easy to make and hard to falsify: did improvement come from re-using designer-provided mechanisms, or from mechanisms the system authored itself? We introduce a pre-registered, two-tier accounting: M1 counts machine-generated content instances within a designer-provided vocabula...
EvoCowork is introduced, a 240-task benchmark for evaluating harness-level recursive self-improvement across four office modalities, with model weights held fixed.
Zhou Liu, Xing Chen, Zhao-Yang Han et al.· 0 citations
Agentic AI moves generative AI applications toward systems that coordinate reasoning, enterprise context, model inference, and external actions. Integrating these capabilities into enterprise software introduces diverse architectural concerns, from orchestration, knowledge and memory, and model access to tools and acti...
Ming-Qin Yu, F. Rabhi, Li-Ming Zhu et al.· 0 citations
Evaluating autonomous agents requires examining whether they can translate goals into coherent action across extended interactions. In persistent environments, this requires coordinating interdependent actions under incomplete observations, accounting for delayed effects, and adapting as earlier decisions shape subsequ...
Jian-Feng Lin, Run-Ze Wang, Hao-Rong Hong et al.· 0 citations
Forecast-based models are fundamental to network traffic management and anomaly detection. However, it is essential to evaluate these forecasting models over extended periods of time in order to assess their adaptability to unexpected phenomena and shifts in underlying patterns. This kind of long-term evaluation requir...
Josef Koumar, Karel Hynek, Kamil Jerabek et al.· 0 citations
Online, same-day delivery requires operators to constantly update their routes, and the routing cost for deliveries increases with the number of customers. Out-of-home (OOH) delivery options can help companies handle increasing customer numbers while keeping delivery costs low. We develop a dynamic pricing-and-routing...
Attila Lischka (Jeff), Ze Zhou (Jeff), Xuegang (Jeff) et al.· 0 citations
Reflective language agents reuse feedback and memory to adapt across tasks without updating model weights. However, existing guarantees for policy selection and action generation do not by themselves establish when local reflection yields long-term performance guarantees for the complete agent under resource constraint...
Zhongnian Li, Shi Xu, Mengting Xu et al.· 0 citations
LLM-powered data analytics aims to translate natural language instructions into executable analytical programs over underlying data. However, existing approaches, whether treating LLMs as intent-to-code translators or as standalone analytical engines, often fail to reliably align domain semantics with complex schemas a...
Ruichen Xia, Xiao Han, Dingyu Yang et al.· 0 citations
