Publications (13)
Ling and Ring 2.6 Technical Report: Efficient and Instant Agentic Intelligence at Trillion-Parameter Scale
Ang Li, Ben Liu, Bin Han +215
Efficient and scalable agentic intelligence requires models that can deliver both low-latency responses and strong reasoning capabilities while remaining practical to train, serve,…
Every Step Evolves: Scaling Reinforcement Learning for Trillion-Scale Thinking Model
Ling Team, Anqi Shen, Baihui Li +101
We present Ring-1T, the first open-source, state-of-the-art thinking model with a trillion-scale parameter. It features 1 trillion total parameters and activates approximately 50 b…
Extraordinary acoustic reflection enhancement by acoustically-transparent thin plates
Zhaojian He, Shasha Peng, Chunyin Qiu +6
We report an observation of the extraordinary high reflection of acoustic waves in water by thin epoxy plates partitioned by subwavelength cuts, whereas such plates without structu…
Pseudo-time-reversal symmetry and topological edge states in two-dimensional acoustic crystals
Jun Mei, Ze-Guo Chen, Ying Wu
We propose a simple two-dimensional acoustic crystal to realize topologically protected edge states for acoustic waves. The acoustic crystal is composed of a triangular array of co…
Do Linear Dispersions of Classical Waves Mean Dirac Cones?
Jun Mei, Ying Wu, C. T. Chan +1
By using the \vec{k}\cdot\vec{p} method, we propose a first-principles theory to study the linear dispersions in phononic and photonic crystals. The theory reveals that only those…
Next-Generation Agentic Reinforcement Learning Systems Enable Self-Evolving Agents
Ran Yan, Wei Fu, Jiale Li +21
LLM agents are rapidly being deployed in production, including coding assistants, customer-support chatbots, and scientific research assistants, yet they remain fundamentally stati…
Maximum A Posteriori Inference in Sum-Product Networks
Jun Mei, Yong Jiang, Kewei Tu
Sum-product networks (SPNs) are a class of probabilistic graphical models that allow tractable marginal inference. However, the maximum a posteriori (MAP) inference in SPNs is NP-h…
Acoustic Metagrating Circulators: Nonreciprocal, Robust, and Tunable Manipulation with Unitary Efficiency
Lijuan Fan, Jun Mei
Nonreciprocal signal operation is highly desired for various acoustic applications, where protection from unwanted backscattering can be realized so that transmitting and receiving…
A Lumped Model for Rotational Modes in Phononic Crystals
Pai Peng, Jun Mei, Ying Wu
We present a lumped model for the rotational modes induced by the rotational motion of individual scatterers in two-dimensional phononic crystals comprised of square arrays of soli…
AReaL: A Large-Scale Asynchronous Reinforcement Learning System for Language Reasoning
Wei Fu, Jiaxuan Gao, Xujie Shen +10
Reinforcement learning (RL) has become a dominant paradigm for training large language models (LLMs), particularly for reasoning tasks. Effective RL for LLMs requires massive paral…
Ring-lite: Scalable Reasoning via C3PO-Stabilized Reinforcement Learning for LLMs
Ling Team, Bin Hu, Cai Chen +43
We present Ring-lite, a Mixture-of-Experts (MoE)-based large language model optimized via reinforcement learning (RL) to achieve efficient and robust reasoning capabilities. Built…
Acoustic frequency filter based on anisotropic topological phononic crystals
Ze-Guo Chen, Jiajun Zhao, Jun Mei +1
There are growing efforts in constructing topological edge states in classical wave system. However, most of the work study the existence, creation and properties of the edge state…
Sound Absorption by Subwavelength Membrane Structures: A Generalized Perspective
Min Yang, Yong Li, Chong Meng +4
Decorated membrane, comprising a thin layer of elastic film with small rigid platelets fixed on top, has been found to be an efficient absorber of low frequency sound. In this work…