1 citations · 3 across the 13 of their papers we have counts for
14 papers
Coefficientwise Extremality and Rigidity for Independent Sets in Bipartite Graph Covers
Yusuke Watanabe
Let be a finite bipartite graph and let be an -fold cover. We compare independent sets of with those of the trivial cover, consisting of di…
Intent Speaks Louder: Controllable User Simulation Beyond Response Imitation
Bo Wang, Ruixing Zhang, Yunqi Liu +4
User simulators are widely used as scalable environments for training and evaluating interactive assistants. Generating the next user turn is inherently one-to-many: the same profi…
SeqLLM: Augmenting LLMs with Behavioral-Sequence Modeling for High-Stakes Decisions at WeChat Pay
Guilin Li, Jiaxing Zhang, Matthias Hwai Yong Tan +2
Merchant risk control at large payment platforms screens tens of millions of merchants daily, where false positives harm legitimate merchants and false negatives leave harmful acti…
Scaling Embeddings Outperforms Scaling Experts in Language Models
Hong Liu, Jiaqi Zhang, Chao Wang +13
While Mixture-of-Experts (MoE) architectures have become the standard for sparsity scaling in large language models, they increasingly face diminishing returns and system-level bot…
SONIC: Segmented Optimized Nexus for Information Compression in Key-Value Caching
Hong Chen, Xiang Liu, Bo Wang +5
The linear growth of Key-Value (KV) cache remains a bottleneck for multi-turn LLM deployment. Existing KV cache compression methods often fail to account for the structural propert…
PANTHER: Generative Pretraining Beyond Language for Sequential User Behavior Modeling
Guilin Li, Yun Zhang, Xiuyuan Chen +6
Large language models (LLMs) have shown that generative pretraining can distill vast world knowledge into compact token representations. While LLMs encapsulate extensive world know…
LongCat-Flash Technical Report
Meituan LongCat Team, Bayan, Bei Li +179
We introduce LongCat-Flash, a 560-billion-parameter Mixture-of-Experts (MoE) language model designed for both computational efficiency and advanced agentic capabilities. Stemming f…
Urban In-Context Learning: Bridging Pretraining and Inference through Masked Diffusion for Urban Profiling
Ruixing Zhang, Bo Wang, Tongyu Zhu +2
Urban profiling aims to predict urban profiles in unknown regions and plays a critical role in economic and social censuses. Existing approaches typically follow a two-stage paradi…
MindSpeed RL: Distributed Dataflow for Scalable and Efficient RL Training on Ascend NPU Cluster
Laingjun Feng, Chenyi Pan, Xinjie Guo +11
Reinforcement learning (RL) is a paradigm increasingly used to align large language models. Popular RL algorithms utilize multiple workers and can be modeled as a graph, where each…
Pangu Ultra MoE: How to Train Your Big MoE on Ascend NPUs
Yehui Tang, Yichun Yin, Yaoyuan Wang +71
Sparse large language models (LLMs) with Mixture of Experts (MoE) and close to a trillion parameters are dominating the realm of most capable language models. However, the massive…
On-Device Qwen2.5: Efficient LLM Inference with Model Compression and Hardware Acceleration
Maoyang Xiang, Ramesh Fernando, Bo Wang
Transformer-based Large Language Models (LLMs) have significantly advanced AI capabilities but pose considerable challenges for deployment on edge devices due to high computational…
Pangu Ultra: Pushing the Limits of Dense Large Language Models on Ascend NPUs
Yichun Yin, Wenyong Huang, Kaikai Song +49
We present Pangu Ultra, a Large Language Model (LLM) with 135 billion parameters and dense Transformer modules trained on Ascend Neural Processing Units (NPUs). Although the field…
Valuable Hallucinations: Realizable Non-realistic Propositions
Qiucheng Chen, Bo Wang
This paper introduces the first formal definition of valuable hallucinations in large language models (LLMs), addressing a gap in the existing literature. We provide a systematic d…
Generative Pretraining at Scale: Transformer-Based Encoding of Transactional Behavior for Fraud Detection
Ze Yu Zhao, Zheng Zhu, Guilin Li +2
In this work, we introduce an innovative autoregressive model leveraging Generative Pretrained Transformer (GPT) architectures, tailored for fraud detection in payment systems. Our…