papers

Publications (41)

cs.IR2024

LLM-Guided Multi-View Hypergraph Learning for Human-Centric Explainable Recommendation

Zhixuan Chu, Yan Wang, Qing Cui +4

As personalized recommendation systems become vital in the age of information overload, traditional methods relying solely on historical user interactions often fail to fully captu…

cs.LG2026

UniGeM: Unifying Data Mixing and Selection via Geometric Exploration and Mining

Changhao Wang, Yunfei Yu, Xinhao Yao +5

The scaling of Large Language Models (LLMs) is increasingly limited by data quality. Most methods handle data mixing and sample selection separately, which can break the structure…

math.DG2018

Some sharp differential sphere theorems for nonnegative scalar curvature manifolds

Qing Cui, Linlin Sun

In this paper, we obtain several new intrinsic and extrinsic differential sphere theorems via Ricci flow. For intrinsic case, we show that a closed simply connected -dime…

cs.LG2026

On Representation Redundancy in Large-Scale Instruction Tuning Data Selection

Youwei Shu, Shaomian Zheng, Dingnan Jin +5

Data quality is a crucial factor in large language models training. While prior work has shown that models trained on smaller, high-quality datasets can outperform those trained on…

stat.ME2024

Combining Incomplete Observational and Randomized Data for Heterogeneous Treatment Effects

Dong Yao, Caizhi Tang, Qing Cui +1

Data from observational studies (OSs) is widely available and readily obtainable yet frequently contains confounding biases. On the other hand, data derived from randomized control…

math.DG2017

On the volume of locally conformally flat 4 dimensional hypersphere

Qing Cui, Linlin Sun

Let be a 5 dimensional Riemannian manifold with , be a locally conformally flat hypersphere in with mean curvature . We prove that, there exists $\va…

cs.CL2026

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,…

math.DG2024

Minimal hypersurfaces in with constant scalar curvature and zero Gauss curvature are totally geodesic

Qing Cui

We show that a closed minimal hypersurface in with constant scalar curvature and zero Gauss curvature is totally geodesic.

cs.IR2021

Multi-Interactive Attention Network for Fine-grained Feature Learning in CTR Prediction

Kai Zhang, Hao Qian, Qing Cui +5

In the Click-Through Rate (CTR) prediction scenario, user's sequential behaviors are well utilized to capture the user interest in the recent literature. However, despite being ext…

math.DG2024

A new characterization for Clifford hypersurfaces

Qing Cui, Carlos Peñafiel

For a closed minimal immersed hypersurface in with second fundamental form , and each integer , define a constant $σ_k=\dfrac{\int_M (|A|^2)^k}{|M…

math.DG2025

Complete minimal hypersurfaces in with constant scalar curvature and zero Gauss-Kronecker curvature

Qing Cui, Boyuan Zhang

We show that any complete minimal hypersurface in the five-dimensional hyperbolic space , endowed with constant scalar curvature and vanishing Gauss-Kronecker curvatur…

cs.CL2026

GRIP: Geometric Refinement and Adaptive Information Potential for Data Efficiency

Changhao Wang, Jiaolong Yang, Xinhao Yao +7

The performance of Large Language Models (LLMs) is increasingly governed by data efficiency rather than raw scaling volume. However, existing selection methods often decouple globa…

cs.IR2023

Leveraging Large Language Models for Pre-trained Recommender Systems

Zhixuan Chu, Hongyan Hao, Xin Ouyang +9

Recent advancements in recommendation systems have shifted towards more comprehensive and personalized recommendations by utilizing large language models (LLM). However, effectivel…

cs.LG2025

Every FLOP Counts: Scaling a 300B Mixture-of-Experts LING LLM without Premium GPUs

Ling Team, Binwei Zeng, Chao Huang +71

In this technical report, we tackle the challenges of training large-scale Mixture of Experts (MoE) models, focusing on overcoming cost inefficiency and resource limitations preval…

cs.CL2014

KNET: A General Framework for Learning Word Embedding using Morphological Knowledge

Qing Cui, Bin Gao, Jiang Bian +2

Neural network techniques are widely applied to obtain high-quality distributed representations of words, i.e., word embeddings, to address text mining, information retrieval, and…

cs.LG2024

Task-Driven Causal Feature Distillation: Towards Trustworthy Risk Prediction

Zhixuan Chu, Mengxuan Hu, Qing Cui +2

Since artificial intelligence has seen tremendous recent successes in many areas, it has sparked great interest in its potential for trustworthy and interpretable risk prediction.…

math.CO2020

Tight gaps in the cycle spectrum of 3-connected planar graphs

Qing Cui, On-Hei Solomon Lo

For any positive integer , define (respectively, ) to be the minimal integer such that every 3-connected planar graph (respectively, 3-connected cubic…

cs.CL2023

Data-Centric Financial Large Language Models

Zhixuan Chu, Huaiyu Guo, Xinyuan Zhou +9

Large language models (LLMs) show promise for natural language tasks but struggle when applied directly to complex domains like finance. LLMs have difficulty reasoning about and in…

cs.LG2023

Enhancing Asynchronous Time Series Forecasting with Contrastive Relational Inference

Yan Wang, Zhixuan Chu, Tao Zhou +9

Asynchronous time series, also known as temporal event sequences, are the basis of many applications throughout different industries. Temporal point processes(TPPs) are the standar…

cs.AI2026

What Really Improves Mathematical Reasoning: Structured Reasoning Signals Beyond Pure Code

Yuze Zhao, Junpeng Fang, Lu Yu +6

Code has become a standard component of modern foundation language model (LM) training, yet its role beyond programming remains unclear. We revisit the claim that code improves rea…

cs.LG2023

Unleashing the Power of Graph Data Augmentation on Covariate Distribution Shift

Yongduo Sui, Qitian Wu, Jiancan Wu +5

The issue of distribution shifts is emerging as a critical concern in graph representation learning. From the perspective of invariant learning and stable learning, a recently well…

cs.IR2019

Value-aware Recommendation based on Reinforced Profit Maximization in E-commerce Systems

Changhua Pei, Xinru Yang, Qing Cui +5

Existing recommendation algorithms mostly focus on optimizing traditional recommendation measures, such as the accuracy of rating prediction in terms of RMSE or the quality of top-…

cs.CL2025

MASS: Mathematical Data Selection via Skill Graphs for Pretraining Large Language Models

Jiazheng Li, Lu Yu, Qing Cui +4

High-quality data plays a critical role in the pretraining and fine-tuning of large language models (LLMs), even determining their performance ceiling to some degree. Consequently,…

cs.LG2025

A Unified Invariant Learning Framework for Graph Classification

Yongduo Sui, Jie Sun, Shuyao Wang +4

Invariant learning demonstrates substantial potential for enhancing the generalization of graph neural networks (GNNs) with out-of-distribution (OOD) data. It aims to recognize sta…

cs.IR2024

Enhancing Recommender Systems with Large Language Model Reasoning Graphs

Yan Wang, Zhixuan Chu, Xin Ouyang +10

Recommendation systems aim to provide users with relevant suggestions, but often lack interpretability and fail to capture higher-level semantic relationships between user behavior…

math.CO2025

The saturation number of W 4

Ning Song, Jinze Hu, Shengjin Ji +1

For a fixed graph , a graph is called -saturated if does not contain as a (not necessarily induced) subgraph, but contains a copy of for any $e\in E(\ov…

cs.CL2025

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…

stat.ML2022

Robust Direct Learning for Causal Data Fusion

Xinyu Li, Yilin Li, Qing Cui +2

In the era of big data, the explosive growth of multi-source heterogeneous data offers many exciting challenges and opportunities for improving the inference of conditional average…

cs.AI2025

SHARP: Synthesizing High-quality Aligned Reasoning Problems for Large Reasoning Models Reinforcement Learning

Xiong Jun Wu, Zhenduo Zhang, ZuJie Wen +11

Training large reasoning models (LRMs) with reinforcement learning in STEM domains is hindered by the scarcity of high-quality, diverse, and verifiable problem sets. Existing synth…

cs.LG2025

The Debate on RLVR Reasoning Capability Boundary: Shrinkage, Expansion, or Both? A Two-Stage Dynamic View

Xinhao Yao, Lu Yu, Xiaolin Hu +4

The ongoing debate on whether reinforcement learning with verifiable rewards (RLVR) expands or shrinks the reasoning capabilities of large language models (LLMs) remains unresolved…

cs.CL2025

Every Activation Boosted: Scaling General Reasoner to 1 Trillion Open Language Foundation

Ling Team, Ang Li, Ben Liu +138

We introduce Ling 2.0, a series reasoning-oriented language foundation built upon the principle that every activation boosts reasoning capability. Designed to scale from tens of bi…

math.CO2025

Some results on minimum saturated graphs

Chenke Zhang, Qing Cui, Jinze Hu +2

Let be a graph and be a family of graphs. We say a graph is -saturated if does not contain any member in and for any $e\in E(\o…

cs.AI2026

Improving Autoformalization Using Direct Dependency Retrieval

Shaoqi Wang, Lu Yu, Siwei Lou +4

The convergence of deep learning and formal mathematics has spurred research in formal verification. Statement autoformalization, a crucial first step in this process, aims to tran…

cs.LG2025

Every Sample Matters: Leveraging Mixture-of-Experts and High-Quality Data for Efficient and Accurate Code LLM

Codefuse, Ling Team, : +30

Recent advancements in code large language models (LLMs) have demonstrated remarkable capabilities in code generation and understanding. It is still challenging to build a code LLM…

cs.CL2026

Mix Data or Merge Models? Balancing the Helpfulness, Honesty, and Harmlessness of Large Language Model via Model Merging

Jinluan Yang, Dingnan Jin, Anke Tang +10

Achieving balanced alignment of large language models (LLMs) in terms of Helpfulness, Honesty, and Harmlessness (3H optimization) constitutes a cornerstone of responsible AI. Exist…

math.DG2017

Optimal lower eigenvalue estimates for Hodge-Laplacian and applications

Qing Cui, Linlin Sun

In this paper, we consider the eigenvalue problem for Hodge-Laplacian on a Riemannian manifold isometrically immersed into another Riemannian manifold for arbitrary co…

math.DG2026

A strict upper volume bound for minimal graphs in the unit ball

Qing Cui

Let be a solution of the minimal surface equation on a domain containing the closed unit ball . A classical calibration argument gives $|Graph…

cs.CL2026

DiffScore: Text Evaluation Beyond Autoregressive Likelihood

Wen Lai, Yingli Shen, Dingnan Jin +4

Autoregressive language models are widely used for text evaluation, however, their left-to-right factorization introduces positional bias, i.e., early tokens are scored with only l…

math.DG2026

A curvature characterization of the Cartan minimal hypersurface in

Qing Cui

Lawson showed that a non-totally geodesic Einstein minimal hypersurface in is congruent to the Clifford hypersurface $\mathbb S^2(1/\sqrt2)\times \mathbb S^2(1/\sqrt2…

cs.CL2026

Improving Cross-Format Robustness in Language Models with Multi-Format Training

June M. Liu, Shaomian Zheng, He Cao +3

Large language models often remain sensitive to answer format: a question solved correctly in one form may fail in another semantically equivalent form. To study this gap, we defin…

math.OC2016

Second Order Necessary Conditions for Optimal Control Problems on Riemannian Manifolds

Qing Cui, Li Deng, Xu Zhang

This work is concerned with an optimal control problem on a Riemannian manifold, for which two typical cases are considered. The first case is when the endpoint is free. For this c…