arXiv subjects

Changcheng Li

Publications and source records attributed to Changcheng Li.

15 recordsLinked to original sources

arXiv2026

Recommendation as Generation: Unifying Personalized Video Generation and Recommendation at Industrial Scale

Traditional short-video recommendation systems match user interest to a fixed pool of pre-produced videos, which limits their ability to capture fine-grained and dynamic preferences. We propose Recommendation-as-Generation (RaG), a new paradigm that generates personalized videos on demand from inferred user interest. Our framework unifies generative recommendation and video generation through shared semantic IDs (SIDs), which disentangle video representation into content semantics and creative style semantics, enabling both fine-grained modeling of user interest and controllable generation of interest-aligned videos. We further develop Video Generation Agents (VGAs) that are conditioned on inferred SIDs to drive hierarchical planning and refinement for video creation, including visual composition, audio alignment, and artistic effect enhancement. To optimize the framework, we effectively introduce a synergistic cross-domain reward learning mechanism that jointly enforces interest alignment, user feedback, and video quality assessment. We deploy RaG on an industrial-scale platform with over 400 million daily active users and evaluate it in a revenue-critical advertising scenario. Online A/B tests show up to 1.87% ad revenue improvement compared to a strong production GRM baseline, demonstrating its effectiveness in driving further revenue gains beyond generative recommendation. Our results highlight a closed-loop generative system as a promising paradigm for integrating personalized video generation into recommendation.

cs.IR↗

arXiv2026

Accelerated Rydberg electromagnetically induced transparency quantum memory via shortcuts to adiabaticity

Electromagnetically induced transparency (EIT) enables coherent light-matter storage, forming the basis of photonic quantum memories that are essential for scalable quantum networks and distributed quantum computing. However, accelerating the storage process violates the adiabatic condition, resulting in the excitation of the lossy intermediate state and a reduction in writing efficiency. We propose and numerically investigate a high-speed, high-fidelity quantum storage scheme by incorporating a shortcut-to-adiabaticity (STA) technique based on counter-diabatic (CD) driving. By introducing a precisely engineered auxiliary field into a conventional EIT system, our protocol significantly shortens the writing time beyond the conventional adiabatic limit while effectively suppressing the transient population of the lossy intermediate state. Furthermore, our scheme demonstrates strong flexibility in pulse design, remaining effective across different temporal profiles of both the control and signal fields. It also exhibits robustness against imperfections in the CD drive. Even with imperfect single-photon writing and non-ideal Rydberg blockade, the scheme retains clear advantages, maintaining high storage performance and overcoming the intrinsic speed-fidelity trade-off of traditional EIT protocols. These features pave the way for fast and robust quantum devices suitable for high-throughput quantum repeaters and advanced quantum information processing.

quant-ph↗

arXiv2026

Confidence Before Answering: A Paradigm Shift for Efficient LLM Uncertainty Estimation

Reliable deployment of large language models (LLMs) requires accurate uncertainty estimation. Existing methods are predominantly answer-first, producing confidence only after generating an answer, which measure the correctness of a specific response and limits practical usability. We study a confidence-first paradigm, where the model outputs its confidence before answering, interpreting this score as the model's probability of answering the question correctly under its current policy. We propose CoCA(Co-optimized Confidence and Answers), a GRPO reinforcement learning framework that jointly optimizes confidence calibration and answer accuracy via segmented credit assignment. By assigning separate rewards and group-relative advantages to confidence and answer segments, CoCA enables stable joint optimization and avoids reward hacking. Experiments across math, code, and factual QA benchmarks show improved calibration and uncertainty discrimination while preserving answer quality, thereby enabling a broader range of downstream applications.

cs.CL↗

arXiv2026

Generative Recommendation for Large-Scale Advertising

Generative recommendation has recently attracted widespread attention in industry due to its potential for scaling and stronger model capacity. However, deploying real-time generative recommendation in large-scale advertising requires designs beyond large-language-model (LLM)-style training and serving recipes. We present a production-oriented generative recommender co-designed across architecture, learning, and serving, named GR4AD (Generative Recommendation for ADdvertising). As for tokenization, GR4AD proposes UA-SID (Unified Advertisement Semantic ID) to capture complicated business information. Furthermore, GR4AD introduces LazyAR, a lazy autoregressive decoder that relaxes layer-wise dependencies for short, multi-candidate generation, preserving effectiveness while reducing inference cost, which facilitates scaling under fixed serving budgets. To align optimization with business value, GR4AD employs VSL (Value-Aware Supervised Learning) and proposes RSPO (Ranking-Guided Softmax Preference Optimization), a ranking-aware, list-wise reinforcement learning algorithm that optimizes value-based rewards under list-level metrics for continual online updates. For online inference, we further propose dynamic beam serving, which adapts beam width across generation levels and online load to control compute. Large-scale online A/B tests show up to 4.2% ad revenue improvement over an existing DLRM-based stack, with consistent gains from both model scaling and inference-time scaling. GR4AD has been fully deployed in Kuaishou advertising system with over 400 million users and achieves high-throughput real-time serving.

cs.IR↗

arXiv2026

Punctuation-aware Hybrid Trainable Sparse Attention for Large Language Models

Attention serves as the fundamental mechanism for long-context modeling in large language models (LLMs), yet dense attention becomes structurally prohibitive for long sequences due to its quadratic complexity. Consequently, sparse attention has received increasing attention as a scalable alternative. However, existing sparse attention methods rely on coarse-grained semantic representations during block selection, which blur intra-block semantic boundaries and lead to the loss of critical information. To address this issue, we propose \textbf{P}unctuation-aware \textbf{H}ybrid \textbf{S}parse \textbf{A}ttention \textbf{(PHSA)}, a natively trainable sparse attention framework that leverages punctuation tokens as semantic boundary anchors. Specifically, (1) we design a dual-branch aggregation mechanism that fuses global semantic representations with punctuation-enhanced boundary features, preserving the core semantic structure while introducing almost no additional computational overhead; (2) we introduce an extreme-sparsity-adaptive training and inference strategy that stabilizes model behavior under very low token activation ratios; Extensive experiments on general benchmarks and long-context evaluations demonstrate that PHSA consistently outperforms dense attention and state-of-the-art sparse attention baselines, including InfLLM v2. Specifically, for the 0.6B-parameter model with 32k-token input sequences, PHSA can reduce the information loss by 10.8\% at a sparsity ratio of 97.3\%.

cs.CL↗

arXiv2025

Personalized Tree-Based Progressive Regression Model for Watch-Time Prediction in Short Video Recommendation

In online video platforms, accurate watch time prediction has become a fundamental and challenging problem in video recommendation. Previous research has revealed that the accuracy of watch time prediction highly depends on both the transformation of watch-time labels and the decomposition of the estimation process. TPM (Tree based Progressive Regression Model) achieves State-of-the-Art performance with a carefully designed and effective decomposition paradigm. TPM discretizes the watch time into several ordinal intervals and organizes them into a binary decision tree, where each node corresponds to a specific interval. At each non-leaf node, a binary classifier is used to determine the specific interval in which the watch time variable most likely falls, based on the prediction outcome at its parent node. The tree structure is central to TPM, as it defines the decomposition of watch time estimation and how ordinal intervals are discretized. However, TPM uses a predefined full binary tree, which may be sub-optimal for two reasons. First, full binary trees imply equal partitioning of the watch time space, which may fail to capture the complexity of real-world distributions. Second, rather than relying on a fixed global structure, we advocate for a personalized, data-driven tree that can be learned end-to-end. Thus, we propose PTPM to enable highly personalized decomposition of watch estimation with better efficacy and efficiency. Moreover, we show that TPM suffers from selection bias due to conditional modeling and propose a simple solution. We conduct extensive experiments on offline datasets and online environments. Offline results show improved watch time accuracy, and online A/B tests further validate the effectiveness of our framework. PTPM has been fully deployed in core traffic scenarios and now serves over 400 million users daily.

cs.IR↗

arXiv2025

Direction switchable single-photon emitter using a Rydberg polariton

All-optical redirection or routing of single photons is essential for quantum networks. Although studied in various systems both in theory and experiment, the redirection of single photons with many output ports, compatible with large-scale photonic circuits, still needs to be explored. Here, we demonstrate a direction switchable single-photon emitter using a Rydberg polariton. The Rydberg component of the stored photon is changed using a stimulated Raman transition with a specific intermediate state. By adjusting the direction of the retrieval laser, we can redirect the emitted photon into a rich variety of alternative modes. Building upon this scheme, we propose a quantum routing of single photons with \textit{N} output channels and unity routing efficiency. In addition, the protocol reduces the effect of motional dephasing increasing the photon lifetime to $>10~\mu$s ($>20$ times photon processing time), enabling functional quantum devices based on Rydberg polaritons.

quant-ph↗

arXiv2024

MTMT: Consolidating Multiple Thinking Modes to Form a Thought Tree for Strengthening LLM

Large language models (LLMs) have shown limitations in tasks requiring complex logical reasoning and multi-step problem-solving. To address these challenges, researchers have employed carefully designed prompts and flowcharts, simulating human cognitive processes to enhance LLM performance, such as the Chain of Thought approach. In this paper, we introduce MTMT (Multi-thinking Modes Tree), a novel method that interacts with LLMs to construct a thought tree, simulating various advanced cognitive processes, including but not limited to association, counterfactual thinking, task decomposition, and comparison. By breaking down the original complex task into simpler sub-questions, MTMT facilitates easier problem-solving for LLMs, enabling more effective utilization of the latent knowledge within LLMs. We evaluate the performance of MTMT under different parameter configurations, using GPT-4o mini as the base model. Our results demonstrate that integrating multiple modes of thinking significantly enhances the ability of LLMs to handle complex tasks.

cs.CL↗

arXiv2024

LDACP: Long-Delayed Ad Conversions Prediction Model for Bidding Strategy

In online advertising, once an ad campaign is deployed, the automated bidding system dynamically adjusts the bidding strategy to optimize Cost Per Action (CPA) based on the number of ad conversions. For ads with a long conversion delay, relying solely on the real-time tracked conversion number as a signal for bidding strategy can significantly overestimate the current CPA, leading to conservative bidding strategies. Therefore, it is crucial to predict the number of long-delayed conversions. Nonetheless, it is challenging to predict ad conversion numbers through traditional regression methods due to the wide range of ad conversion numbers. Previous regression works have addressed this challenge by transforming regression problems into bucket classification problems, achieving success in various scenarios. However, specific challenges arise when predicting the number of ad conversions: 1) The integer nature of ad conversion numbers exacerbates the discontinuity issue in one-hot hard labels; 2) The long-tail distribution of ad conversion numbers complicates tail data prediction. In this paper, we propose the Long-Delayed Ad Conversions Prediction model for bidding strategy (LDACP), which consists of two sub-modules. To alleviate the issue of discontinuity in one-hot hard labels, the Bucket Classification Module with label Smoothing method (BCMS) converts one-hot hard labels into non-normalized soft labels, then fits these soft labels by minimizing classification loss and regression loss. To address the challenge of predicting tail data, the Value Regression Module with Proxy labels (VRMP) uses the prediction bias of aggregated pCTCVR as proxy labels. Finally, a Mixture of Experts (MoE) structure integrates the predictions from BCMS and VRMP to obtain the final predicted ad conversion number.

cs.LG↗

arXiv2024

Suppression of motional dephasing using state mapping

Rydberg-mediated quantum optics is a useful route toward deterministic quantum information processing based on single photons and quantum networks, but is bottlenecked by the fast motional dephasing of Rydberg atoms. Here, we propose and experimentally demonstrate suppressing the motional dephasing by creating an {\it a priori} unknown but correct phase to each Rydberg atom in an atomic ensemble. The phase created is exactly proportional to the unknown velocity of the thermal motion, resulting in a condition as if no thermal motion occurs to the Rydberg atom upon the retrieval of the signal photon. Our experiments, though hampered by the noise of lasers and the environment, demonstrate more than one order of magnitude enhancement of the coherence time. The feasibility of realizing long-lived storage of single photons in strongly interacting Rydberg media sheds new light on Rydberg-mediated quantum nonlinear optics.

quant-ph↗

arXiv2022

Causal Structural Learning on MPHIA Individual Dataset

The Population-based HIV Impact Assessment (PHIA) is an ongoing project that conducts nationally representative HIV-focused surveys for measuring national and regional progress toward UNAIDS' 90-90-90 targets, the primary strategy to end the HIV epidemic. We believe the PHIA survey offers a unique opportunity to better understand the key factors that drive the HIV epidemics in the most affected countries in sub-Saharan Africa. In this article, we propose a novel causal structural learning algorithm to discover important covariates and potential causal pathways for 90-90-90 targets. Existing constrained-based causal structural learning algorithms are quite aggressive in edge removal. The proposed algorithm preserves more information about important features and potential causal pathways. It is applied to the Malawi PHIA (MPHIA) data set and leads to interesting results. For example, it discovers age and condom usage to be important for female HIV awareness; the number of sexual partners to be important for male HIV awareness; and knowing the travel time to HIV care facilities leads to a higher chance of being treated for both females and males. We further compare and validate the proposed algorithm using BIC and using Monte Carlo simulations, and show that the proposed algorithm achieves improvement in true positive rates in important feature discovery over existing algorithms.

stat.ME↗

arXiv2021

Graph Neural Networks: A Review of Methods and Applications

Lots of learning tasks require dealing with graph data which contains rich relation information among elements. Modeling physics systems, learning molecular fingerprints, predicting protein interface, and classifying diseases demand a model to learn from graph inputs. In other domains such as learning from non-structural data like texts and images, reasoning on extracted structures (like the dependency trees of sentences and the scene graphs of images) is an important research topic which also needs graph reasoning models. Graph neural networks (GNNs) are neural models that capture the dependence of graphs via message passing between the nodes of graphs. In recent years, variants of GNNs such as graph convolutional network (GCN), graph attention network (GAT), graph recurrent network (GRN) have demonstrated ground-breaking performances on many deep learning tasks. In this survey, we propose a general design pipeline for GNN models and discuss the variants of each component, systematically categorize the applications, and propose four open problems for future research.

cs.LG↗

arXiv2021

Probabilistic HIV Recency Classification -- A Logistic Regression without Labeled Individual Level Training Data

Accurate HIV incidence estimation based on individual recent infection status (recent vs long-term infection) is important for monitoring the epidemic, targeting interventions to those at greatest risk of new infection, and evaluating existing programs of prevention and treatment. Starting from 2015, the Population-based HIV Impact Assessment (PHIA) individual-level surveys are implemented in the most-affected countries in sub-Saharan Africa. PHIA is a nationally-representative HIV-focused survey that combines household visits with key questions and cutting-edge technologies such as biomarker tests for HIV antibody and HIV viral load which offer the unique opportunity of distinguishing between recent infection and long-term infection, and providing relevant HIV information by age, gender, and location. In this article, we propose a semi-supervised logistic regression model for estimating individual level HIV recency status. It incorporates information from multiple data sources -- the PHIA survey where the true HIV recency status is unknown, and the cohort studies provided in the literature where the relationship between HIV recency status and the covariates are presented in the form of a contingency table. It also utilizes the national level HIV incidence estimates from the epidemiology model. Applying the proposed model to Malawi PHIA data, we demonstrate that our approach is more accurate for the individual level estimation and more appropriate for estimating HIV recency rates at aggregated levels than the current practice -- the binary classification tree (BCT).

stat.ME↗

arXiv2020

Approximate Cross-validated Mean Estimates for Bayesian Hierarchical Regression Models

We introduce a novel procedure for obtaining cross-validated predictive estimates for Bayesian hierarchical regression models (BHRMs). Bayesian hierarchical models are popular for their ability to model complex dependence structures and provide probabilistic uncertainty estimates, but can be computationally expensive to run. Cross-validation (CV) is therefore not a common practice to evaluate the predictive performance of BHRMs. Our method circumvents the need to re-run computationally costly estimation methods for each cross-validation fold and makes CV more feasible for large BHRMs. By conditioning on the variance-covariance parameters, we shift the CV problem from probability-based sampling to a simple and familiar optimization problem. In many cases, this produces estimates which are equivalent to full CV. We provide theoretical results and demonstrate its efficacy on publicly available data and in simulations.

stat.ML↗

arXiv2019

GEAR: Graph-based Evidence Aggregating and Reasoning for Fact Verification

Fact verification (FV) is a challenging task which requires to retrieve relevant evidence from plain text and use the evidence to verify given claims. Many claims require to simultaneously integrate and reason over several pieces of evidence for verification. However, previous work employs simple models to extract information from evidence without letting evidence communicate with each other, e.g., merely concatenate the evidence for processing. Therefore, these methods are unable to grasp sufficient relational and logical information among the evidence. To alleviate this issue, we propose a graph-based evidence aggregating and reasoning (GEAR) framework which enables information to transfer on a fully-connected evidence graph and then utilizes different aggregators to collect multi-evidence information. We further employ BERT, an effective pre-trained language representation model, to improve the performance. Experimental results on a large-scale benchmark dataset FEVER have demonstrated that GEAR could leverage multi-evidence information for FV and thus achieves the promising result with a test FEVER score of 67.10%. Our code is available at https://github.com/thunlp/GEAR.

cs.CL↗