arXiv subjects

Liqun Liu

Publications and source records attributed to Liqun Liu.

At least 19 recordsLinked to original sources

arXiv2026

R^3: Advertisement Compliance Rectification via Group-Relative Experience Extractor and Curriculum Reinforcement

Rigorous content moderation is crucial for online advertising but leads to millions of daily rejections. This scale renders manual rectification infeasible, particularly for video advertisements. However, existing safety-driven methods often suffer from aggressive over-editing, which compromises the advertiser's original semantic intent merely to satisfy compliance. In this work, we target the rectification of textual violations in video ads, covering both speech transcripts and on-screen text. We propose R^3, a novel framework designed to harmonize compliance with original semantic intent preservation. Our approach integrates three key innovations: (1) an experience-driven data synthesis framework that bootstraps high-quality supervision via a group-Relative compliance experience extractor; (2) a curriculum Reinforcement learning strategy with hierarchical rewards designed to enforce compliance while maximizing semantic consistency; and (3) a comprehensive video Rectification framework seamlessly integrating text recognition, rewriting, and re-rendering for industrial deployment. Extensive experiments on industrial datasets and online A/B testing demonstrate that R^3 significantly outperforms state-of-the-art baselines, achieving an optimal trade-off between violation rectification and intent preservation.

cs.CL↗

arXiv2026

Decoding Multimodal Cues: Unveiling the Implicit Meaning Behind Hateful Videos

Hateful videos have become prevalent on online platforms, highlighting an urgent need for effective detection. However, existing studies primarily focus on binary classification and fail to provide contextual rationales that reveal the implicit meanings behind these judgments, significantly undermining model explainability. To fill this gap, we aim to achieve explainable hateful video detection, enabling models to provide contextual rationales that integrate relevant evidence and logical reasoning alongside decisions. This approach can comprehensively enhance the understanding of video content and the explainability of the decision-making process. We first introduce two datasets, Ex-HateMM and Ex-ImpliHateVid, for explainable hateful video detection. Each dataset provides fine-grained annotations of multimodal harmful elements, along with contextual rationales. We then propose an Information Augmentation and Reasoning Enhancement (IARE) framework designed for explainable detection. The framework employs an information augmentation phase that leverages the multimodal chain-of-thought to integrate harmful elements, thereby enriching rationale evidence. Additionally, IARE incorporates a reasoning enhancement phase, in which Direct Preference Optimization guides the model toward correct reasoning paths and away from incorrect ones, thereby improving the logical coherence of its justifications. We conduct extensive experiments on the two datasets, comparing multiple baselines with our proposed IARE framework. The results demonstrate that IARE achieves state-of-the-art performance while also generating accurate rationales.

cs.CL↗

arXiv2026

CamGeo: Sparse Camera-Conditioned Image-to-Video Generation with 3D Geometry Priors

Sparse camera-conditioned image-to-video generation presents a pivotal challenge: synthesizing geometrically consistent 3D motion from minimal pose cues. Existing methods, which largely rely on dense supervision or naive interpolation, suffer from severe pose drift and motion discontinuities due to the lack of robust 3D priors. In this paper, we introduce CamGeo, a novel framework that distills rich 3D geometric knowledge from a pre-trained video-to-3D model (VGGT) directly into the diffusion backbone. To achieve this without incurring inference latency, we propose a training-only distillation strategy. Specifically, CamGeo incorporates: (1) keyframe trajectory distillation that enforces cycle-consistency with sparse input poses, (2) cross-frame consistency distillation with both camera trajectory and depth constraints to generate consistent structure across unsupervised frames, and (3) a three-stage coarse-to-fine curriculum learning, progressively scales geometric complexity, from global structure coherence to fine-grained refinement, achieving stable optimization. Extensive experiments demonstrate that CamGeo achieves consistent improvements under various sparsity ratios.

cs.CE↗

arXiv2026

Motivating Careerists

Motivating careerists is challenging for political organizations. Without explicit contracts, careerists often pander to public opinions or their superiors' preferences. Worse, when tasked with implementing these distorted decisions, they tend to underinvest in the necessary efforts. We analyze the motivation problem by examining how a careerist agent fulfills these roles on behalf of a principal across various information structures. Importantly, the principal can credibly commit to performance-based reward schemes to incentivize correct decisions and diligent implementation. However, such schemes are feasible only if the principal observes policy consequences while backing away from implementation details. Along the way, we characterize the principal-optimal information structure. Putting theoretical findings into practice, we explore the underlying incentive structures and their policy implications.

econ.GN↗

arXiv2026

ARGUS: Policy-Adaptive Ad Governance via Evolving Reinforcement with Adversarial Umpiring

Online advertising governance faces significant challenges due to the non-stationary nature of regulatory policies, where emerging mandates (e.g., restrictions on education or aesthetic anxiety) create severe label inconsistencies and reasoning ambiguities in historical datasets. In this paper, we propose ARGUS, a policy-adaptive governance system that enables evolving reinforcement through multi-agent adversarial umpiring. ARGUS addresses the sparsity of new policy data by employing a three-stage framework: (1) Policy Seeding for initial perception; (2) Adversarial Label Rectification, which utilizes a ``Prosecutor-Defender-Umpire'' architecture to resolve conflicts between stale labels and new mandates; and (3) Latent Knowledge Discovery, which employs a tripartite dialectical discussion to unearth sophisticated, ``gray-area'' violations. By leveraging RAG-enhanced policy knowledge and Chain-of-Thought synthesis as dynamic rewards for reinforcement learning, ARGUS synchronizes its reasoning pathways with evolving regulations. Extensive experiments on both industrial and public datasets demonstrate that ARGUS significantly outperforms traditional fine-tuning baselines, achieving superior policy-adaptive learning with minimal gold data.

cs.CL↗

arXiv2026

Disentangling to Re-couple: Resolving the Similarity-Controllability Paradox in Subject-Driven Text-to-Image Generation

Subject-Driven Text-to-Image (T2I) Generation aims to preserve a subject's identity while editing its context based on a text prompt. A core challenge in this task is the "similarity-controllability paradox", where enhancing textual control often degrades the subject's fidelity, and vice-versa. We argue this paradox stems from the ambiguous role of text prompts, which are often tasked with describing both the subject and the desired modifications, leading to conflicting signals for the model. To resolve this, we propose DisCo, a novel framework that first Disntangles and then re-Couples visual and textual information. First, our textual-visual decoupling module isolates the sources of information: subject identity is extracted exclusively from the reference image with the entity word of the subject, while the text prompt is simplified to contain only the modification command, where the subject refers to general pronouns, eliminating descriptive ambiguity. However, this strict separation can lead to unnatural compositions between the subject and its contexts. We address this by designing a dedicated reward signal and using reinforcement learning to seamlessly recouple the visually-defined subject and the textually-generated context. Our approach effectively resolves the paradox, enabling simultaneous high-fidelity subject preservation and precise textual control. Extensive experiments demonstrate that our method achieves state-of-the-art performance, producing highly realistic and coherent images.

cs.CV↗

arXiv2026

DSH-Bench: A Difficulty- and Scenario-Aware Benchmark with Hierarchical Subject Taxonomy for Subject-Driven Text-to-Image Generation

Significant progress has been achieved in subject-driven text-to-image (T2I) generation, which aims to synthesize new images depicting target subjects according to user instructions. However, evaluating these models remains a significant challenge. Existing benchmarks exhibit critical limitations: 1) insufficient diversity and comprehensiveness in subject images, 2) inadequate granularity in assessing model performance across different subject difficulty levels and prompt scenarios, and 3) a profound lack of actionable insights and diagnostic guidance for subsequent model refinement. To address these limitations, we propose DSH-Bench, a comprehensive benchmark that enables systematic multi-perspective analysis of subject-driven T2I models through four principal innovations: 1) a hierarchical taxonomy sampling mechanism ensuring comprehensive subject representation across 58 fine-grained categories, 2) an innovative classification scheme categorizing both subject difficulty level and prompt scenario for granular capability assessment, 3) a novel Subject Identity Consistency Score (SICS) metric demonstrating a 9.4\% higher correlation with human evaluation compared to existing measures in quantifying subject preservation, and 4) a comprehensive set of diagnostic insights derived from the benchmark, offering critical guidance for optimizing future model training paradigms and data construction strategies. Through an extensive empirical evaluation of 19 leading models, DSH-Bench uncovers previously obscured limitations in current approaches, establishing concrete directions for future research and development.

cs.CV↗

arXiv2026

Search-P1: Path-Centric Reward Shaping for Stable and Efficient Agentic RAG Training

Retrieval-Augmented Generation (RAG) enhances large language models (LLMs) by incorporating external knowledge, yet traditional single-round retrieval struggles with complex multi-step reasoning. Agentic RAG addresses this by enabling LLMs to dynamically decide when and what to retrieve, but current RL-based training methods suffer from sparse outcome rewards that discard intermediate signals and low sample efficiency where failed samples contribute nothing. We propose Search-P1, a framework that introduces path-centric reward shaping for agentic RAG training, comprising two key components: (1) Path-Centric Reward, which evaluates the structural quality of reasoning trajectories through order-agnostic step coverage and soft scoring that extracts learning signals even from failed samples, and (2) Dual-Track Path Scoring with offline-generated reference planners that assesses paths from both self-consistency and reference-alignment perspectives. Experiments on multiple QA benchmarks demonstrate that Search-P1 achieves significant improvements over Search-R1 and other strong baselines, with an average accuracy gain of 7.7 points.

cs.CL↗

arXiv2026

Towards Faithful Industrial RAG: A Reinforced Co-adaptation Framework for Advertising QA

Industrial advertising question answering (QA) is a high-stakes task in which hallucinated content, particularly fabricated URLs, can lead to financial loss, compliance violations, and legal risk. Although Retrieval-Augmented Generation (RAG) is widely adopted, deploying it in production remains challenging because industrial knowledge is inherently relational, frequently updated, and insufficiently aligned with generation objectives. We propose a reinforced co-adaptation framework that jointly optimizes retrieval and generation through two components: (1) Graph-aware Retrieval (GraphRAG), which models entity-relation structure over a high-citation knowledge subgraph for multi-hop, domain-specific evidence selection; and (2) evidence-constrained reinforcement learning via Group Relative Policy Optimization (GRPO) with multi-dimensional rewards covering faithfulness, style compliance, safety, and URL validity. Experiments on an internal advertising QA dataset show consistent gains across expert-judged dimensions including accuracy, completeness, and safety, while reducing the hallucination rate by 72\%. A two-week online A/B test demonstrates a 28.6\% increase in like rate, a 46.2\% decrease in dislike rate, and a 92.7\% reduction in URL hallucination. The system has been running in production for over half a year and has served millions of QA interactions.

cs.CL↗

arXiv2025

RAVEN++: Pinpointing Fine-Grained Violations in Advertisement Videos with Active Reinforcement Reasoning

Advertising (Ad) is a cornerstone of the digital economy, yet the moderation of video advertisements remains a significant challenge due to their complexity and the need for precise violation localization. While recent advancements, such as the RAVEN model, have improved coarse-grained violation detection, critical gaps persist in fine-grained understanding, explainability, and generalization. To address these limitations, we propose RAVEN++, a novel framework that introduces three key innovations: 1) Active Reinforcement Learning (RL), which dynamically adapts training to samples of varying difficulty; 2) Fine-Grained Violation Understanding, achieved through hierarchical reward functions and reasoning distillation; and 3) Progressive Multi-Stage Training, which systematically combines knowledge injection, curriculum-based passive RL, and active RL. Extensive experiments on both public and proprietary datasets, on both offline scenarios and online deployed A/B Testing, demonstrate that RAVEN++ outperforms general-purpose LLMs and specialized models like RAVEN in terms of fine-grained violation understanding, reasoning capabilities, and generalization ability.

cs.LG↗

arXiv2025

Multi-Agent VLMs Guided Self-Training with PNU Loss for Low-Resource Offensive Content Detection

Accurate detection of offensive content on social media demands high-quality labeled data; however, such data is often scarce due to the low prevalence of offensive instances and the high cost of manual annotation. To address this low-resource challenge, we propose a self-training framework that leverages abundant unlabeled data through collaborative pseudo-labeling. Starting with a lightweight classifier trained on limited labeled data, our method iteratively assigns pseudo-labels to unlabeled instances with the support of Multi-Agent Vision-Language Models (MA-VLMs). Un-labeled data on which the classifier and MA-VLMs agree are designated as the Agreed-Unknown set, while conflicting samples form the Disagreed-Unknown set. To enhance label reliability, MA-VLMs simulate dual perspectives, moderator and user, capturing both regulatory and subjective viewpoints. The classifier is optimized using a novel Positive-Negative-Unlabeled (PNU) loss, which jointly exploits labeled, Agreed-Unknown, and Disagreed-Unknown data while mitigating pseudo-label noise. Experiments on benchmark datasets demonstrate that our framework substantially outperforms baselines under limited supervision and approaches the performance of large-scale models

cs.LG↗

arXiv2024

Strengthened Symbol Binding Makes Large Language Models Reliable Multiple-Choice Selectors

Multiple-Choice Questions (MCQs) constitute a critical area of research in the study of Large Language Models (LLMs). Previous works have investigated the selection bias problem in MCQs within few-shot scenarios, in which the LLM's performance may be influenced by the presentation of answer choices, leaving the selection bias during Supervised Fine-Tuning (SFT) unexplored. In this paper, we reveal that selection bias persists in the SFT phase , primarily due to the LLM's inadequate Multiple Choice Symbol Binding (MCSB) ability. This limitation implies that the model struggles to associate the answer options with their corresponding symbols (e.g., A/B/C/D) effectively. To enhance the model's MCSB capability, we first incorporate option contents into the loss function and subsequently adjust the weights of the option symbols and contents, guiding the model to understand the option content of the current symbol. Based on this, we introduce an efficient SFT algorithm for MCQs, termed Point-wise Intelligent Feedback (PIF). PIF constructs negative instances by randomly combining the incorrect option contents with all candidate symbols, and proposes a point-wise loss to provide feedback on these negative samples into LLMs. Our experimental results demonstrate that PIF significantly reduces the model's selection bias by improving its MCSB capability. Remarkably, PIF exhibits a substantial enhancement in the accuracy for MCQs.

cs.CL↗

arXiv2024

Enhancing Reinforcement Learning with Label-Sensitive Reward for Natural Language Understanding

Recent strides in large language models (LLMs) have yielded remarkable performance, leveraging reinforcement learning from human feedback (RLHF) to significantly enhance generation and alignment capabilities. However, RLHF encounters numerous challenges, including the objective mismatch issue, leading to suboptimal performance in Natural Language Understanding (NLU) tasks. To address this limitation, we propose a novel Reinforcement Learning framework enhanced with Label-sensitive Reward (RLLR) to amplify the performance of LLMs in NLU tasks. By incorporating label-sensitive pairs into reinforcement learning, our method aims to adeptly capture nuanced label-sensitive semantic features during RL, thereby enhancing natural language understanding. Experiments conducted on five diverse foundation models across eight tasks showcase promising results. In comparison to Supervised Fine-tuning models (SFT), RLLR demonstrates an average performance improvement of 1.54%. Compared with RLHF models, the improvement averages at 0.69%. These results reveal the effectiveness of our method for LLMs in NLU tasks. Code and data available at: https://github.com/MagiaSN/ACL2024_RLLR.

cs.CL↗

arXiv2023

TencentPretrain: A Scalable and Flexible Toolkit for Pre-training Models of Different Modalities

Recently, the success of pre-training in text domain has been fully extended to vision, audio, and cross-modal scenarios. The proposed pre-training models of different modalities are showing a rising trend of homogeneity in their model structures, which brings the opportunity to implement different pre-training models within a uniform framework. In this paper, we present TencentPretrain, a toolkit supporting pre-training models of different modalities. The core feature of TencentPretrain is the modular design. The toolkit uniformly divides pre-training models into 5 components: embedding, encoder, target embedding, decoder, and target. As almost all of common modules are provided in each component, users can choose the desired modules from different components to build a complete pre-training model. The modular design enables users to efficiently reproduce existing pre-training models or build brand-new one. We test the toolkit on text, vision, and audio benchmarks and show that it can match the performance of the original implementations.

cs.CL↗

arXiv2023

Inefficient Peace or Preventive War?

We study a model of two-player bargaining game in the shadow of a preventive trade war that examines why states deliberately maintain trade barriers in the age of globalization. Globalization can induce substantial power shifts between states, which makes the threat of a preventive trade war salient. In this situation, there may exist "healthy" levels of trade barriers that dampen the war incentives by reducing states' expected payoffs from such a war. Thus, we demonstrate that trade barriers can sometimes serve as breaks and cushions necessary to sustain inefficient yet peaceful economic cooperation between states. We assess the theoretical implications by examining the US-China trade relations since 1972.

econ.GN↗

arXiv2022

Mixture of Virtual-Kernel Experts for Multi-Objective User Profile Modeling

In industrial applications like online advertising and recommendation systems, diverse and accurate user profiles can greatly help improve personalization. Deep learning is widely applied to mine expressive tags to users from their historical interactions in the system, e.g., click, conversion action in the advertising chain. The usual approach is to take a certain action as the objective, and introduce multiple independent Two-Tower models to predict the possibility of users' action on tags (known as CTR or CVR prediction). The predicted users' high probably attractive tags are to represent their preferences. However, the single-action models cannot learn complementarily and support effective training on data-sparse actions. Besides, limited by the lack of information fusion between the two towers, the model learns insufficiently to represent users' preferences on various tag \textbf{topics} well. This paper introduces a novel multi-task model called Mixture of Virtual-Kernel Experts (MVKE) to learn user preferences on various actions and topics unitedly. In MVKE, we propose a concept of Virtual-Kernel Expert, which focuses on modeling one particular facet of the user's preferences, and all of them learn coordinately. Besides, the gate-based structure used in MVKE builds an information fusion bridge between two towers, improving the model's capability and maintaining high efficiency. We apply the model in Tencent Advertising System, where both online and offline evaluations show that our method has a significant improvement compared with the existing ones and brings about an obvious lift to actual advertising revenue.

cs.IR↗

arXiv2021

Domestic Constraints in Crisis Bargaining

I study how political bias and audience costs impose domestic institutional constraints that affect states' capacity to reach peaceful agreements during crises. With a mechanism design approach, I show that the existence of peaceful agreements hinges crucially on whether the resource being divided can appease two sides of the highest type (i.e. the maximum war capacity). The derivation has two major implications. On the one hand, if war must be averted, then political leaders are not incentivized by audience costs to communicate private information; they will pool on the strategy that induces the maximum bargaining gains. On the other hand, political bias matters for the scope of peace because it alters a state's expected war payoff.

econ.GN↗

arXiv2021

Constraining to Motivate

Effective decision-making often involves a principal delegating authority to better-informed agents. However, when agents'careers depend on how their decisions are perceived, they may avoid acquiring costly information and instead choose politically safe actions. We propose a simple solution: constrained delegation. By restricting the agent's choice set, the principal can reshape the reputational stakes attached to the remaining options and restore incentives to acquire information. The mechanism applies broadly to settings where political or market-based incentives discourage information acquisition and shows how limiting discretion can, counterintuitively, improve decision quality.

econ.GN↗