arXiv subjects

Jie Jiang

Publications and source records attributed to Jie Jiang.

At least 37 records · Page 2Linked to original sources

arXiv2026

UniVA: Unified Value Alignment for Generative Recommendation in Online Advertising at Tencent

Generative Recommendation (GR) reformulates recommendation as next-token generation over item Semantic IDs (SIDs) and has shown promise in industrial applications. However, extending GR to advertising is non-trivial, as advertising recommendation must jointly account for user relevance and commercial value. This creates a mismatch: high generation likelihood does not necessarily imply high advertising utility. As a result, valuable ads may be poorly distinguished in the SID space, pruned during autoregressive decoding, or missed when request-invalid branches consume limited beam capacity during online serving. To address this problem, we propose UniVA, a Unified Value Alignment framework for generative advertising recommendation. UniVA aligns commercial value across the entire pipeline of SID construction, autoregressive decoding, and online serving. Commercial SID Tokenization injects business attributes and bid information into SID construction. A Generation-as-Ranking SID Decoder then fuses generation scores with token-level value estimates during autoregressive decoding. {Finally, Value-Aware Constrained Serving restricts the fused decoding process to request-valid SID paths through a personalized trie.} Experiments on the Tencent WeChat Channels advertising platform show that UniVA achieves a 37.04\% relative improvement in offline Hit Rate@100 over the baseline and lifts gross merchandise value (GMV) by 1.5\% in online A/B tests.

cs.IR↗

arXiv2026

The User Asks, Platforms Compete: How Agentic Recommendation Markets Take Shape

Online recommendation has traditionally taken place after a user enters a platform, which determines the candidate pool and the ranking shown to the user. LLM-based user agents enable a different recommendation process: a user specifies a need before choosing a platform, leaving platforms to compete for the user's attention, which we refer to as an agentic recommendation market. In our controlled LLM-based experiments across three product domains, we find this new setting of recommendation creates a tension between access and attention. Compared with traditional platform-centric recommendation, user-centric recommendation greatly expands the opportunity for relevant items to enter comparison; yet broader participation does not translate directly into effective exposure. Competition directly triggers platforms' strategic play: selectively positive explanations occupy 73--78% of first-ranked positions. When the user agent relates platforms' actions to subsequent user feedback, this share falls to 36--41%, while the chance of a user purchasing the relevant item increases. A user agent is therefore more than a ranker over a larger pool of candidates: its querying, ranking, and feedback mechanism governing who can compete, how scarce attention is allocated, and how earlier outcomes shape the evaluation of platforms directly affect user utility. Designing agentic recommendation therefore requires treating access, attention, and accountability as a joint mechanism design problem.

cs.AI↗

arXiv2026

Beyond Action Imitation: Learning a Decision-Aware User Simulator for Online Advertising

Recent advances in LLM-based user simulation have shown promise for offline evaluation of recommendation and advertising systems. However, existing simulators typically infer user preferences from single-domain interaction histories and are primarily optimized to reproduce observable actions such as clicks. Consequently, they capture only a partial view of user preferences, while action-only prediction easily induces model shortcuts and limits both the fidelity and diagnostic value of simulation. To address these challenges, we propose DASH, a decision-aware user simulator that jointly generates thinking traces and predicts behavioral actions from heterogeneous cross-domain histories. DASH first introduces a Context Engineering stage that folds heterogeneous cross-domain histories into decision-relevant context, together with prompt optimization for effective reasoning over the folded context. To train a user simulator, DASH distills thinking trajectories from strong LLMs as SFT data, and further tailors a rubric-based reward model that evaluates thinking traces along form, content, and logic for RL training. Combined with the action reward, these signals jointly improve action prediction and thinking quality. Extensive experiments on real-world Tencent advertising data spanning five heterogeneous content domains demonstrate the effectiveness, efficiency, fidelity, and diagnostic value of DASH.

cs.IR↗

arXiv2026

Breaking the Curse of Repulsion: Remoteness-Aware Control of Negative Off-Policy Updates

Off-policy policy optimization reuses historical behavior, including negative-advantage samples that suppress known failures. We show that repeated reuse can turn this useful signal into excessive repulsion: as the learner moves away from a historical negative action, subsequent updates make that action increasingly remote without necessarily reducing its update strength. Our aggregate theory characterizes the resulting transition from a stable displacement beyond the positive-only target to persistent drift and the loss of finite stable equilibria; controlled strength sweeps show that an intermediate displacement can improve held-out reward. The relevant learner-relative coordinate is squared standardized distance for Gaussian policies and surprisal for categorical policies. We introduce Dynamic Remoteness-Aware Policy Optimization (DRPO), which leaves the negative update unchanged in the near field and exponentially attenuates its remote tail. DRPO restores eventual inward Gaussian drift for every fixed finite negative-to-positive mass ratio, yields an explicit ultimate-bound radius, and changes categorical support suppression from exponential to polynomial probability decay. External diagnostics and controlled interventions isolate remoteness-dependent policy geometry as a source of negative-update amplification and show that selective tapering can remove its far-field effect without discarding useful local feedback.

cs.LG↗

arXiv2026

NOVA: A Verification-Aware Agent Harness for Architecture Evolution in Industrial Recommender Systems

Industrial advertising recommender systems are continually improved through architecture modifications, yet production iteration remains expert-intensive because coordinated changes to model topology, feature configuration, and interaction modules must satisfy strict interface, resource, and serving constraints. AutoML is limited to predefined search spaces, while generic coding agents verify runnability rather than recommender-specific semantic validity. Executable candidates may therefore violate architectural contracts, while the lack of structured reuse of semantic diagnostics and evaluation outcomes can lead to repeated invalid or ineffective modifications. We present NOVA, a verification-aware agent harness that organizes production architecture modification as multi-round search over concrete implementations within a fixed evaluation budget. At each round, NOVA generates multiple candidates under production constraints, rejects semantic violations, and ranks the valid survivors for local testing and offline evaluation. Across rounds, trajectory memory synthesizes semantic diagnostics, local-test outcomes, and offline metric changes into modification directions and forbidden patterns that guide subsequent search. Under the same maximum offline-evaluation budget for automated methods, NOVA achieves the highest effective pass rate, reaching 53.3% on ScaleUp and 51.7% on Literature-to-Production tasks. In a production A/B test covering 5% of traffic in an advertising system serving over one billion users, the selected Literature-to-Production candidate yields GMV gains of +1.25%, +1.70%, and +2.02% across three major pCVR objectives, with corresponding relative reductions in absolute pCVR bias of 58.8%, 66.7%, and 37.3%, respectively.

cs.IR↗

arXiv2026

Diffusion Language Model for Recommendation

Large language model (LLM)-empowered recommender systems have emerged as a promising paradigm for generative recommendation, leveraging their strong semantic reasoning and generative capacity to model complex, diverse user preferences. However, most existing approaches rely on an autoregressive paradigm that is suboptimal for recommendation. The next-token objective emphasizes sequential order rather than the structural inter-item dependencies underlying user preferences. In addition, prefix-constrained generation restricts bidirectional context and commits to left-to-right decoding, causing early errors to accumulate without correction. Inspired by the success of diffusion language models, we propose \textbf{DLMRec}, a discrete diffusion language model tailored for recommendation that offers a compelling alternative to autoregressive generation. Specifically, DLMRec introduces three key components to bridge diffusion language modeling with recommendation. First, a collaborative-aware stochastic tokenizer encodes multi-hop collaborative signals into expressive discrete tokens compatible with diffusion modeling. Second, a curriculum-driven training strategy aligns the denoising process with preference recovery through progressive item- and token-level learning. Third, a stability-aware voting mechanism aggregates iterative predictions to improve generation consistency and robustness.

cs.IR↗

arXiv2026

Investigation on the Relation between Active Regions' Compliance with Empirical Laws and Flare Productivity

It remains evasive whether solar active regions (ARs) obeying or violating Hale's polarity law, Joy's tilt law, and the hemispheric helicity rule (HHR) differ in flare productivity. Here we conduct a comprehensive statistical analysis of ARs during the Solar Cycle 24 and the ascending phase of Cycle 25. ARs are automatically detected from full-disk line-of-sight magnetograms acquired by the Michelson Doppler Imager (MDI) and the Helioseismic and Magnetic Imager (HMI). We calculate tilt angles via flux-weighted polarity centroids, estimate magnetic twist by the force-free parameter $α_{\mathrm{best}}$ from HMI vector magnetograms, and measure flare productivity using the flare index (FI) built from GOES C-class-and-above events. Our results substantiate that the majority of ARs follow the aforementioned three empirical laws. The compliance rate tends to be higher for ARs emerging at higher latitudes or having larger centroid distance, while total unsigned magnetic flux exerts limited influence, with a clear positive correlation only for Hale's law. Overall, FI shows no significant discrepancies across different compliance groups, except that Cycle 24 ARs that satisfy Hale's and Joy's laws but violate the HHR exhibit higher FI than other groups. We also identify empirical thresholds for centroid distance and total unsigned flux, above which the median FI of binned ARs becomes nonzero. Combining the flux and distance thresholds effectively separates flare-productive from flare-quiet ARs. We hence conclude that the flare productivity of ARs is not dependent on the compliance with the empirical laws, but more closely associated with sufficiently large and strong magnetic systems.

astro-ph.SR↗

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

TimeThink: Reasoning with Time for Video LLMs

Video reasoning requires models to identify and verify temporally localized evidence within long video sequences. Recent Video Large Language Models (Video-LLMs) have shown promising reasoning abilities when aligned with reinforcement learning, yet existing approaches typically rely on outcome-based rewards that supervise only the final prediction. Such supervision provides limited guidance on how models should discover the relevant temporal evidence during intermediate reasoning. In this work, we propose TimeThink, a reinforcement learning framework that explicitly guides temporal evidence discovery in Video-LLMs. Our key idea is to treat temporal clue steps as the fundamental optimization primitive of video reasoning, where each reasoning step references a candidate time interval in the video. We introduce a step-wise temporal process reward that provides localized credit assignment for these clues and a joint process--outcome optimization objective that balances reasoning fidelity with task correctness. To enable scalable training, we construct TimeThink-RFT-20K, a dataset with automatically derived temporal evidence segments. Extensive experiments across video reasoning, temporal grounding, and general video understanding benchmarks show that TimeThink consistently improves both temporal localization and reasoning performance, achieving state-of-the-art results among open-source video RL models.

cs.CV↗

arXiv2026

The recent anomalously weak polar field does not imply a weak field at solar cycle 25 minimum

The ongoing solar cycle 25 has progressed past its peak of sunspot numbers, being stronger than the previous cycle 24. However, the present polar field is rather weak compared to previous cycles at the same evolution phase, particularly in the northern hemisphere, where it has been decreasing since mid-2025 till present. A prominent poleward surge is observed to cause the decrease of the polar field. This raises concerns to the polar field at cycle 25 minimum, which is the precursor to the strength of the next cycle 26. To predict whether the polar field at cycle 25 minimum will be weak as expected, we use observation-based statistical properties to predict the active region emergence during the latter half of cycle 25, and use a surface flux transport model to simulate the evolution of the large-scale magnetic field. We predict the polar field at cycle 25 minimum to be $-5.62\pm1.61$G in the north and $5.51\pm1.48$G in the south, both stronger than those at cycle 24 minimum. This is because the poleward surge causing the temporal decrease of the polar field originates from a group of active regions that produces net increase to the polar field, instead of active regions with non-Joy's tilt. Our results suggest the weak polar field at present is unlikely to cause a weak minium and an exceptionally weak cycle 26, clarifying that the short term evolution should not be simply correlated to the long term properties of the solar large-scale field.

astro-ph.SR↗

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

AD-Bench: A Real-World, Trajectory-Aware Advertising Analytics Benchmark for LLM Agents

While Large Language Model (LLM) agents have made remarkable progress on complex reasoning, evaluating them in real-world environments remains an open problem. Existing benchmarks are largely confined to idealized simulations and fail to capture specialized domains such as advertising and marketing analytics, where tasks require multi-round interaction with professional tools and where ground-truth answers quickly become obsolete as data and platform rules evolve. To address this, we propose AD-Bench, a benchmark built from real user marketing-analysis requests on a production advertising platform. AD-Bench introduces two key designs: (i) a dynamic ground-truth pipeline that replays expert tool-call trajectories to regenerate answers consistent with the current environment, mitigating answer obsolescence; and (ii) a trajectory-aware evaluation that jointly measures end-to-end answer correctness (Pass@k) and trajectory coverage. Requests are stratified into three difficulty levels (L1-L3) to probe multi-round, multi-tool collaboration. Experiments show that the best model, Claude-Opus-4.7, attains Pass@1 = 76.9% and Pass@3 = 80.4% with 82.7% trajectory coverage overall, yet drops sharply on L3 to Pass@1 = 61.4% and Pass@3 = 65.1%, revealing that even state-of-the-art agents have substantial gaps in complex advertising analytics.

cs.CL↗

arXiv2026

Beyond Positive Signals: Unlocking Implicit Negative Behaviors for Enhanced Sequential User Modeling

User behavior sequence modeling has become a central component in modern click-through rate (CTR) prediction. Over the past years, the community has invested substantial effort into improving how sequences are encoded, from target-aware attention and interest evolution networks to unified architectures that jointly process sequential and non-sequential features. However, a more fundamental question remains under-explored: what should constitute the behavior sequence? Current practice constructs sequences exclusively from positive interactions (clicks, purchases, completions), while the far more abundant implicit negative behaviors (skips, low engagement, scroll-past) are largely underutilized. As gains from longer positive sequences approach diminishing returns, we revisit this underutilized data source within the sequential modeling framework. In this paper, we demonstrate that mixed-polarity behavior sequences, which chronologically interleave positive and negative tokens within a fixed length budget, consistently outperform positive-only sequences across diverse model architectures with negligible additional computational overhead. We further identify a semantic indistinguishability problem inherent to naive polarity embeddings and propose Target-Aware Polarity Fusion (TAPF), a lightweight target-conditioned gating mechanism that provides additional gains by differentiating behavioral evidence. Notably, even the simpler polarity bias baseline captures the majority of improvement, underscoring that the primary contribution is the mixed-polarity data paradigm itself. Experiments on three public benchmarks demonstrate consistent improvements of +1.9% to +9.6% relative AUC across five architectures, which validate the practical value of our approach.

cs.IR↗

arXiv2026

LENS: A Staged Design for Interaction Granularity in Sequential CTR Prediction

In sequential CTR prediction, a central design question is at what granularity the target should interact with the user behaviour sequence. Existing models mainly follow two routes. Raw-item architectures such as DIN let the target score each item in the sequence directly. This relies on well-trained item embeddings and becomes brittle for sparse items. Latent-query architectures such as HyFormer, MixFormer, and OneTrans build query representations by combining the target with other information. This is more robust across item-density regimes but blunter: target-specific control is diluted. We propose LENS to restore target-specific control within these coarser bottlenecks. LENS has two modules: a Target-Conditioned Query Gate (TCQG) for query activation and a Target-Conditioned Position Bias (TCPB) for history retrieval. We further introduce Query-Specific Position Bias (QueryPos), a simple static position-aware reference for latent-query backbones. Across three representative latent-query backbones and four datasets, the combined QueryPos+LENS design achieves positive total-gain point estimates in all twelve evaluated backbone--dataset cells. We also identify a density-dependent conditioning rule: as item density decreases, the optimal condition source shifts from item-only to item-plus-sequence.

cs.IR↗

arXiv2026

Harnessing Routing Foresight for Micro-step-level MoE load balancing in RL Post-training

Mixture-of-Experts (MoE) and reinforcement learning (RL) post-training now dominate large language model (LLM) development, yet expert load imbalance remains a critical challenge. Existing load-balancing systems target pre-training by relying on historical step-level statistics. However, these methods fail under the unique workload dynamics of RL post-training: the step-level load is stable, but the tiny batch sizes processed during micro-steps cause severe, high-frequency load fluctuations. We introduce ForeMoE, a micro-step-level load balancing system for MoE RL post-training. Instead of relying on historical statistics, ForeMoE exploits the multi-stage RL pipeline (rollout, recompute, policy update) by using foreseeable routing information from the rollout stage to proactively guide load balancing in the remaining stages. To support frequent per-micro-step reconfiguration, ForeMoE employs a hierarchical planner that decomposes the NP-hard load balancing problem into tractable sub-components, alongside a transfer engine that leverages complementary hardware paths (CPU-assisted and GPU-direct) for overlapped expert transfer. Evaluations on 64 GPUs demonstrate that ForeMoE achieves up to a 1.45$\times$ speedup over state-of-the-art RL post-training systems.

cs.DC↗

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

VeriSpace: Spatially Grounded Action Verification for Vision-Language-Action Models

Vision-language-action (VLA) models have shown strong promise for robotic manipulation, but their reliability at test time remains limited by one-shot action prediction, where even small action errors can cause grasp failure, collision, or incorrect task progression. A natural alternative is to equip VLA systems with test-time verification, allowing multiple candidate actions to be proposed and evaluated before execution. However, reliable action verification is challenging because it requires not only distinguishing subtle geometric differences between candidate actions, but also assessing whether an action makes meaningful progress toward the task goal. We present VeriSpace, a 3D-aware action verifier for test-time action selection in VLA systems. VeriSpace evaluates candidate actions through two key components: Dual-Path 3D-Injected Scene Encoding, which constructs a scene representation that jointly preserves visual semantics and explicit 3D geometry, and Spatially-Grounded Action Reasoning, which evaluates each action by reasoning over task-relevant spatial relations, geometric validity, and expected goal progress. Together, these components enable more reliable discrimination between subtle yet outcome-critical action candidates while remaining fully compatible with existing VLA policies. Experiments on public benchmarks and real-world robotic manipulation tasks show that VeriSpace consistently improves decision reliability over both underlying VLA policies and prior verification-based methods, yielding substantial gains in both in-distribution and out-of-distribution settings.

cs.RO↗

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↗