OneBar: An End-to-End Content-Grounded Generative Query Recommendation Framework for E-Commerce Video Feeds

Short-video platforms now expose clickable search entries beneath the video player, enabling users to easily express content-induced search intent. However, conventional query recommendation systems on short-video platforms suffer from latency constraints and objective misalignment, while recent generative approaches struggle with noisy content-side metadata and preference drift. To address these issues, we propose OneBar, an end-to-end generative framework for real-time query recommendation for E-Commerce video feeds. OneBar features three key innovations: (1) a collaborative-multimodal intent grounding module that fuses multimodal video understanding and behavior-derived collaborative anchors; (2) a Unified End-to-End architecture equipped with a prompt-compression mechanism for efficient online serving; and (3) a progressive preference learning strategy for efficient preference-internalization, which internalizes hierarchical behavior preferences into the generative policy, eliminating the need for a separately trained reward model. Compared with online base, OneBar increases Query Exposure by 16.91\% and Query Click by 18.68\%, while maintaining a slight Query CTR gain of 0.19\%. The additional search traffic further contributes to 20.36\% more guided orders and 21.67\% higher GMV.

OneRetrieval: Unifying Multi-Branch E-commerce Retrieval with an Editable Generative Model

Industrial e-commerce search serves hundreds of millions of items through a multi-branch retrieval stage fused by hand-tuned merging without joint optimization. Generative retrieval (GR) raises the prospect of collapsing this stage into a single model, yet unification is gated by more than retrieval quality: the inverted-index branch converts below the platform average yet persists because it is almost the only branch where operations can inject a new term within hours without any model update; a one-model substitute must preserve this real-time editability. Existing GR methods structurally lack it: closed-codebook methods fix each slot to a quantized embedding at training, while open-vocabulary methods leave new-term routing to model generalization. We present OneRetrieval, a one-model GR framework built on Keyword-Aligned Encoding (KAE), which ties each identifier position to an interpretable attribute word, pairing competitive recall quality with the editability of the inverted index -- to our knowledge the first editable generative retrieval method. An information-theoretic merging organizes 18 attribute categories into six codebook groups with non-uniform capacity; reserved slots in each codebook can be bound to new words after deployment without retraining; and a four-stage fine-tuning pipeline secures quality and editability jointly. On five million real-traffic requests, OneRetrieval matches the deep recall of the strongest generative baseline, with an intervention hit rate over an order of magnitude above closed-codebook encodings. Online, replacing the inverted-index branch significantly lifts order volume; extending to nearly the entire stage holds conversion while improving CTR. The system is deployed at Kuaishou, serving hundreds of millions of PVs daily.

Plan Before Search: Search Agents Need Plan

Training large language models as retrieval-augmented reasoning agents typically combines reinforcement learning with an SFT cold start distilled from a stronger model. However, this paradigm overlooks two fundamental factors: the dependency structure among sub-skills, and the possibility that distillation is not the only route to capability acquisition. We study this through Plan, a structured agentic behavior for multi-hop retrieval that decomposes a question into ordered sub-questions before any retrieval is performed, so that each search step can be anchored to a pre-designed sub-question instead of drifting under the influence of partially relevant documents retrieved earlier. However, across three model families spanning 3B to 14B parameters, we find that an identical reward signal induces qualitatively different RL failure modes. This phenomenon indicates that successful training hinges not only on reward design but also on model-specific feasibility conditions: sufficient initial entropy, training stability, and prerequisite sub-skills. Motivated by this, we propose a self-bootstrapping paradigm in which a small-scale seed model generates filtered trajectories that activate Plan in any target model, eliminating the need for distillation from an external stronger model. Our pipeline activates Plan across every tested model and consistently outperforms competitive baselines on multi-hop QA benchmarks.

TIGER-FG: Text-Guided Implicit Fine-Grained Grounding for E-commerce Retrieval

E-commerce image search often takes a cropped image as the query, while each candidate is represented by full item images and structured text. This image-to-multimodal retrieval setting presents two asymmetries: a modality disparity -- a visual query must match image--text items, and a granularity disparity -- a cropped query must be compared with full images containing background context and possible distractors. Detection-based pipelines handle the granularity disparity through explicit localization but incur extra cost and error propagation, whereas CLIP-style encoders avoid detection, but are vulnerable to backgrounds or irrelevant items. To address these limitations, we propose TIGER-FG, a text-guided implicit fine-grained grounding framework for image-to-multimodal e-commerce retrieval. TIGER-FG uses item text as semantic guidance to produce target-focused item representations without object detection for retrieval. We further introduce dual distillation objectives that preserve target-region spatial consistency and query--item similarity structure, yielding more stable and discriminative multimodal representations. In addition, we construct ECom-RF-IMMR, a realistic benchmark suite with a 10M-pair training set and two evaluation benchmarks covering standard and cluttered item layouts. TIGER-FG improves Recall@1 over the strongest baseline by 6.1 and 34.4 percentage points on the two evaluation benchmarks, respectively, with only 85.7M query-side parameters and 256-dim embeddings. Results on public e-commerce benchmarks further demonstrate its generalization to noisy and one-to-many retrieval scenarios. Code and data will be released.

SD-Search: On-Policy Hindsight Self-Distillation for Search-Augmented Reasoning

Search-augmented reasoning agents interleave internal reasoning with calls to an external retriever, and their performance relies on the quality of each issued query. However, under outcome-reward reinforcement learning, every search decision in a rollout shares the same trajectory-level reward, leaving individual queries without step-specific credit. Recent process-supervision approaches address this gap by drawing step-level signals from outside the policy, relying either on a much larger teacher model, or on sub-question annotations produced by a stronger external system. In contrast, we propose SD-Search, which derives step-level supervision from the policy itself through on-policy hindsight self-distillation, requiring neither an external teacher nor additional annotations. In SD-Search, a single model plays two roles that differ only in conditioning: a student that sees only the context available at inference time, and a teacher that additionally conditions on a compact hindsight block summarizing the search queries and final outcomes of a group of rollouts sampled from the same question. Since the teacher knows how each rollout unfolded and which ones succeeded, its query distribution implicitly marks which decisions were worth making, and the student is trained to recover this behavior by minimizing the token-level Jensen--Shannon divergence to the teacher at search-query positions. This layers a dense, step-level signal on top of GRPO's coarse trajectory reward. Crucially, this signal is produced by the policy itself within the standard RL training loop, without external model inference, auxiliary annotation pipeline, or additional training stage.

Beyond Static Collision Handling: Adaptive Semantic ID Learning for Multimodal Recommendation at Industrial Scale

Modern recommendation systems involve massive catalogs of multimodal items, where scalable item identification must balance compactness, semantic fidelity, and downstream effectiveness. Semantic IDs (SIDs) address this need by representing items as short discrete token sequences derived from multimodal signals, providing a compact interface for retrieval, ranking, and generative recommendation. However, effective SID learning is hindered by collisions, where different items are assigned identical or highly confusable codes. Existing methods mainly rely on improved quantization or fixed overlap regularization, but they do not adaptively distinguish whether an overlap should be suppressed or preserved. We propose AdaSID, an adaptive semantic ID learning framework for recommendation. AdaSID regulates SID overlaps through a two-stage process. First, it relaxes repulsion for observed overlaps when the involved items are semantically compatible, preserving admissible sharing rather than uniformly separating all collisions. Second, it allocates the remaining regulation pressure according to local collision load and training progress, strengthening control in congested regions while gradually rebalancing optimization toward recommendation alignment. This design adaptively decides which overlaps to penalize, how strongly to regulate them, and when to shift the learning focus. Extensive offline and online experiments validate AdaSID. On two public benchmarks, AdaSID improves Recall and NDCG by about 4.5% on average over strong baselines, while improving codebook utilization and SID diversity. In Kuaishou e-commerce, an online A/B test on short-video retrieval covering tens of millions of users achieves statistically significant gains, including a 0.98% GMV improvement, and industrial ranking evaluation shows consistent AUC improvements.

IG-Search: Step-Level Information Gain Rewards for Search-Augmented Reasoning

Reinforcement learning has emerged as an effective paradigm for training large language models to perform search-augmented reasoning. However, existing approaches rely on trajectory-level rewards that cannot distinguish precise search queries from vague or redundant ones within a rollout group, and collapse to a near-zero gradient signal whenever every sampled trajectory fails. In this paper, we propose IG-Search, a reinforcement learning framework that introduces a step-level reward based on Information Gain (IG). For each search step, IG measures how much the retrieved documents improve the model's confidence in the gold answer relative to a counterfactual baseline of random documents, thereby reflecting the effectiveness of the underlying search query. This signal is fed back to the corresponding search-query tokens via per-token advantage modulation in GRPO, enabling fine-grained, step-level credit assignment within a rollout. Unlike prior step-level methods that require either externally annotated intermediate supervision or shared environment states across trajectories, IG-Search derives its signals from the policy's own generation probabilities, requiring no intermediate annotations beyond standard question-answer pairs. Experiments on seven single-hop and multi-hop QA benchmarks demonstrate that IG-Search achieves an average EM of 0.430 with Qwen2.5-3B, outperforming the strongest trajectory-level baseline (MR-Search) by 1.6 points and the step-level method GiGPO by 0.9 points on average across benchmarks, with particularly pronounced gains on multi-hop reasoning tasks. Despite introducing a dense step-level signal, IG-Search adds only ~6.4% to per-step training wall-clock time over the trajectory-level baseline and leaves inference latency unchanged, while still providing a meaningful gradient signal even when every sampled trajectory answers incorrectly.

Scale over Preference: The Impact of AI-Generated Content on Online Content Ecology

The rapid proliferation of Artificial Intelligence-Generated Content (AIGC) is fundamentally restructuring online content ecologies, necessitating a rigorous examination of its behavioral and distributional implications. Leveraging a comprehensive longitudinal dataset comprising tens of millions of users from a leading Chinese video-sharing platform, this study elucidated the distinct creation and consumption behaviors characterizing AIGC versus Human-Generated Content (HGC). We identified a prevalent scale-over-preference dynamic, wherein AIGC creators achieve aggregate engagement comparable to HGC creators through high-volume production, despite a marked consumer preference for HGC. Deeper analysis uncovered the ability of the algorithmic content distribution mechanism in moderating these competing interests regarding AIGC. These findings advocated for the implementation of AIGC-sensitive distribution algorithms and precise governance frameworks to ensure the long-term health of the online content platforms.

OneSearch-V2: The Latent Reasoning Enhanced Self-distillation Generative Search Framework

Generative Retrieval (GR) has emerged as a promising paradigm for modern search systems. Compared to multi-stage cascaded architecture, it offers advantages such as end-to-end joint optimization and high computational efficiency. OneSearch, as a representative industrial-scale deployed generative search framework, has brought significant commercial and operational benefits. However, its inadequate understanding of complex queries, inefficient exploitation of latent user intents, and overfitting to narrow historical preferences have limited its further performance improvement. To address these challenges, we propose \textbf{OneSearch-V2}, a latent reasoning enhanced self-distillation generative search framework. It contains three key innovations: (1) a thought-augmented complex query understanding module, which enables deep query understanding and overcomes the shallow semantic matching limitations of direct inference; (2) a reasoning-internalized self-distillation training pipeline, which uncovers users' potential yet precise e-commerce intentions beyond log-fitting through implicit in-context learning; (3) a behavior preference alignment optimization system, which mitigates reward hacking arising from the single conversion metric, and addresses personal preference via direct user feedback. Extensive offline evaluations demonstrate OneSearch-V2's strong query recognition and user profiling capabilities. Online A/B tests further validate its business effectiveness, yielding +3.98\% item CTR, +3.05\% buyer conversion rate, and +2.11\% order volume. Manual evaluation further confirms gains in search experience quality, with +1.65\% in page good rate and +1.37\% in query-item relevance. More importantly, OneSearch-V2 effectively mitigates common search system issues such as information bubbles and long-tail sparsity, without incurring additional inference costs or serving latency.

Stop Treating Collisions Equally: Qualification-Aware Semantic ID Learning for Recommendation at Industrial Scale

Semantic IDs (SIDs) are compact discrete representations derived from multimodal item features, serving as a unified abstraction for ID-based and generative recommendation. However, learning high-quality SIDs remains challenging due to two issues. (1) Collision problem: the quantized token space is prone to collisions, in which semantically distinct items are assigned identical or overly similar SID compositions, resulting in semantic entanglement. (2) Collision-signal heterogeneity: collisions are not uniformly harmful. Some reflect genuine conflicts between semantically unrelated items, while others stem from benign redundancy or systematic data effects. To address these challenges, we propose Qualification-Aware Semantic ID Learning (QuaSID), an end-to-end framework that learns collision-qualified SIDs by selectively repelling qualified conflict pairs and scaling the repulsion strength by collision severity. QuaSID consists of two mechanisms: Hamming-guided Margin Repulsion, which translates low-Hamming SID overlaps into explicit, severity-scaled geometric constraints on the encoder space; and Conflict-Aware Valid Pair Masking, which masks protocol-induced benign overlaps to denoise repulsion supervision. In addition, QuaSID incorporates a dual-tower contrastive objective to inject collaborative signals into tokenization. Experiments on public benchmarks and industrial data validate QuaSID. On public datasets, QuaSID consistently outperforms strong baselines, improving top-K ranking quality by 5.9% over the best baseline while increasing SID composition diversity. In an online A/B test on Kuaishou e-commerce with a 5% traffic split, QuaSID increases ranking GMV-S2 by 2.38% and improves completed orders on cold-start retrieval by up to 6.42%. Finally, we show that the proposed repulsion loss is plug-and-play and enhances a range of SID learning frameworks across datasets.