研究人员正在开发新的方法，通过利用大型语言模型（LLM）和优化计算效率来改进推荐系统。WhisperRec专注于潜在推理以减少推理开销，而ROCS和CCFormer分别提出了面向请求的计算共享和高效的跨字段交互。HiLaR和Feedback-Grounded Policy Discovery等其他方法分别旨在通过分层潜在推理和基于反馈的策略发现来增强基于LLM的推荐。像Google Discover这样的系统中的异构排名也正在通过自适应架构来解决。
AI
arXiv:2607.26621v2 Announce Type: cross Abstract: Large language models (LLMs) have demonstrated strong reasoning capabilities, motivating their adoption as backbones for foundation recommendation models (FRMs). Existing approaches typically enhance recommendation with explicit C…
arXiv:2607.27577v1 Announce Type: cross Abstract: Heterogeneous recommendation feeds present complex challenges that extend beyond those found in highly homogeneous environments (e.g., music-only or video-only closed-ecosystem platforms). In Google Discover, a unified feed integr…
arXiv cs.LG
TIER_1English(EN)·Yuxin Chen, Liang Luo, Buyun Zhang, Jian Jiao, Boda Li, Haoyu Wang, Tongyi Tang, Ao Cai, Zijian Shen, Zhengkai Zhang, Wenyi Xie, Ryan Dick, Han Liu, Neng Shi, Bin Yu, Jianbo Xiao, Shuyao Bi, Hongtao Yu, Yuanwei Fang, Zhuoran Zhao, Sijia Chen, Yang Chen, …·
arXiv:2607.27744v1 Announce Type: new Abstract: Modern recommendation models gain prediction quality by scaling feature-interaction and sequence modules, but production cost constraints cap how far systems can scale. In this work, we propose Request-Oriented Compute Sharing (ROCS…
Recent studies in industrial recommendation systems have demonstrated that sequential recommendation models built upon self-attention can benefit from predictable scaling laws by increasing sequence length and model capacity. However, practical recommender systems impose strict l…
Semantic-ID-based generative recommenders enable efficient next-item generation, but their item-level supervision mainly captures behavioral co-occurrence and local transitions. Large language models (LLMs) can complement these models by reasoning over heterogeneous interaction h…
Large Language Models (LLMs) have shown strong potential for recommendation by leveraging their semantic understanding and contextual modeling capabilities. Recent studies further introduce reasoning mechanisms to improve user preference modeling. However, explicit natural-langua…
Modern recommendation models gain prediction quality by scaling feature-interaction and sequence modules, but production cost constraints cap how far systems can scale. In this work, we propose Request-Oriented Compute Sharing (ROCS), a modeling and inference paradigm that exploi…
Making LLM-based generative recommendation models stronger and more personalized through natural language and explicit reasoning is a widely anticipated yet still unsolved goal. Such models cast recommendation as autoregressively generating an item's semantic-ID (SID), a short tu…
Generative recommendation formulates next-item prediction as conditional autoregressive generation over discrete Semantic IDs, enabling end-to-end recommendation over large-scale item spaces. However, most existing methods follow a history-as-context paradigm that repeatedly reco…
Heterogeneous recommendation feeds present complex challenges that extend beyond those found in highly homogeneous environments (e.g., music-only or video-only closed-ecosystem platforms). In Google Discover, a unified feed integrates diverse content sourced from the decentralize…
Sequential recommenders are commonly trained with one-hot next-item labels under a causal (prefix-only) objective aligned with inference. While deployment-compatible, this supervision offers little insight into relative preferences among non-target items. Yet logged interaction s…
Large language models (LLMs) have demonstrated strong reasoning capabilities, motivating their adoption as backbones for foundation recommendation models (FRMs). Existing approaches typically enhance recommendation with explicit Chain-of-Thought (CoT) under the Think-then-Answer …
Large language models (LLMs) have demonstrated strong reasoning capabilities, motivating their adoption as backbones for foundation recommendation models (FRMs). Existing approaches typically enhance recommendation with explicit Chain-of-Thought (CoT) under the Think-then-Answer …
Industrial recommender systems build candidate pools by assigning explicit quotas to objective-specific retrieval routes. This design offers quota control but increasingly fragments modeling, training, and serving as the route set grows. Semantic-ID-based generative retrieval pro…
arXiv cs.AI
TIER_1English(EN)·Junting Wang, Xinrui He, Yunzhe Li, Hari Sundaram·
arXiv:2607.24995v1 Announce Type: new Abstract: Semantic IDs (SIDs) are now a central component of generative recommendation. Current SID-based systems assign three roles to the same token sequence. Shared prefixes are intended to organize related items, the complete SID identifi…
arXiv cs.AI
TIER_1English(EN)·Ziyu Zheng, Zhengshun Du, Yaming Yang, Bin Tong, Guan Wang, Meng Yan, Ziyu Guan, Wei Zhao·
arXiv:2607.25216v1 Announce Type: cross Abstract: Semantic ID-based generative recommendation tokenizes each item into a sequence of discrete semantic IDs and predicts the next item by generating semantic IDs. However, existing methods typically regard SIDs as independent discret…
arXiv:2607.25209v1 Announce Type: cross Abstract: Generative recommendation commonly represents items using fixed-length semantic identifiers (SIDs) constructed through clustering and quantization. However, these artificial codes may overcompress item semantics, remain misaligned…
arXiv:2607.24865v1 Announce Type: cross Abstract: Large-scale recommendation systems face "Memory Wall" bottlenecks due to massive, dense embedding tables. While generative retrieval uses discrete tokens for IDs, high-dimensional context still relies on inefficient dense formats.…
arXiv:2607.24845v1 Announce Type: cross Abstract: Large language models (LLMs) have been applied to sequential recommendation by formulating it as a natural language task. Previous work has improved personalization by incorporating collaborative and sequential signals through inp…
Reranking is a combinatorial decision problem that aims to select and order a high-utility slate from a request-specific candidate set. A major line of generative rerankers adopts autoregressive (AR) models, which construct the slate one position at a time to capture inter-positi…
Multimodal large language models (MLLMs) can convert multimodal item content into structured descriptions used as semantic features for recommendation. Conventional content-only generation, however, cannot use downstream user signals to determine which semantics should be emphasi…
Modern recommendation interfaces organise content into shelves: themed rows such as "More of What You Like" or "New Releases for You." In production systems, these shelves are typically defined through hand-crafted templates coupled with dedicated retrieval logic. While effective…
Generative recommendation formulates recommendation task into an SID sequence autoregressive generation paradigm, but the decoding process is often dominated by generation likelihood. This may conflict with real-world business objectives, where high-value candidates can receive l…
Generative recommendation tokenizes items as discrete Semantic IDs (SIDs) and autoregressively generates target items from users' historical SID sequences. Although existing SIDs incorporate multimodal and structured information, they are typically statically assigned and indepen…
Generative recommendation has recently emerged as a promising paradigm that reformulates retrieval as autoregressive generation over semantic identifiers (SIDs), achieving strong performance and drawing increasing attention as an alternative to matching. Despite this progress, SI…
Semantic ID-based generative recommendation tokenizes each item into a sequence of discrete semantic IDs and predicts the next item by generating semantic IDs. However, existing methods typically regard SIDs as independent discrete symbols, while often overlooking the topology of…
Generative recommendation commonly represents items using fixed-length semantic identifiers (SIDs) constructed through clustering and quantization. However, these artificial codes may overcompress item semantics, remain misaligned with pretrained LLM vocabularies, and require cos…
Early ranking stages in recommendation systems precompute item embeddings and cache them in-model for scoring within strict latency constraints. Because this cache exists only at serving time, outside the training loop, training and serving use different item representations, a s…
Large Language Models (LLMs) have shown great promise in recommendation due to superior reasoning abilities. However, existing methods mainly rely on explicit Chain-of-Thought (CoT), resulting in verbose reasoning texts and inefficient response times. latent reasoning aims to bal…
Semantic-ID-based generative recommendation represents each item as a hierarchical discrete token sequence and reformulates next-item prediction as constrained sequence generation. Existing methods, however, mainly use Semantic IDs as target sequences to be memorized, leaving the…
Generative recommendation unifies retrieval and ranking within a single model by autoregressively decoding semantic identifier (SID) sequences. Yet reliably incorporating behavior signals from clicks, cart additions, and orders remains challenging. Existing approaches either join…
Large-scale recommendation systems face "Memory Wall" bottlenecks due to massive, dense embedding tables. While generative retrieval uses discrete tokens for IDs, high-dimensional context still relies on inefficient dense formats. Inspired by computer vision data compression, we …
Large language models (LLMs) have been applied to sequential recommendation by formulating it as a natural language task. Previous work has improved personalization by incorporating collaborative and sequential signals through input conditioning or LLM fine-tuning. However, exist…
<p><em>A daily deep dive into llm topics, coding problems, and platform features from <a href="https://pixelbank.dev" rel="noopener noreferrer">PixelBank</a>.</em></p> <h2> Topic Deep Dive: Inference Optimization </h2> <p><em>From the Deployment & Optimization chapter</em></p…
