VarLenRec learns variable-length semantic IDs for generative recommendation by allocating longer codes to tail items via popularity-weighted information budget allocation, hyperbolic residual quantization, and a differentiable soft length controller.
hub Canonical reference
Session-based Recommendations with Recurrent Neural Networks
Canonical reference. 70% of citing Pith papers cite this work as background.
62
Pith papers citing it
Background
70% of classified citations
abstract
We apply recurrent neural networks (RNN) on a new domain, namely recommender systems. Real-life recommender systems often face the problem of having to base recommendations only on short session-based data (e.g. a small sportsware website) instead of long user histories (as in the case of Netflix). In this situation the frequently praised matrix factorization approaches are not accurate. This problem is usually overcome in practice by resorting to item-to-item recommendations, i.e. recommending similar items. We argue that by modeling the whole session, more accurate recommendations can be provided. We therefore propose an RNN-based approach for session-based recommendations. Our approach also considers practical aspects of the task and introduces several modifications to classic RNNs such as a ranking loss function that make it more viable for this specific problem. Experimental results on two data-sets show marked improvements over widely used approaches.
hub tools
citation-role summary
background 7
method 2
baseline 1
citation-polarity summary
claims ledger
- abstract We apply recurrent neural networks (RNN) on a new domain, namely recommender systems. Real-life recommender systems often face the problem of having to base recommendations only on short session-based data (e.g. a small sportsware website) instead of long user histories (as in the case of Netflix). In this situation the frequently praised matrix factorization approaches are not accurate. This problem is usually overcome in practice by resorting to item-to-item recommendations, i.e. recommending similar items. We argue that by modeling the whole session, more accurate recommendations can be pro
co-cited works
representative citing papers
AsymRec decouples input and output representations in generative recommendation via multi-expert semantic projection and multi-faceted hierarchical quantization, outperforming prior models by 15.8% on average.
F-GRPO factorizes group-relative policy optimization into generation and ranking phases within one autoregressive sequence, using order-invariant coverage and position-aware utility rewards to improve top-ranked performance on recommendation and multi-hop QA tasks.
AWARE augments generative next-POI recommendation with LLM agents that produce user-anchored narratives capturing events, culture, and trends, delivering up to 12.4% relative gains on three real datasets.
OSA improves LLM-based recommenders by anchoring ordinal preference levels as numeric tokens in the model's latent space to retain fine-grained strength information when fusing collaborative signals.
SUIN improves CTR prediction by augmenting target user sequences with similar users' behaviors via embedding-based retrieval, user-specific position encoding, and user-aware target attention.
Beam-search negatives induce partial AUC optimization in GRPO for LLM recommenders; Windowed Partial AUC and TAWin improve Top-K alignment on four datasets.
TF-LLMER resolves optimization barriers in LLM-enhanced recommenders through embedding normalization and Rec-PCA that aligns semantic representations with collaborative co-occurrence graphs.
AdaTTA is an actor-critic RL framework that selects sequence-specific test-time augmentations and improves recommendation metrics by up to 26% over fixed augmentation strategies on four datasets.
RAR retrieves candidate items from a 300k-movie corpus then uses LLM generation with RL feedback to produce context-aware recommendations that outperform baselines on benchmarks.
FAERec fuses collaborative ID embeddings with LLM semantic embeddings using adaptive gating and dual-level alignment to enhance tail-item sequential recommendations.
S²GR adds stepwise thinking tokens with contrastive supervision on codebook clusters to balance computational focus and ground reasoning paths in generative recommendation.
HSTU-based generative recommenders with 1.5 trillion parameters scale as a power law with compute up to GPT-3 scale, outperform baselines by up to 65.8% NDCG, run 5-15x faster than FlashAttention2 on long sequences, and improve online A/B metrics by 12.4%.
MDCNS is a multi-source negative sampling framework for sequential recommendation that uses peer and teacher models plus divergence and consensus mechanisms to improve diversity and avoid local optima.
SAPO computes per-reasoning-step group-relative advantages in RL to improve credit assignment for structured generation of semantic identifiers in recommendation systems.
RAGR builds mixed item-review sequences for generative recommendation and uses DPO alignment to favor item tokens, reporting gains over prior GR baselines on three datasets.
RRCM trains an LLM to dynamically retrieve from collaborative and meta memories using group relative policy optimization driven by final top-k recommendation quality.
A simple graph heuristic without training or sequence encoders matches or outperforms trained generative recommenders on 10 of 14 sequential recommendation benchmarks by exploiting local transition and feature shortcuts.
BLUE aligns LLM-generated textual user profiles with embedding-based recommendation objectives via reinforcement learning and next-item text supervision, yielding better zero-shot performance and cross-domain transfer than baselines.
PA-Bridge bridges passive conversation starter recommendations with active user expressions via adversarial distribution alignment and semantic discretization, yielding 0.54% higher feature penetration in online tests.
Fair re-ranking is equivalent to gradient descent on a ranking manifold under Walrasian equilibrium in an attention market, yielding the ManifoldRank algorithm that adjusts gradients for supply-side fairness costs and demand-side score predictions.
BITRec improves generative multi-behavior recommendation by modeling behavioral intensity via separated pathways and transitions via learnable relation matrices, reporting 15-23% gains on large retail datasets.
WPGRec is a new sequential recommender that performs multi-scale temporal modeling via stationary wavelet packets and injects high-order collaborative information through scale-aligned graph propagation with energy-aware gated fusion.
GraphRAG-IRL fuses graph-grounded MaxEnt IRL pre-ranking with persona-guided LLM re-ranking to deliver up to 16.8% NDCG@10 gains over IRL-only baselines on MovieLens and consistent 4-6% gains on KuaiRand.
citing papers explorer
-
Every Preference Has Its Strength: Injecting Ordinal Semantics into LLM-Based Recommenders
OSA improves LLM-based recommenders by anchoring ordinal preference levels as numeric tokens in the model's latent space to retain fine-grained strength information when fusing collaborative signals.
-
FLAME: Condensing Ensemble Diversity into a Single Network for Efficient Sequential Recommendation
FLAME condenses ensemble diversity into a single network via modular ensemble simulation and guided mutual learning during training, delivering ensemble-level performance with single-network inference speed on sequential recommendation tasks.