TRACER uses token reassignment for concept-related items plus a coherence regularizer to unlearn specific concepts in generative recommendation while preserving utility better than baselines.
Minionerec: An open-source framework for scaling generative recommendation
11 Pith papers cite this work. Polarity classification is still indexing.
11
Pith papers citing it
citation-role summary
background 2
citation-polarity summary
years
2026 11roles
background 2polarities
background 2representative citing papers
BLADE uses Bayesian list-wise alignment with dynamic estimation to create a self-evolving target that overcomes limitations of static references in LLM-based recommendation, yielding sustained gains in ranking and complex metrics.
Exact LTI Koopman models for nonlinear control systems require affine linear dynamics under controllability and coordinate inclusion assumptions.
Beam-search negatives induce partial AUC optimization in GRPO for LLM recommenders; Windowed Partial AUC and TAWin improve Top-K alignment on four datasets.
Releases TencentGR-1M and TencentGR-10M datasets with baselines for all-modality generative recommendation in advertising, including weighted evaluation for conversions.
Closed-loop LLM simulations find generative recommenders form fewer exposure-level information cocoons than traditional sequential baselines on Amazon data, though tokenization strategy and model scale affect concentration in generated SID space.
LLMs for generative recommendation show heavy one-hop memorization that accounts for most gains over baselines, and IIRG training that incorporates multi-hop co-occurrences and semantic relations yields larger gains on non-memorizable cases.
GloRank reformulates list-wise reranking as token generation over a global item identifier space, using supervised pre-training followed by reinforcement learning to maximize list-wise utility and outperforming baselines on benchmarks and industrial data.
CRAB mitigates popularity bias in generative recommenders by rebalancing the semantic token codebook through splitting popular tokens and applying a tree-structured regularizer to boost representations for unpopular items.
Adding visual evidence and LLM-mined item-side intent descriptors before semantic-ID quantization, plus a relevance-gated quality reward, improves SID-based generative recommendation on three Amazon categories.
TriAlignGR introduces cross-modal alignment, deep interest mining via CoT, and triangular multitask training to fix semantic degradation and opacity in SID-based generative recommendation.
citing papers explorer
-
TRACER: Token ReAssignment for Concept ERasure in Generative Recommendation
TRACER uses token reassignment for concept-related items plus a coherence regularizer to unlearn specific concepts in generative recommendation while preserving utility better than baselines.
-
Beyond Static Best-of-N: Bayesian List-wise Alignment for LLM-based Recommendation
BLADE uses Bayesian list-wise alignment with dynamic estimation to create a self-evolving target that overcomes limitations of static references in LLM-based recommendation, yielding sustained gains in ranking and complex metrics.
-
Limitations of LTI Koopman Modeling for Nonlinear Control Systems
Exact LTI Koopman models for nonlinear control systems require affine linear dynamics under controllability and coordinate inclusion assumptions.
-
Objective Shaping with Hard Negatives: Windowed Partial AUC Optimization for RL-based LLM Recommenders
Beam-search negatives induce partial AUC optimization in GRPO for LLM recommenders; Windowed Partial AUC and TAWin improve Top-K alignment on four datasets.
-
Tencent Advertising Algorithm Challenge 2025: All-Modality Generative Recommendation
Releases TencentGR-1M and TencentGR-10M datasets with baselines for all-modality generative recommendation in advertising, including weighted evaluation for conversions.
-
Do Generative Recommenders Deepen the Information Cocoon? A Closed-Loop Simulation with LLM-powered User Simulators
Closed-loop LLM simulations find generative recommenders form fewer exposure-level information cocoons than traditional sequential baselines on Amazon data, though tokenization strategy and model scale affect concentration in generated SID space.
-
On the Memorization Behavior of LLMs in Generative Recommendation: Observations, Implications, and Training Strategies
LLMs for generative recommendation show heavy one-hop memorization that accounts for most gains over baselines, and IIRG training that incorporates multi-hop co-occurrences and semantic relations yields larger gains on non-memorizable cases.
-
From Local Indices to Global Identifiers: Generative Reranking for Recommender Systems via Global Action Space
GloRank reformulates list-wise reranking as token generation over a global item identifier space, using supervised pre-training followed by reinforcement learning to maximize list-wise utility and outperforming baselines on benchmarks and industrial data.
-
CRAB: Codebook Rebalancing for Bias Mitigation in Generative Recommendation
CRAB mitigates popularity bias in generative recommenders by rebalancing the semantic token codebook through splitting popular tokens and applying a tree-structured regularizer to boost representations for unpopular items.
-
Deep Interest Mining for Intent-Enriched Semantic IDs in Multimodal Generative Recommendation
Adding visual evidence and LLM-mined item-side intent descriptors before semantic-ID quantization, plus a relevance-gated quality reward, improves SID-based generative recommendation on three Amazon categories.
-
TriAlignGR: Triangular Multitask Alignment with Multimodal Deep Interest Mining for Generative Recommendation
TriAlignGR introduces cross-modal alignment, deep interest mining via CoT, and triangular multitask training to fix semantic degradation and opacity in SID-based generative recommendation.