Latest Research Papers
2024-12-18
arXiv
SAFERec: Self-Attention and Frequency Enriched Model for Next Basket Recommendation
SAFERec, a new algorithm for Next-Basket Recommendation, enhances transformer-based models by incorporating item frequency information, improving their performance on NBR tasks. Experiments show SAFERec outperforms other baselines, with an 8% improvement in Recall@10.
Transformer-based approaches such as BERT4Rec and SASRec demonstrate strong
performance in Next Item Recommendation (NIR) tasks. However, applying these
architectures to Next-Basket Recommendation (NBR) tasks, which often involve
highly repetitive interactions, is challenging due to the vast number of
possible item combinations in a basket. Moreover, frequency-based methods such
as TIFU-KNN and UP-CF still demonstrate strong performance in NBR tasks,
frequently outperforming deep-learning approaches. This paper introduces
SAFERec, a novel algorithm for NBR that enhances transformer-based
architectures from NIR by incorporating item frequency information,
consequently improving their applicability to NBR tasks. Extensive experiments
on multiple datasets show that SAFERec outperforms all other baselines,
specifically achieving an 8\% improvement in Recall@10.
2024-10-25
arXiv
Knowledge Graph Enhanced Language Agents for Recommendation
This paper introduces Knowledge Graph Enhanced Language Agents (KGLA), a framework that integrates knowledge graphs with language agents to improve recommendation systems by enriching user profiles and capturing complex relationships between users and items. The method significantly enhances recommendation performance, as demonstrated by substantial improvements in NDCG@1 on three widely used benchmarks.
Language agents have recently been used to simulate human behavior and
user-item interactions for recommendation systems. However, current language
agent simulations do not understand the relationships between users and items,
leading to inaccurate user profiles and ineffective recommendations. In this
work, we explore the utility of Knowledge Graphs (KGs), which contain extensive
and reliable relationships between users and items, for recommendation. Our key
insight is that the paths in a KG can capture complex relationships between
users and items, eliciting the underlying reasons for user preferences and
enriching user profiles. Leveraging this insight, we propose Knowledge Graph
Enhanced Language Agents(KGLA), a framework that unifies language agents and KG
for recommendation systems. In the simulated recommendation scenario, we
position the user and item within the KG and integrate KG paths as natural
language descriptions into the simulation. This allows language agents to
interact with each other and discover sufficient rationale behind their
interactions, making the simulation more accurate and aligned with real-world
cases, thus improving recommendation performance. Our experimental results show
that KGLA significantly improves recommendation performance (with a 33%-95%
boost in NDCG@1 among three widely used benchmarks) compared to the previous
best baseline method.
2024-10-21
arXiv
STAR: A Simple Training-free Approach for Recommendations using Large Language Models
This paper introduces STAR, a training-free approach for recommendation systems using large language models (LLMs) that combines semantic embeddings and collaborative user information. The method achieves competitive performance on next-item prediction tasks, demonstrating the potential of LLMs without fine-tuning. Experimental results show significant improvements in Hits@10 on various categories of the Amazon Review dataset.
Recent progress in large language models (LLMs) offers promising new
approaches for recommendation system (RecSys) tasks. While the current
state-of-the-art methods rely on fine-tuning LLMs to achieve optimal results,
this process is costly and introduces significant engineering complexities.
Conversely, methods that bypass fine-tuning and use LLMs directly are less
resource-intensive but often fail to fully capture both semantic and
collaborative information, resulting in sub-optimal performance compared to
their fine-tuned counterparts. In this paper, we propose a Simple Training-free
Approach for Recommendation (STAR), a framework that utilizes LLMs and can be
applied to various recommendation tasks without the need for fine-tuning. Our
approach involves a retrieval stage that uses semantic embeddings from LLMs
combined with collaborative user information to retrieve candidate items. We
then apply an LLM for pairwise ranking to enhance next-item prediction.
Experimental results on the Amazon Review dataset show competitive performance
for next item prediction, even with our retrieval stage alone. Our full method
achieves Hits@10 performance of +23.8% on Beauty, +37.5% on Toys and Games, and
-1.8% on Sports and Outdoors relative to the best supervised models. This
framework offers an effective alternative to traditional supervised models,
highlighting the potential of LLMs in recommendation systems without extensive
training or custom architectures.
2024-03-04
arXiv
Wukong: Towards a Scaling Law for Large-Scale Recommendation
This paper introduces Wukong, a network architecture based on stacked factorization machines and an upscaling strategy, to establish a scaling law for recommendation models. Wukong effectively captures diverse interactions and outperforms state-of-the-art models in quality and scalability. The results show that Wukong maintains its superiority across a wide range of model complexities.
Scaling laws play an instrumental role in the sustainable improvement in
model quality. Unfortunately, recommendation models to date do not exhibit such
laws similar to those observed in the domain of large language models, due to
the inefficiencies of their upscaling mechanisms. This limitation poses
significant challenges in adapting these models to increasingly more complex
real-world datasets. In this paper, we propose an effective network
architecture based purely on stacked factorization machines, and a synergistic
upscaling strategy, collectively dubbed Wukong, to establish a scaling law in
the domain of recommendation. Wukong's unique design makes it possible to
capture diverse, any-order of interactions simply through taller and wider
layers. We conducted extensive evaluations on six public datasets, and our
results demonstrate that Wukong consistently outperforms state-of-the-art
models quality-wise. Further, we assessed Wukong's scalability on an internal,
large-scale dataset. The results show that Wukong retains its superiority in
quality over state-of-the-art models, while holding the scaling law across two
orders of magnitude in model complexity, extending beyond 100 GFLOP/example,
where prior arts fall short.