chiangel / awesome-llm-for-recsys Goto Github PK

Hi @CHIANGEL,

Thanks so much for sharing such a nice repo on LLMs for Rec. I am wondering if our recent paper Large Language Models as Zero-Shot Conversational Recommenders could be included in this list?

Our contributions include (1) datasets: new conversational recommendation datasets from Reddit; (2) LLMs evaluation on conversational recommendation tasks and related analysis.

Thanks in advance!

Hi Jianghao,

Thanks for your awesome repository and the excellent survey! I believe both of them will help many researchers gain insights into the latest LLM-related techniques and bring long-term benefits to the RS community.

I am happy to announce our paper, "MISSRec: Pre-training and Transferring Multi-modal Interest-aware Sequence Representation for Recommendation" (https://arxiv.org/abs/2308.11175), on ACM MM 2023. This paper is about integrating multi-modal information (using the pre-trained and frozen CLIP-ViT-B/32 model) into behavior sequence representation to support universal recommendation. The data and code are also available at https://github.com/gimpong/MM23-MISSRec.

If the paper falls within the scope of this repository, I would appreciate it if you considered including it in the paper list. :)

Kind regards,
Jinpeng

Dear Repo Owner,

Thank you for maintaining this nice repo of LLM for Recommendation. We have recently released a session-recommendation dataset named Amazon-M2 for evaluating LLMs in recommendation scenarios. It provides both a dataset, a benchmark, and three text-related tasks. Would you mind adding this work to your repo?

Thank you for your attention.

Dear authors, thanks for your great contribution to the recommender system community! Could you add two more datasets to make the GitHub page more comprehensive?

Ninerec, the RS-scenario is Feeds, and the link is https://github.com/westlake-repl/NineRec. It provides multiple datasets with raw text and images for transfer learning in recommender systems

MicroLens, the RS-scenario is Video Streaming, and the link is https://github.com/westlake-repl/MicroLens. It provides a large-scale short video dataset for recommendation tasks, providing various original modal content including text, images, audio, and videos.

Hi Jianghao, 👋

Thanks for this awesome repo and also the comprehensive survey! It helps me a lot when I am conducting research about LLMs for Recommender Systems! And also, it means a lot to the research community! 😄

I kindly hope that you would consider adding our new paper titled "Representation Learning with Large Language Models for Recommendation"(https://arxiv.org/abs/2310.15950) into your awesome repo. 😊

In this paper, we purpose a model-agnostic framework (RLMRec) which utilizes the LLMs to improve the performance of SOTA recommenders through representation learning. 🚀

paper link: https://arxiv.org/abs/2310.15950
code link: https://github.com/HKUDS/RLMRec

I would appreciate it if you consider attaching this paper the code link to your awesome reposity :)

Best regards,
Xubin

感觉一些论文分类并不太符合原Survey的要求？
比如UP5，POD，BDLM等等

Hi,

There is a new paper that discusses leveraging LLMs to obtain better explanations iteratively, and It then explores using enriched explanations to enhance Visualization Recommendations.

LLM4Vis: Explainable Visualization Recommendation using ChatGPT
Lei Wang, Songheng Zhang, Yun Wang, Ee-Peng Lim and Yong Wang
EMNLP Industry 2023 | paper | code