Search

In recent years, large language models (LLM) have emerged as powerful tools for diverse natural language processing tasks. However, their potential for recommender systems under the generative recommendation paradigm remains relatively unexplored. This paper presents an innovative approach to recommendation systems using large language models (LLMs) based on text data. In this paper, we present a novel LLM for generative recommendation (GenRec) that utilized the expressive power of LLM to directly generate the target item to recommend, rather than calculating ranking score for each candidate item one by one as in traditional discriminative recommendation. GenRec uses LLM's understanding ability to interpret context, learn user preferences, and generate relevant recommendation. Our proposed approach leverages the vast knowledge encoded in large language models to accomplish recommendation tasks. We first we formulate specialized prompts to enhance the ability of LLM to comprehend recommendation tasks. Subsequently, we use these prompts to fine-tune the LLaMA backbone LLM on a dataset of user-item interactions, represented by textual data, to capture user preferences and item characteristics. Our research underscores the potential of LLM-based generative recommendation in revolutionizing the domain of recommendation systems and offers a foundational framework for future explorations in this field. We conduct extensive experiments on benchmark datasets, and the experiments shows that our GenRec has significant better results on large dataset.

Human intelligence excels at combining basic skills to solve complex tasks. This capability is vital for Artificial Intelligence (AI) and should be embedded in comprehensive intelligent models, enabling them to harness expert models for complex task-solving towards Artificial General Intelligence (AGI). Large Language Models (LLMs) show promising learning and reasoning abilities, and can effectively use external models to tackle complex problems. In this work, we introduce OpenAGI, an open-source AGI research platform designed for multi-step, real-world tasks. Specifically, OpenAGI uses a dual strategy, integrating standard benchmark tasks for benchmarking and evaluation, and open-ended tasks including more expandable models for creative problem-solving. Tasks are presented as natural language queries to the LLM, which then selects and executes appropriate models. We also propose a Reinforcement Learning from Task Feedback (RLTF) mechanism that uses task results to improve the LLM's ability, which creates a self-improving AI feedback loop. While we acknowledge that AGI is a broad and multifaceted research challenge with no singularly defined solution path, the integration of LLMs with domain-specific expert models, inspired by mirroring the blend of general and specialized intelligence in humans, offers a promising approach towards AGI. We are open-sourcing the OpenAGI project's code, dataset, benchmarks, evaluation methods, and demo to foster community involvement in AGI advancement: https://github.com/agiresearch/OpenAGI., 22 pages, 11 figures, 7 tables

The spread of a graph $G$ is the difference between the largest and smallest eigenvalue of the adjacency matrix of $G$. Gotshall, O'Brien and Tait conjectured that for sufficiently large $n$, the $n$-vertex outerplanar graph with maximum spread is the graph obtained by joining a vertex to a path on $n-1$ vertices. In this paper, we disprove this conjecture by showing that the extremal graph is the graph obtained by joining a vertex to a path on $\lceil (2n-1)/3\rceil$ vertices and $\lfloor(n-2)/3\rfloor$ isolated vertices. For planar graphs, we show that the extremal $n$-vertex planar graph attaining the maximum spread is the graph obtained by joining two nonadjacent vertices to a path on $\lceil(2n-2)/3\rceil$ vertices and $\lfloor(n-4)/3\rfloor$ isolated vertices.

In recommendation systems, the choice of loss function is critical since a good loss may significantly improve the model performance. However, manually designing a good loss is a big challenge due to the complexity of the problem. A large fraction of previous work focuses on handcrafted loss functions, which needs significant expertise and human effort. In this paper, inspired by the recent development of automated machine learning, we propose an automatic loss function generation framework, AutoLossGen, which is able to generate loss functions directly constructed from basic mathematical operators without prior knowledge on loss structure. More specifically, we develop a controller model driven by reinforcement learning to generate loss functions, and develop iterative and alternating optimization schedule to update the parameters of both the controller model and the recommender model. One challenge for automatic loss generation in recommender systems is the extreme sparsity of recommendation datasets, which leads to the sparse reward problem for loss generation and search. To solve the problem, we further develop a reward filtering mechanism for efficient and effective loss generation. Experimental results show that our framework manages to create tailored loss functions for different recommendation models and datasets, and the generated loss gives better recommendation performance than commonly used baseline losses. Besides, most of the generated losses are transferable, i.e., the loss generated based on one model and dataset also works well for another model or dataset. Source code of the work is available at https://github.com/rutgerswiselab/AutoLossGen., 12 pages, 6 figures. Conference: Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR '22)

Recommender systems (RS), serving at the forefront of Human-centered AI, are widely deployed in almost every corner of the web and facilitate the human decision-making process. However, despite their enormous capabilities and potential, RS may also lead to undesired counter-effects on users, items, producers, platforms, or even the society at large, such as compromised user trust due to non-transparency, unfair treatment of different consumers, or producers, privacy concerns due to extensive use of user's private data for personalization, just to name a few. All of these create an urgent need for Trustworthy Recommender Systems (TRS) so as to mitigate or avoid such adverse impacts and risks. In this survey, we will introduce techniques related to trustworthy and responsible recommendation, including but not limited to explainable recommendation, fairness in recommendation, privacy-aware recommendation, robustness in recommendation, user controllable recommendation, as well as the relationship between these different perspectives in terms of trustworthy and responsible recommendation. Through this survey, we hope to deliver readers with a comprehensive view of the research area and raise attention to the community about the importance, existing research achievements, and future research directions on trustworthy recommendation.

The Observation--Hypothesis--Prediction--Experimentation loop paradigm for scientific research has been practiced by researchers for years towards scientific discoveries. However, with data explosion in both mega-scale and milli-scale scientific research, it has been sometimes very difficult to manually analyze the data and propose new hypotheses to drive the cycle for scientific discovery. In this paper, we discuss the role of Explainable AI in scientific discovery process by demonstrating an Explainable AI-based paradigm for science discovery. The key is to use Explainable AI to help derive data or model interpretations, hypotheses, as well as scientific discoveries or insights. We show how computational and data-intensive methodology -- together with experimental and theoretical methodology -- can be seamlessly integrated for scientific research. To demonstrate the AI-based science discovery process, and to pay our respect to some of the greatest minds in human history, we show how Kepler's laws of planetary motion and Newton's law of universal gravitation can be rediscovered by (Explainable) AI based on Tycho Brahe's astronomical observation data, whose works were leading the scientific revolution in the 16-17th century. This work also highlights the important role of Explainable AI (as compared to Blackbox AI) in science discovery to help humans prevent or better prepare for the possible technological singularity that may happen in the future, since science is not only about the know how, but also the know why. Presentation of the work is available at https://slideslive.com/38986142/from-kepler-to-newton-explainable-ai-for-science-discovery., Accepted by ICML-AI4Science 2022

While recent years have witnessed the emergence of various explainable methods in machine learning, to what degree the explanations really represent the reasoning process behind the model prediction -- namely, the faithfulness of explanation -- is still an open problem. One commonly used way to measure faithfulness is \textit{erasure-based} criteria. Though conceptually simple, erasure-based criterion could inevitably introduce biases and artifacts. We propose a new methodology to evaluate the faithfulness of explanations from the \textit{counterfactual reasoning} perspective: the model should produce substantially different outputs for the original input and its corresponding counterfactual edited on a faithful feature. Specially, we introduce two algorithms to find the proper counterfactuals in both discrete and continuous scenarios and then use the acquired counterfactuals to measure faithfulness. Empirical results on several datasets show that compared with existing metrics, our proposed counterfactual evaluation method can achieve top correlation with the ground truth under diffe

The theory of abstract kernels in non-trivial extensions for many kinds of algebraical objects, such as groups, rings and graded rings, associative algebras, Lie algebras, restricted Lie algebras, DG-algebras and DG-Lie algebras, has been widely studied since 1940's. Gerhard Hochschild firstly treats associative algebra as an generic type in the series of kernel problems. He proves the theorem of constructing kernel by presenting many tedious relations that may lost the readers today. In this paper, we shall illustrate the formulation and recast it for Lie algebra(-oid) kernels. We also prove the independence of 3-cocycle in the case of associative algebra. Finally, we use the universal enveloping algebra of Lie algebra to reduce the difficulty of a direct construction for the derivation algebras.

Although methanol synthesis via CO hydrogenation has been industrialized, CO2 hydrogenation to methanol still confronts great obstacles of low methanol selectivity and poor stability, particularly for supported metal catalysts under industrial conditions. We report a binary metal oxide, ZnO-ZrO2 solid solution catalyst, which can achieve methanol selectivity of up to 86 to 91% with CO2 single-pass conversion of more than 10% under reaction conditions of 5.0 MPa, 24,000 ml/(g hour), H2/CO2 = 3:1 to 4:1, 320° to 315°C. Experimental and theoretical results indicate that the synergetic effect between Zn and Zr sites results in the excellent performance. The ZnO-ZrO2 solid solution catalyst shows high stability for at least 500 hours on stream and is also resistant to sintering at higher temperatures. Moreover, no deactivation is observed in the presence of 50 ppm SO2 or H2S in the reaction stream.

To explore the effectiveness of a new method to repair severe soft tissue defects of the leg, foot, and ankle with contralateral saphenous neurocutaneous vascular flaps combined with assembly external frisket for fixation in parallel-leg position.Between August 2009 and August 2013, 29 cases with leg, foot, or ankle wound were treated. There were 18 males and 11 females, with an average age of 37.6 years (range, 11-65 years). The interval of injury and operation was 14-36 days (mean, 22.3 days). The locations were the planta pedis and heel in 5 cases, the dorsal foot in 2 cases, the ankle in 4 cases, middle and lower leg in 14 cases, and upper leg in 4 cases. The area of trauma ranged from 5 cm x 3 cm to 19 cm x 9 cm. The assembly external frisket was used for fixation in parallel-leg position; a bridge flap was transplanted to repair defects, and the area of flap ranged from 6 cm x 4 cm to 22 cm x 11 cm. The donor sites were directly sutured or repaired with skin graft. The pedicle of the bridge flap was cut off and the assembly external frisket was removed after 3-4 weeks.The flaps in all patients survived completely; primary healing of wound and incision at donor site was obtained. The patients were followed up 6-18 months (mean, 13.2 months). The appearance of flaps was satisfactory, and the sensation of the heel recovered. Moreover, the patients had a comfortable feeling, and the ankle and knee joints can move freely, and had good function.The assembly external frisket in parallel-leg position instead of cross-leg can make patients comfortable position, and achieve reliable fixation. The saphenous neurocutaneous vascular flap has the advantages of no need for anastomosis vein and for sacrifice of the main vein. Contralateral saphenous neurocutaneous vascular flap combined with assembly external frisket for fixation in parallel-leg position is a favorite method to repair severe soft tissue defects of the leg, foot, and ankle.