Like humans, large language models (LLMs) do not always generate the best output on their first try. Motivated by how humans refine their written text, we introduce Self-Refine, an approach for improving initial outputs from LLMs through iterative feedback and refinement. The main idea is to generate an initial output using an LLMs; then, the same LLMs provides feedback for its output and uses it to refine itself, iteratively. Self-Refine does not require any supervised training data, additional training, or reinforcement learning, and instead uses a single LLM as the generator, refiner, and feedback provider. We evaluate Self-Refine across 7 diverse tasks, ranging from dialog response generation to mathematical reasoning, using state-of-the-art (GPT-3.5, ChatGPT, and GPT-4) LLMs. Across all evaluated tasks, outputs generated with Self-Refine are preferred by humans and automatic metrics over those generated with the same LLM using conventional one-step generation, improving by ~20% absolute on average in task performance. Our work demonstrates that even state-of-the-art LLMs like GPT-4 can be further improved at test time using our simple, standalone approach.

Self-Refine: Iterative Refinement with Self-Feedback

Language models (LMs) have recently shown remarkable performance on reasoning tasks by explicitly generating intermediate inferences, e.g., chain-of-thought prompting. However, these intermediate inference steps may be inappropriate deductions from the initial context and lead to incorrect final predictions. Here we introduce REFINER, a framework for finetuning LMs to explicitly generate intermediate reasoning steps while interacting with a critic model that provides automated feedback on the reasoning. Specifically, the critic provides structured feedback that the reasoning LM uses to iteratively improve its intermediate arguments. Empirical evaluations of REFINER on three diverse reasoning tasks show significant improvements over baseline LMs of comparable scale. Furthermore, when using GPT3.5 as the reasoner, the trained critic significantly improves reasoning without finetuning the reasoner. Finally, our critic model is trained without expensive human-in-the-loop data but can be substituted with humans at inference time.

REFINER: Reasoning Feedback on Intermediate Representations

Social media platforms have been establishing content moderation guidelines and employing various moderation policies to counter hate speech and misinformation. The goal of this paper is to study these community guidelines and moderation practices, as well as the relevant research publications, to identify the research gaps, differences in moderation techniques, and challenges that should be tackled by the social media platforms and the research community. To this end, we study and analyze fourteen most popular social media content moderation guidelines and practices, and consolidate them. We then introduce three taxonomies drawn from this analysis as well as covering over two hundred interdisciplinary research papers about moderation strategies. We identify the differences between the content moderation employed in mainstream and fringe social media platforms. Finally, we have in-depth applied discussions on both research and practical challenges and solutions.

SoK: Content Moderation in Social Media, from Guidelines to Enforcement, and Research to Practice

Large Language Models (LLMs) have emerged as powerful tools in the field of Natural Language Processing (NLP) and have recently gained significant attention in the domain of Recommendation Systems (RS). These models, trained on massive amounts of data using self-supervised learning, have demonstrated remarkable success in learning universal representations and have the potential to enhance various aspects of recommendation systems by some effective transfer techniques such as fine-tuning and prompt tuning, and so on. The crucial aspect of harnessing the power of language models in enhancing recommendation quality is the utilization of their high-quality representations of textual features and their extensive coverage of external knowledge to establish correlations between items and users. To provide a comprehensive understanding of the existing LLM-based recommendation systems, this survey presents a taxonomy that categorizes these models into two major paradigms, respectively Discriminative LLM for Recommendation (DLLM4Rec) and Generative LLM for Recommendation (GLLM4Rec), with the latter being systematically sorted out for the first time. Furthermore, we systematically review and analyze existing LLM-based recommendation systems within each paradigm, providing insights into their methodologies, techniques, and performance. Additionally, we identify key challenges and several valuable findings to provide researchers and practitioners with inspiration. We have also created a GitHub repository to index relevant papers on LLMs for recommendation, https://github.com/WLiK/LLM4Rec.

A Survey on Large Language Models for Recommendation

Twitter bots are automatic programs operated by malicious actors to manipulate public opinion and spread misinformation. Research efforts have been made to automatically identify bots based on texts and networks on social media. Existing methods only leverage texts or networks alone, and while few works explored the shallow combination of the two modalities, we hypothesize that the interaction and information exchange between texts and graphs could be crucial for holistically evaluating bot activities on social media. In addition, according to a recent survey (Cresci, 2020), Twitter bots are constantly evolving while advanced bots steal genuine users’ tweets and dilute their malicious content to evade detection. This results in greater inconsistency across the timeline of novel Twitter bots, which warrants more attention. In light of these challenges, we propose BIC, a Twitter Bot detection framework with text-graph Interaction and semantic Consistency. Specifically, in addition to separately modeling the two modalities on social media, BIC employs a text-graph interaction module to enable information exchange across modalities in the learning process. In addition, given the stealing behavior of novel Twitter bots, BIC proposes to model semantic consistency in tweets based on attention weights while using it to augment the decision process. Extensive experiments demonstrate that BIC consistently outperforms state-of-the-art baselines on two widely adopted datasets. Further analyses reveal that text-graph interactions and modeling semantic consistency are essential improvements and help combat bot evolution.

BIC: Twitter Bot Detection with Text-Graph Interaction and Semantic Consistency

Easy access, variety of content, and fast widespread interactions are some of the reasons making social media increasingly popular. However, this rise has also enabled the propagation of fake news, text published by news sources with an intent to spread misinformation and sway beliefs. Detecting it is an important and challenging problem to prevent large scale misinformation and maintain a healthy society. We view fake news detection as reasoning over the relations between sources, articles they publish, and engaging users on social media in a graph framework. After embedding this information, we formulate inference operators which augment the graph edges by revealing unobserved interactions between its elements, such as similarity between documents’ contents and users’ engagement patterns. Our experiments over two challenging fake news detection tasks show that using inference operators leads to a better understanding of the social media framework enabling fake news spread, resulting in improved performance.

/pdf/tackling-fake-news-detection-by-continually-improving-social-1ks1tbhs.pdf

Tackling Fake News Detection by Continually Improving Social Context Representations using Graph Neural Networks

Over the last few years, there has been growing interest in learning models for physically grounded language understanding tasks, such as the popular blocks world domain. These works typically view this problem as a single-step process, in which a human operator gives an instruction and an automated agent is evaluated on its ability to execute it. In this paper we take the first step towards increasing the bandwidth of this interaction, and suggest a protocol for including advice, high-level observations about the task, which can help constrain the agent’s prediction. We evaluate our approach on the blocks world task, and show that even simple advice can help lead to significant performance improvements. To help reduce the effort involved in supplying the advice, we also explore model self-generated advice which can still improve results.

Improving Natural Language Interaction with Robots Using Advice

Large Language Models (LLMs) have demonstrated exceptional capabilities in generalizing to new tasks in a zero-shot or few-shot manner. However, the extent to which LLMs can comprehend user preferences based on their previous behavior remains an emerging and still unclear research question. Traditionally, Collaborative Filtering (CF) has been the most effective method for these tasks, predominantly relying on the extensive volume of rating data. In contrast, LLMs typically demand considerably less data while maintaining an exhaustive world knowledge about each item, such as movies or products. In this paper, we conduct a thorough examination of both CF and LLMs within the classic task of user rating prediction, which involves predicting a user's rating for a candidate item based on their past ratings. We investigate various LLMs in different sizes, ranging from 250M to 540B parameters and evaluate their performance in zero-shot, few-shot, and fine-tuning scenarios. We conduct comprehensive analysis to compare between LLMs and strong CF methods, and find that zero-shot LLMs lag behind traditional recommender models that have the access to user interaction data, indicating the importance of user interaction data. However, through fine-tuning, LLMs achieve comparable or even better performance with only a small fraction of the training data, demonstrating their potential through data efficiency.

Do LLMs Understand User Preferences? Evaluating LLMs On User Rating Prediction

Over the last few years, there has been growing interest in learning models for physically grounded language understanding tasks, such as the popular blocks world domain. These works typically view this problem as a single-step process, in which a human operator gives an instruction and an automated agent is evaluated on its ability to execute it. In this paper we take the first step towards increasing the bandwidth of this interaction, and suggest a protocol for including advice, high-level observations about the task, which can help constrain the agent's prediction. We evaluate our approach on the blocks world task, and show that even simple advice can help lead to significant performance improvements. To help reduce the effort involved in supplying the advice, we also explore model self-generated advice which can still improve results.

Training good representations for items is critical in recommender models. Typically, an item is assigned a unique randomly generated ID, and is commonly represented by learning an embedding corresponding to the value of the random ID. Although widely used, this approach have limitations when the number of items are large and items are power-law distributed -- typical characteristics of real-world recommendation systems. This leads to the item cold-start problem, where the model is unable to make reliable inferences for tail and previously unseen items. Removing these ID features and their learned embeddings altogether to combat cold-start issue severely degrades the recommendation quality. Content-based item embeddings are more reliable, but they are expensive to store and use, particularly for users' past item interaction sequence. In this paper, we use Semantic IDs, a compact discrete item representations learned from content embeddings using RQ-VAE that captures hierarchy of concepts in items. We showcase how we use them as a replacement of item IDs in a resource-constrained ranking model used in an industrial-scale video sharing platform. Moreover, we show how Semantic IDs improves the generalization ability of our system, without sacrificing top-level metrics.

Nikhil Mehta

Papers

Tackling Fake News Detection by Continually Improving Social Context Representations using Graph Neural Networks

Improving Natural Language Interaction with Robots Using Advice

Do LLMs Understand User Preferences? Evaluating LLMs On User Rating Prediction

Improving Natural Language Interaction with Robots Using Advice

Better Generalization with Semantic IDs: A case study in Ranking for Recommendations