Large Language Models (LLMs) have shown their ability to collaborate effectively with humans in real-world scenarios. However, LLMs are apt to generate hallucinations, i.e., makeup incorrect text and unverified information, which can cause significant damage when deployed for mission-critical tasks. In this paper, we propose a self-check approach based on reverse validation to detect factual errors automatically in a zero-resource fashion. To facilitate future studies and assess different methods, we construct a hallucination detection benchmark named PHD, which is generated by ChatGPT and annotated by human annotators. Contrasting previous studies of zero-resource hallucination detection, our method and benchmark concentrate on passage-level detection instead of sentence-level. We empirically evaluate our method and existing zero-resource detection methods on two datasets. The experimental results demonstrate that the proposed method considerably outperforms the baselines while costing fewer tokens and less time. Furthermore, we manually analyze some hallucination cases that LLM failed to capture, revealing the shared limitation of zero-resource methods.

We introduce ECHo (Event Causality Inference via Human-Centric Reasoning), a diagnostic dataset of event causality inference grounded in visio-linguistic social scenarios. ECHo employs real-world human-centric deductive information building on a television crime drama. ECHo requires the Theory-of-Mind (ToM) ability to understand and reason about social interactions based on multimodal information. Using ECHo, we propose a unified Chain-of-Thought (CoT) framework to assess the reasoning capability of current AI systems. Our ToM-enhanced CoT pipeline accommodates various large foundation models in both zero-shot and few-shot visio-linguistic reasoning. We use this framework to scrutinize recent large foundation models such as InstructGPT and MiniGPT-4 on three diagnostic human-centric tasks. Further analysis demonstrates ECHo as a challenging dataset to expose imperfections and inconsistencies in reasoning. Our data and code are publicly available at https://github.com/YuxiXie/ECHo.

We present PeerSum, a novel dataset for generating meta-reviews of scientific papers. The meta-reviews can be interpreted as abstractive summaries of reviews, multi-turn discussions and the paper abstract. These source documents have rich inter-document relationships with an explicit hierarchical conversational structure, cross-references and (occasionally) conflicting information. To introduce the structural inductive bias into pre-trained language models, we introduce Rammer ( Relationship-aware Multi-task Meta-review Generator), a model that uses sparse attention based on the conversational structure and a multi-task training objective that predicts metadata features (e.g., review ratings). Our experimental results show that Rammer outperforms other strong baseline models in terms of a suite of automatic evaluation metrics. Further analyses, however, reveal that RAMMER and other models struggle to handle conflicts in source documents of PeerSum, suggesting meta-review generation is a challenging task and a promising avenue for further research.

In this study, we explore the impact of network topology on the approximation capabilities of artificial neural networks (ANNs), with a particular focus on complex topologies. We propose a novel methodology for constructing complex ANNs based on various topologies, including Barab\'asi-Albert, Erd\H{o}s-R\'enyi, Watts-Strogatz, and multilayer perceptrons (MLPs). The constructed networks are evaluated on synthetic datasets generated from manifold learning generators, with varying levels of task difficulty and noise, and on real-world datasets from the UCI suite. Our findings reveal that complex topologies lead to superior performance in high-difficulty regimes compared to traditional MLPs. This performance advantage is attributed to the ability of complex networks to exploit the compositionality of the underlying target function. However, this benefit comes at the cost of increased forward-pass computation time and reduced robustness to graph damage. Additionally, we investigate the relationship between various topological attributes and model performance. Our analysis shows that no single attribute can account for the observed performance differences, suggesting that the influence of network topology on approximation capabilities may be more intricate than a simple correlation with individual topological attributes. Our study sheds light on the potential of complex topologies for enhancing the performance of ANNs and provides a foundation for future research exploring the interplay between multiple topological attributes and their impact on model performance.

Many natural language processing tasks benefit from long inputs, but processing long documents with Transformers is expensive -- not only due to quadratic attention complexity but also from applying feedforward and projection layers to every token. However, not all tokens are equally important, especially for longer documents. We propose CoLT5, a long-input Transformer model that builds on this intuition by employing conditional computation, devoting more resources to important tokens in both feedforward and attention layers. We show that CoLT5 achieves stronger performance than LongT5 with much faster training and inference, achieving SOTA on the long-input SCROLLS benchmark. Moreover, CoLT5 can effectively and tractably make use of extremely long inputs, showing strong gains up to 64k input length.

Knowledge graphs represent known facts using triplets. While existing knowledge graph embedding methods only consider the connections between entities, we propose considering the relationships between triplets. For example, let us consider two triplets $T_1$ and $T_2$ where $T_1$ is (Academy_Awards, Nominates, Avatar) and $T_2$ is (Avatar, Wins, Academy_Awards). Given these two base-level triplets, we see that $T_1$ is a prerequisite for $T_2$. In this paper, we define a higher-level triplet to represent a relationship between triplets, e.g., $\langle T_1$, PrerequisiteFor, $T_2\rangle$ where PrerequisiteFor is a higher-level relation. We define a bi-level knowledge graph that consists of the base-level and the higher-level triplets. We also propose a data augmentation strategy based on the random walks on the bi-level knowledge graph to augment plausible triplets. Our model called BiVE learns embeddings by taking into account the structures of the base-level and the higher-level triplets, with additional consideration of the augmented triplets. We propose two new tasks: triplet prediction and conditional link prediction. Given a triplet $T_1$ and a higher-level relation, the triplet prediction predicts a triplet that is likely to be connected to $T_1$ by the higher-level relation, e.g., $\langle T_1$, PrerequisiteFor, ?$\rangle$. The conditional link prediction predicts a missing entity in a triplet conditioned on another triplet, e.g., $\langle T_1$, PrerequisiteFor, (Avatar, Wins, ?)$\rangle$. Experimental results show that BiVE significantly outperforms all other methods in the two new tasks and the typical base-level link prediction in real-world bi-level knowledge graphs.

Recent years have seen the rapid development of large generative models for text; however, much less research has explored the connection between text and another "language" of communication -- music. Music, much like text, can convey emotions, stories, and ideas, and has its own unique structure and syntax. In our work, we bridge text and music via a text-to-music generation model that is highly efficient, expressive, and can handle long-term structure. Specifically, we develop Mo\^usai, a cascading two-stage latent diffusion model that can generate multiple minutes of high-quality stereo music at 48kHz from textual descriptions. Moreover, our model features high efficiency, which enables real-time inference on a single consumer GPU with a reasonable speed. Through experiments and property analyses, we show our model's competence over a variety of criteria compared with existing music generation models. Lastly, to promote the open-source culture, we provide a collection of open-source libraries with the hope of facilitating future work in the field. We open-source the following: Codes: https://github.com/archinetai/audio-diffusion-pytorch; music samples for this paper: http://bit.ly/44ozWDH; all music samples for all models: https://bit.ly/audio-diffusion.

Can non-programmers annotate natural language utterances with complex programs that represent their meaning? We introduce APEL, a framework in which non-programmers select among candidate programs generated by a seed semantic parser (e.g., Codex). Since they cannot understand the candidate programs, we ask them to select indirectly by examining the programs' input-ouput examples. For each utterance, APEL actively searches for a simple input on which the candidate programs tend to produce different outputs. It then asks the non-programmers only to choose the appropriate output, thus allowing us to infer which program is correct and could be used to fine-tune the parser. As a first case study, we recruited human non-programmers to use APEL to re-annotate SPIDER, a text-to-SQL dataset. Our approach achieved the same annotation accuracy as the original expert annotators (75%) and exposed many subtle errors in the original annotations.

Constrained submodular maximization is a fundamental problem at the heart of discrete optimization. The reason for this is as simple as it is clear: submodular functions capture the notion of diminishing returns present in a wide variety of real-world settings. Consequently to its striking importance and coinciding NP-hardness [20], extensive research has been conducted on submodular maximization since the seventies (e.g., [15, 42]), with focus lately shifting towards handling the massive datasets emerging in modern applications. With a wide variety of possible constraints, often regarding cardinality, independence in a matroid, or knapsacktype restrictions, the number of applications is vast. To name just a few, there are recent works on feature selection in machine learning [13, 14, 32], influence maximization in viral marketing [2, 31], and data summarization [43, 38, 45]. Many of these applications have non-monotone submodular objectives, meaning that adding an element to an existing set might actually decrease its value. Two such examples are discussed in detail in Section 5. This work was supported by the ERC Advanced Grant 788893 AMDROMA "Algorithmic and Mechanism Design Research in Online Markets" and the MIUR PRIN project ALGADIMAR "Algorithms, Games, and Digital Markets."

Large Language Models (LLMs) play powerful, black-box readers in the retrieve-then-read pipeline, making remarkable progress in knowledge-intensive tasks. This work introduces a new framework, Rewrite-Retrieve-Read instead of the previous retrieve-then-read for the retrieval-augmented LLMs from the perspective of the query rewriting. Unlike prior studies focusing on adapting either the retriever or the reader, our approach pays attention to the adaptation of the search query itself, for there is inevitably a gap between the input text and the needed knowledge in retrieval. We first prompt an LLM to generate the query, then use a web search engine to retrieve contexts. Furthermore, to better align the query to the frozen modules, we propose a trainable scheme for our pipeline. A small language model is adopted as a trainable rewriter to cater to the black-box LLM reader. The rewriter is trained using the feedback of the LLM reader by reinforcement learning. Evaluation is conducted on downstream tasks, open-domain QA and multiple-choice QA. Experiments results show consistent performance improvement, indicating that our framework is proven effective and scalable, and brings a new framework for retrieval-augmented LLM.