Multimodal Chain of Thought (MCoT) is a popular prompting strategy for improving the performance of multimodal large language models (MLLMs) across a range of complex reasoning tasks. Despite its popularity, there is a notable absence of automated methods for evaluating the quality of reasoning steps in MCoT. To address this gap, we propose Multimodal Chain-of-Thought Evaluation (MiCEval), a framework designed to assess the correctness of reasoning chains by evaluating the quality of both the description and each reasoning step. The evaluation of the description component focuses on the accuracy of the image descriptions, while the reasoning step evaluates the quality of each step as it is conditionally generated based on the preceding steps. MiCEval is built upon a fine-grained dataset with annotations that rate each step according to correctness, relevance, and informativeness. Extensive experiments on four state-of-the-art MLLMs show that step-wise evaluations using MiCEval align more closely with human judgments compared to existing methods based on cosine similarity or fine-tuning approaches. MiCEval datasets and code can be found in https://github.com/alenai97/MiCEval.

Attributed Question Answering (AQA) aims to provide both a trustworthy answer and a reliable attribution report for a given question. Retrieval is a widely adopted approach, including two general paradigms: Retrieval-Then-Read (RTR) and post-hoc retrieval. Recently, Large Language Models (LLMs) have shown remarkable proficiency, prompting growing interest in AQA among researchers. However, RTR-based AQA often suffers from irrelevant knowledge and rapidly changing information, even when LLMs are adopted, while post-hoc retrieval-based AQA struggles with comprehending long-form answers with complex logic, and precisely identifying the content needing revision and preserving the original intent. To tackle these problems, this paper proposes an Atomic fact decomposition-based Retrieval and Editing (ARE) framework, which decomposes the generated long-form answers into molecular clauses and atomic facts by the instruction-tuned LLMs. Notably, the instruction-tuned LLMs are fine-tuned using a well-constructed dataset, generated from large scale Knowledge Graphs (KGs). This process involves extracting one-hop neighbors from a given set of entities and transforming the result into coherent long-form text. Subsequently, ARE leverages a search engine to retrieve evidences related to atomic facts, inputting these evidences into an LLM-based verifier to determine whether the facts require expansion for re-retrieval or editing. Furthermore, the edited facts are backtracked into the original answer, with evidence aggregated based on the relationship between molecular clauses and atomic facts. Extensive evaluations demonstrate the superior performance of our proposed method over the state-of-the-arts on several datasets, with an additionally proposed new metric $Attr_{p}$ for evaluating the precision of evidence attribution.

Large Language Models (LLMs) can interact with the real world by connecting with versatile external APIs, resulting in better problem-solving and task automation capabilities. Previous research primarily focuses on APIs with limited arguments from a single source or overlooks the complex dependency relationship between different APIs. However, it is essential to utilize multiple APIs collaboratively from various sources (e.g., different Apps in the iPhone), especially for complex user instructions. In this paper, we introduce \texttt{AppBench}, the first benchmark to evaluate LLMs' ability to plan and execute multiple APIs from various sources in order to complete the user's task. Specifically, we consider two significant challenges in multiple APIs: \textit{1) graph structures:} some APIs can be executed independently while others need to be executed one by one, resulting in graph-like execution order; and \textit{2) permission constraints:} which source is authorized to execute the API call. We have experimental results on 9 distinct LLMs; e.g., GPT-4o achieves only a 2.0\% success rate at the most complex instruction, revealing that the existing state-of-the-art LLMs still cannot perform well in this situation even with the help of in-context learning and finetuning. Our code and data are publicly available at https://github.com/ruleGreen/AppBench.

Recent studies have explored the use of Large Language Models (LLMs) with Retrieval Augmented Generation (RAG) for Knowledge Graph Question Answering (KGQA). They typically require rewriting retrieved subgraphs into natural language formats comprehensible to LLMs. However, when tackling complex questions, the knowledge rewritten by existing methods may include irrelevant information, omit crucial details, or fail to align with the question's semantics. To address them, we propose a novel rewriting method CoTKR, Chain-of-Thought Enhanced Knowledge Rewriting, for generating reasoning traces and corresponding knowledge in an interleaved manner, thereby mitigating the limitations of single-step knowledge rewriting. Additionally, to bridge the preference gap between the knowledge rewriter and the question answering (QA) model, we propose a training strategy PAQAF, Preference Alignment from Question Answering Feedback, for leveraging feedback from the QA model to further optimize the knowledge rewriter. We conduct experiments using various LLMs across several KGQA benchmarks. Experimental results demonstrate that, compared with previous knowledge rewriting methods, CoTKR generates the most beneficial knowledge representation for QA models, which significantly improves the performance of LLMs in KGQA.

Retrieval-Augmented Generation (RAG) is widely used to inject external non-parametric knowledge into large language models (LLMs). Recent works suggest that Knowledge Graphs (KGs) contain valuable external knowledge for LLMs. Retrieving information from KGs differs from extracting it from document sets. Most existing approaches seek to directly retrieve relevant subgraphs, thereby eliminating the need for extensive SPARQL annotations, traditionally required by semantic parsing methods. In this paper, we model the subgraph retrieval task as a conditional generation task handled by small language models. Specifically, we define a subgraph identifier as a sequence of relations, each represented as a special token stored in the language models. Our base generative subgraph retrieval model, consisting of only 220M parameters, achieves competitive retrieval performance compared to state-of-the-art models relying on 7B parameters, demonstrating that small language models are capable of performing the subgraph retrieval task. Furthermore, our largest 3B model, when plugged with an LLM reader, sets new SOTA end-to-end performance on both the WebQSP and CWQ benchmarks. Our model and data will be made available online: https://github.com/hwy9855/GSR.

The NLP community has recently shown a growing interest in leveraging Large Language Models (LLMs) for knowledge-intensive tasks, viewing LLMs as potential knowledge bases (KBs). However, the reliability and extent to which LLMs can function as KBs remain underexplored. While previous studies suggest LLMs can encode knowledge within their parameters, the amount of parametric knowledge alone is not sufficient to evaluate their effectiveness as KBs. This study defines criteria that a reliable LLM-as-KB should meet, focusing on factuality and consistency, and covering both seen and unseen knowledge. We develop several metrics based on these criteria and use them to evaluate 26 popular LLMs, while providing a comprehensive analysis of the effects of model size, instruction tuning, and in-context learning (ICL). Our results paint a worrying picture. Even a high-performant model like GPT-3.5-turbo is not factual or consistent, and strategies like ICL and fine-tuning are unsuccessful at making LLMs better KBs.

With the emergence of large language models, such as LLaMA and OpenAI GPT-3, In-Context Learning (ICL) gained significant attention due to its effectiveness and efficiency. However, ICL is very sensitive to the choice, order, and verbaliser used to encode the demonstrations in the prompt. Retrieval-Augmented ICL methods try to address this problem by leveraging retrievers to extract semantically related examples as demonstrations. While this approach yields more accurate results, its robustness against various types of adversarial attacks, including perturbations on test samples, demonstrations, and retrieved data, remains under-explored. Our study reveals that retrieval-augmented models can enhance robustness against test sample attacks, outperforming vanilla ICL with a 4.87% reduction in Attack Success Rate (ASR); however, they exhibit overconfidence in the demonstrations, leading to a 2% increase in ASR for demonstration attacks. Adversarial training can help improve the robustness of ICL methods to adversarial attacks; however, such a training scheme can be too costly in the context of LLMs. As an alternative, we introduce an effective training-free adversarial defence method, DARD, which enriches the example pool with those attacked samples. We show that DARD yields improvements in performance and robustness, achieving a 15% reduction in ASR over the baselines. Code and data are released to encourage further research: https://github.com/simonucl/adv-retreival-icl

We focus on Text-to-SQL semantic parsing from the perspective of Large Language Models. Motivated by challenges related to the size of commercial database schemata and the deployability of business intelligence solutions, we propose an approach that dynamically retrieves input database information and uses abstract syntax trees to select few-shot examples for in-context learning. Furthermore, we investigate the extent to which an in-parallel semantic parser can be leveraged for generating $\textit{approximated}$ versions of the expected SQL queries, to support our retrieval. We take this approach to the extreme--we adapt a model consisting of less than $500$M parameters, to act as an extremely efficient approximator, enhancing it with the ability to process schemata in a parallelised manner. We apply our approach to monolingual and cross-lingual benchmarks for semantic parsing, showing improvements over state-of-the-art baselines. Comprehensive experiments highlight the contribution of modules involved in this retrieval-augmented generation setting, revealing interesting directions for future work.

Knowledge graph (KG) completion aims to find out missing triples in a KG. Some tasks, such as link prediction and instance completion, have been proposed for KG completion. They are triple-level tasks with some elements in a missing triple given to predict the missing element of the triple. However, knowing some elements of the missing triple in advance is not always a realistic setting. In this paper, we propose a novel graph-level automatic KG completion task called Triple Set Prediction (TSP) which assumes none of the elements in the missing triples is given. TSP is to predict a set of missing triples given a set of known triples. To properly and accurately evaluate this new task, we propose 4 evaluation metrics including 3 classification metrics and 1 ranking metric, considering both the partial-open-world and the closed-world assumptions. Furthermore, to tackle the huge candidate triples for prediction, we propose a novel and efficient subgraph-based method GPHT that can predict the triple set fast. To fairly compare the TSP results, we also propose two types of methods RuleTensor-TSP and KGE-TSP applying the existing rule- and embedding-based methods for TSP as baselines. During experiments, we evaluate the proposed methods on two datasets extracted from Wikidata following the relation-similarity partial-open-world assumption proposed by us, and also create a complete family data set to evaluate TSP results following the closed-world assumption. Results prove that the methods can successfully generate a set of missing triples and achieve reasonable scores on the new task, and GPHT performs better than the baselines with significantly shorter prediction time. The datasets and code for experiments are available at https://github.com/zjukg/GPHT-for-TSP.

Improving the performance of large language models (LLMs) in complex question-answering (QA) scenarios has always been a research focal point. Recent studies have attempted to enhance LLMs' performance by combining step-wise planning with external retrieval. While effective for advanced models like GPT-3.5, smaller LLMs face challenges in decomposing complex questions, necessitating supervised fine-tuning. Previous work has relied on manual annotation and knowledge distillation from teacher LLMs, which are time-consuming and not accurate enough. In this paper, we introduce a novel framework for enhancing LLMs' planning capabilities by using planning data derived from knowledge graphs (KGs). LLMs fine-tuned with this data have improved planning capabilities, better equipping them to handle complex QA tasks that involve retrieval. Evaluations on multiple datasets, including our newly proposed benchmark, highlight the effectiveness of our framework and the benefits of KG-derived planning data.