FlowBench: Revisiting and Benchmarking Workflow-Guided Planning for LLM-based Agents

LLM-based agents have emerged as promising tools, which are crafted to fulfill complex tasks by iterative planning and action. However, these agents are susceptible to undesired planning hallucinations when lacking specific knowledge for expertise-intensive tasks. To address this, preliminary attempts are made to enhance planning reliability by incorporating external workflow-related knowledge. Despite the promise, such infused knowledge is mostly disorganized and diverse in formats, lacking rigorous formalization and comprehensive comparisons. Motivated by this, we formalize different formats of workflow knowledge and present FlowBench, the first benchmark for workflow-guided planning. FlowBench covers 51 different scenarios from 6 domains, with knowledge presented in diverse formats. To assess different LLMs on FlowBench, we design a multi-tiered evaluation framework. We evaluate the efficacy of workflow knowledge across multiple formats, and the results indicate that current LLM agents need considerable improvements for satisfactory planning. We hope that our challenging benchmark can pave the way for future agent planning research.

PTA: Enhancing Multimodal Sentiment Analysis through Pipelined Prediction and Translation-based Alignment

Multimodal aspect-based sentiment analysis (MABSA) aims to understand opinions in a granular manner, advancing human-computer interaction and other fields. Traditionally, MABSA methods use a joint prediction approach to identify aspects and sentiments simultaneously. However, we argue that joint models are not always superior. Our analysis shows that joint models struggle to align relevant text tokens with image patches, leading to misalignment and ineffective image utilization. In contrast, a pipeline framework first identifies aspects through MATE (Multimodal Aspect Term Extraction) and then aligns these aspects with image patches for sentiment classification (MASC: Multimodal Aspect-Oriented Sentiment Classification). This method is better suited for multimodal scenarios where effective image use is crucial. We present three key observations: (a) MATE and MASC have different feature requirements, with MATE focusing on token-level features and MASC on sequence-level features; (b) the aspect identified by MATE is crucial for effective image utilization; and (c) images play a trivial role in previous MABSA methods due to high noise. Based on these observations, we propose a pipeline framework that first predicts the aspect and then uses translation-based alignment (TBA) to enhance multimodal semantic consistency for better image utilization. Our method achieves state-of-the-art (SOTA) performance on widely used MABSA datasets Twitter-15 and Twitter-17. This demonstrates the effectiveness of the pipeline approach and its potential to provide valuable insights for future MABSA research. For reproducibility, the code and checkpoint will be released.

Self-Explanation Prompting Improves Dialogue Understanding in Large Language Models

Task-oriented dialogue (TOD) systems facilitate users in executing various activities via multi-turn dialogues, but Large Language Models (LLMs) often struggle to comprehend these intricate contexts. In this study, we propose a novel "Self-Explanation" prompting strategy to enhance the comprehension abilities of LLMs in multi-turn dialogues. This task-agnostic approach requires the model to analyze each dialogue utterance before task execution, thereby improving performance across various dialogue-centric tasks. Experimental results from six benchmark datasets confirm that our method consistently outperforms other zero-shot prompts and matches or exceeds the efficacy of few-shot prompts, demonstrating its potential as a powerful tool in enhancing LLMs' comprehension in complex dialogue tasks.

UniPCM: Universal Pre-trained Conversation Model with Task-aware Automatic Prompt

Recent research has shown that multi-task pre-training greatly improves the model's robustness and transfer ability, which is crucial for building a high-quality dialog system. However, most previous works on multi-task pre-training rely heavily on human-defined input format or prompt, which is not optimal in quality and quantity. In this work, we propose to use Task-based Automatic Prompt generation (TAP) to automatically generate high-quality prompts. Using the high-quality prompts generated, we scale the corpus of the pre-trained conversation model to 122 datasets from 15 dialog-related tasks, resulting in Universal Pre-trained Conversation Model (UniPCM), a powerful foundation model for various conversational tasks and different dialog systems. Extensive experiments have shown that UniPCM is robust to input prompts and capable of various dialog-related tasks. Moreover, UniPCM has strong transfer ability and excels at low resource scenarios, achieving SOTA results on 9 different datasets ranging from task-oriented dialog to open-domain conversation. Furthermore, we are amazed to find that TAP can generate prompts on par with those collected with crowdsourcing. The code is released with the paper.

SpokenWOZ: A Large-Scale Speech-Text Benchmark for Spoken Task-Oriented Dialogue in Multiple Domains

Task-oriented dialogue (TOD) models have great progress in the past few years. However, these studies primarily focus on datasets written by annotators, which has resulted in a gap between academic research and more realistic spoken conversation scenarios. While a few small-scale spoken TOD datasets are proposed to address robustness issues, e.g., ASR errors, they fail to identify the unique challenges in spoken conversation. To tackle the limitations, we introduce SpokenWOZ, a large-scale speech-text dataset for spoken TOD, which consists of 8 domains, 203k turns, 5.7k dialogues and 249 hours of audios from human-to-human spoken conversations. SpokenWOZ incorporates common spoken characteristics such as word-by-word processing and commonsense reasoning. We also present cross-turn slot and reasoning slot detection as new challenges based on the spoken linguistic phenomena. We conduct comprehensive experiments on various models, including text-modal baselines, newly proposed dual-modal baselines and LLMs. The results show the current models still has substantial areas for improvement in spoken conversation, including fine-tuned models and LLMs, i.e., ChatGPT.

Speech-Text Dialog Pre-training for Spoken Dialog Understanding with Explicit Cross-Modal Alignment

Recently, speech-text pre-training methods have shown remarkable success in many speech and natural language processing tasks. However, most previous pre-trained models are usually tailored for one or two specific tasks, but fail to conquer a wide range of speech-text tasks. In addition, existing speech-text pre-training methods fail to explore the contextual information within a dialogue to enrich utterance representations. In this paper, we propose Speech-text dialog Pre-training for spoken dialog understanding with ExpliCiT cRoss-Modal Alignment (SPECTRA), which is the first-ever speech-text dialog pre-training model. Concretely, to consider the temporality of speech modality, we design a novel temporal position prediction task to capture the speech-text alignment. This pre-training task aims to predict the start and end time of each textual word in the corresponding speech waveform. In addition, to learn the characteristics of spoken dialogs, we generalize a response selection task from textual dialog pre-training to speech-text dialog pre-training scenarios. Experimental results on four different downstream speech-text tasks demonstrate the superiority of SPECTRA in learning speech-text alignment and multi-turn dialog context.

Gate Recurrent Unit Network based on Hilbert-Schmidt Independence Criterion for State-of-Health Estimation

State-of-health (SOH) estimation is a key step in ensuring the safe and reliable operation of batteries. Due to issues such as varying data distribution and sequence length in different cycles, most existing methods require health feature extraction technique, which can be time-consuming and labor-intensive. GRU can well solve this problem due to the simple structure and superior performance, receiving widespread attentions. However, redundant information still exists within the network and impacts the accuracy of SOH estimation. To address this issue, a new GRU network based on Hilbert-Schmidt Independence Criterion (GRU-HSIC) is proposed. First, a zero masking network is used to transform all battery data measured with varying lengths every cycle into sequences of the same length, while still retaining information about the original data size in each cycle. Second, the Hilbert-Schmidt Independence Criterion (HSIC) bottleneck, which evolved from Information Bottleneck (IB) theory, is extended to GRU to compress the information from hidden layers. To evaluate the proposed method, we conducted experiments on datasets from the Center for Advanced Life Cycle Engineering (CALCE) of the University of Maryland and NASA Ames Prognostics Center of Excellence. Experimental results demonstrate that our model achieves higher accuracy than other recurrent models.

Bilingual Alignment Pre-training for Zero-shot Cross-lingual Transfer

Multilingual pre-trained models have achieved remarkable transfer performance by pre-trained on rich kinds of languages. Most of the models such as mBERT are pre-trained on unlabeled corpora. The static and contextual embeddings from the models could not be aligned very well. In this paper, we aim to improve the zero-shot cross-lingual transfer performance by aligning the embeddings better. We propose a pre-training task named Alignment Language Model (AlignLM), which uses the statistical alignment information as the prior knowledge to guide bilingual word prediction. We evaluate our method on multilingual machine reading comprehension and natural language interface tasks. The results show AlignLM can improve the zero-shot performance significantly on MLQA and XNLI datasets.

CharBERT: Character-aware Pre-trained Language Model

Most pre-trained language models (PLMs) construct word representations at subword level with Byte-Pair Encoding (BPE) or its variations, by which OOV (out-of-vocab) words are almost avoidable. However, those methods split a word into subword units and make the representation incomplete and fragile. In this paper, we propose a character-aware pre-trained language model named CharBERT improving on the previous methods (such as BERT, RoBERTa) to tackle these problems. We first construct the contextual word embedding for each token from the sequential character representations, then fuse the representations of characters and the subword representations by a novel heterogeneous interaction module. We also propose a new pre-training task named NLM (Noisy LM) for unsupervised character representation learning. We evaluate our method on question answering, sequence labeling, and text classification tasks, both on the original datasets and adversarial misspelling test sets. The experimental results show that our method can significantly improve the performance and robustness of PLMs simultaneously. Pretrained models, evaluation sets, and code are available at https://github.com/wtma/CharBERT

Benchmarking Robustness of Machine Reading Comprehension Models

Machine Reading Comprehension (MRC) is an important testbed for evaluating models' natural language understanding (NLU) ability. There has been rapid progress in this area, with new models achieving impressive performance on various MRC benchmarks. However, most of these benchmarks only evaluate models on in-domain test sets without considering their robustness under test-time perturbations. To fill this important gap, we construct AdvRACE (Adversarial RACE), a new model-agnostic benchmark for evaluating the robustness of MRC models under six different types of test-time perturbations, including our novel superimposed attack and distractor construction attack. We show that current state-of-the-art (SOTA) models are vulnerable to these simple black-box attacks. Our benchmark is constructed automatically based on the existing RACE benchmark, and thus the construction pipeline can be easily adopted by other tasks and datasets. We will release the data and source codes to facilitate future work. We hope that our work will encourage more research on improving the robustness of MRC and other NLU models.