Existing works on long-term open-domain dialogues focus on evaluating model responses within contexts spanning no more than five chat sessions. Despite advancements in long-context large language models (LLMs) and retrieval augmented generation (RAG) techniques, their efficacy in very long-term dialogues remains unexplored. To address this research gap, we introduce a machine-human pipeline to generate high-quality, very long-term dialogues by leveraging LLM-based agent architectures and grounding their dialogues on personas and temporal event graphs. Moreover, we equip each agent with the capability of sharing and reacting to images. The generated conversations are verified and edited by human annotators for long-range consistency and grounding to the event graphs. Using this pipeline, we collect LoCoMo, a dataset of very long-term conversations, each encompassing 600 turns and 16K tokens on avg., over up to 32 sessions. Based on LoCoMo, we present a comprehensive evaluation benchmark to measure long-term memory in models, encompassing question answering, event summarization, and multi-modal dialogue generation tasks. Our experimental results indicate that LLMs exhibit challenges in understanding lengthy conversations and comprehending long-range temporal and causal dynamics within dialogues. Employing strategies like long-context LLMs or RAG can offer improvements but these models still substantially lag behind human performance.
论文检索
输入标题、作者或关键词，从 1,444 篇学术成果中精准定位
LLM-Rubric: A Multidimensional, Calibrated Approach to Automated Evaluation of Natural Language Texts
PDF ↗This paper introduces a framework for the automated evaluation of natural language texts. A manually constructed rubric describes how to assess multiple dimensions of interest. To evaluate a text, a large language model (LLM) is prompted with each rubric question and produces a distribution over potential responses. The LLM predictions often fail to agree well with human judges—indeed, the humans do not fully agree with one another. However, the multiple LLM distributions can be _combined_ to _predict_ each human judge’s annotations on all questions, including a summary question that assesses overall quality or relevance. LLM-Rubric accomplishes this by training a small feed-forward neural network that includes both judge-specific and judge-independent parameters. When evaluating dialogue systems in a human-AI information-seeking task, we find that LLM-Rubric with 9 questions (assessing dimensions such as naturalness, conciseness, and citation quality) predicts human judges’ assessment of overall user satisfaction, on a scale of 1–4, with RMS error < 0.5, a 2× improvement over the uncalibrated baseline.
Understanding the reason for emotional support response is crucial for establishing connections between users and emotional support dialogue systems. Previous works mostly focus on generating better responses but ignore interpretability, which is extremely important for constructing reliable dialogue systems. To empower the system with better interpretability, we propose an emotional support response generation scheme, named \textbf{E}motion-Focused and \textbf{S}trategy-Driven \textbf{C}hain-\textbf{o}f-\textbf{T}hought (\textbf{ESCoT}), mimicking the process of \textit{identifying}, \textit{understanding}, and \textit{regulating} emotions. Specially, we construct a new dataset with ESCoT in two steps: (1) \textit{Dialogue Generation} where we first generate diverse conversation situations, then enhance dialogue generation using richer emotional support strategies based on these situations; (2) \textit{Chain Supplement} where we focus on supplementing selected dialogues with elements such as emotion, stimuli, appraisal, and strategy reason, forming the manually verified chains. Additionally, we further develop a model to generate dialogue responses with better interpretability. We also conduct extensive experiments and human evaluations to validate the effectiveness of the proposed ESCoT and generated dialogue responses. Our dataset, code, and model will be released.
The advancement of Artificial Intelligence is pivotal in reshaping healthcare, enhancing diagnostic precision, and facilitating personalized treatment strategies. One major challenge for healthcare professionals is quickly navigating through long clinical documents to provide timely and effective solutions. Doctors often struggle to draw quick conclusions from these extensive documents. To address this issue and save time for healthcare professionals, an effective summarization model is essential. Most current models assume the data is only text-based. However, patients often include images of their medical conditions in clinical documents. To effectively summarize these multimodal documents, we introduce EDI-Summ, an innovative Image-Guided Encoder-Decoder Model. This model uses modality-aware contextual attention on the encoder and an image cross-attention mechanism on the decoder, enhancing the BART base model to create detailed visual-guided summaries. We have tested our model extensively on three multimodal clinical benchmarks involving multimodal question and dialogue summarization tasks. Our analysis demonstrates that EDI-Summ outperforms state-of-the-art large language and vision-aware models in these summarization tasks. Disclaimer: The work includes vivid medical illustrations, depicting the essential aspects of the subject matter.
Conversation agents fueled by Large Language Models (LLMs) are providing a new way to interact with visual data. While there have been initial attempts for image-based conversation models, this work addresses the under-explored field of video-based conversation by introducing Video-ChatGPT. It is a multimodal model that merges a video-adapted visual encoder with an LLM. The resulting model is capable of understanding and generating detailed conversations about videos. We introduce a new dataset of 100,000 video-instruction pairs used to train Video-ChatGPT acquired via manual and semi-automated pipeline that is easily scalable and robust to label noise. We also develop a quantitative evaluation framework for video-based dialogue models to objectively analyze the strengths and weaknesses of video-based dialogue models. Code: https://github.com/mbzuai-oryx/Video-ChatGPT.
Analyzing LLM Behavior in Dialogue Summarization: Unveiling Circumstantial Hallucination Trends
PDF ↗Recent advancements in large language models (LLMs) have significantly advanced the capabilities of summarization systems.However, they continue to face a persistent challenge: hallucination. While prior work has extensively examined LLMs in news domains, evaluation of dialogue summarization has primarily focused on BART-based models, resulting in a notable gap in understanding LLM effectiveness.Our work seeks to address this gap by benchmarking LLMs for dialogue summarization faithfulness using human annotations,focusing on identifying and categorizing span-level inconsistencies.Specifically, we evaluate two prominent LLMs: GPT-4 and Alpaca-13B.Our evaluation reveals that LLMs often generate plausible, but not fully supported inferences based on conversation contextual cues, a trait absent in older models. As a result, we propose a refined taxonomy of errors, introducing a novel category termed “Contextual Inference” to address this aspect of LLM behavior. Using our taxonomy, we compare the behavioral differences between LLMs and older fine-tuned models. Additionally, we systematically assess the efficacy of automatic error detection methods on LLM summaries and find that they struggle to detect these nuanced errors effectively. To address this, we introduce two prompt-based approaches for fine-grained error detection. Our methods outperform existing metrics, particularly in identifying the novel “Contextual Inference” error type.
VisDiaHalBench: A Visual Dialogue Benchmark For Diagnosing Hallucination in Large Vision-Language Models
PDF ↗Despite the significant success of large vision-language models (LVLMs), some studies have revealed that LVLMs suffer from the hallucination problem, where the LVLMs’ response contains descriptions of non-existent objects. Although various benchmarks have been proposed to investigate this problem, they mostly focus on single-turn evaluation and overlook the hallucination raised by textual inputs. To investigate the hallucination problem of LVLMs when given long-term misleading textual history, we propose a novel visual dialogue hallucination evaluation benchmark VisDiaHalBench. The benchmark consists of samples with five-turn questions about an edited image and its original version. VisDiaHalBench differs from previous hallucination benchmarks in the following three points: 1) The questions and answers are unambiguously grounded by annotated scene graphs. 2) The images are uncommonly edited to inspect the visual model and common-object hallucination in LLMs. 3) The carefully designed dialogue refers a same object in different turns to assess the image consistency and influence of history for LVLMs. The detailed analysis of several state-of-the-art LVLMs across image consistency, visual understanding, history influence, and other dimensions reveals their substantial performance gap with single-turn VQA tasks. The benchmark is released in: https://github.com/qingxingcao/VisDiaHalBench
Recently, the advent of large language models (LLMs) has revolutionized generative agents. Among them, Role-Playing Conversational Agents (RPCAs) attract considerable attention due to their ability to emotionally engage users. However, the absence of a comprehensive benchmark impedes progress in this field. To bridge this gap, we introduce CharacterEval, a Chinese benchmark for comprehensive RPCA assessment, complemented by a tailored high-quality dataset. The dataset comprises 1,785 multi-turn role-playing dialogues, encompassing 11,376 examples and featuring 77 characters derived from Chinese novels and scripts. It was carefully constructed, beginning with initial dialogue extraction via GPT-4, followed by rigorous human-led quality control, and enhanced with in-depth character profiles sourced from Baidu Baike. CharacterEval employs a multifaceted evaluation approach, encompassing thirteen targeted metrics on four dimensions. To facilitate the convenient evaluation for these subjective metrics in CharacterEval, we further developed CharacterRM, a role-playing reward model based on human annotations, which has a higher correlation with human judgment compared to GPT-4. Comprehensive experiments on CharacterEval demonstrate that Chinese LLMs exhibit more promising capabilities than GPT-4 in Chinese role-playing conversation.
Large Language Models (LLMs) have shown strong generalization abilities to excel in various tasks, including emotion support conversations. However, deploying such LLMs like GPT-3 (175B parameters) is resource-intensive and challenging at scale. In this study, we utilize LLMs as “Counseling Teacher” to enhance smaller models’ emotion support response abilities, significantly reducing the necessity of scaling up model size. To this end, we first introduce an iterative expansion framework, aiming to prompt the large teacher model to curate an expansive emotion support dialogue dataset. This curated dataset, termed ExTES, encompasses a broad spectrum of scenarios and is crafted with meticulous strategies to ensure its quality and comprehensiveness. Based on this, we then devise a Diverse Response Inpainting (DRI) mechanism to harness the teacher model to produce multiple diverse responses by filling in the masked conversation context. This richness and variety serve as instructive examples, providing a robust foundation for fine-tuning smaller student models. Experiments across varied scenarios reveal that the teacher-student scheme with DRI notably improves the response abilities of smaller models, even outperforming the teacher model in some cases. The dataset and codes are available in https://github.com/pandazzh2020/ExTES.
Chat Vector: A Simple Approach to Equip LLMs with Instruction Following and Model Alignment in New Languages
PDF ↗Recently, the development of open-source large language models (LLMs) has advanced rapidly. Nevertheless, due to data constraints, the capabilities of most open-source LLMs are primarily focused on English. To address this issue, we introduce the concept of \textit{chat vector} to equip pre-trained language models with instruction following and human value alignment via simple model arithmetic. The chat vector is derived by subtracting the weights of a pre-trained base model (e.g. LLaMA2) from those of its corresponding chat model (e.g. LLaMA2-chat). By simply adding the chat vector to a continual pre-trained model’s weights, we can endow the model with chat capabilities in new languages without the need for further training.Our empirical studies demonstrate the superior efficacy of the chat vector from three different aspects: instruction following, toxicity mitigation, and multi-turn dialogue. Moreover, to showcase the adaptability of our approach, we extend our experiments to encompass various languages, base models, and chat vectors. The results underscore the chat vector’s simplicity, effectiveness, and wide applicability, making it a compelling solution for efficiently enabling conversational capabilities in pre-trained language models. Our code is available at https://github.com/aqweteddy/ChatVector.
Dialogue State Tracking (DST) is designed to monitor the evolving dialogue state in the conversations and plays a pivotal role in developing task-oriented dialogue systems. However, obtaining the annotated data for the DST task is usually a costly endeavor. In this paper, we focus on employing LLMs to generate dialogue data to reduce dialogue collection and annotation costs. Specifically, GPT-4 is used to simulate the user and agent interaction, generating thousands of dialogues annotated with DST labels. Then a two-stage fine-tuning on LLaMA 2 is performed on the generated data and the real data for the DST prediction. Experimental results on two public DST benchmarks show that with the generated dialogue data, our model performs better than the baseline trained solely on real data. In addition, our approach is also capable of adapting to the dynamic demands in real-world scenarios, generating dialogues in new domains swiftly. After replacing dialogue segments in any domain with the corresponding generated ones, the model achieves comparable performance to the model trained on real data. The source code and generated dialogue data are available at https://github.com/ParticleMedia/LUAS.
Large language models (LLMs) are increasingly prevalent in conversational systems due to their advanced understanding and generative capabilities in general contexts. However, their effectiveness in task-oriented dialogues (TOD), which requires not only response generation but also effective dialogue state tracking (DST) within specific tasks and domains, remains less satisfying. In this work, we propose a novel approach FnCTOD for solving DST with LLMs through function calling. This method improves zero-shot DST, allowing adaptation to diverse domains without extensive data collection or model tuning. Our experimental results demonstrate that our approach achieves exceptional performance with both modestly sized open-source and also proprietary LLMs: with in-context prompting it enables various 7B or 13B parameter models to surpass the previous state-of-the-art (SOTA) achieved by ChatGPT, and improves ChatGPT’s performance beating the SOTA by 5.6% average joint goal accuracy (JGA). Individual model results for GPT-3.5 and GPT-4 are boosted by 4.8% and 14%, respectively. We also show that by fine-tuning on a small collection of diverse task-oriented dialogues, we can equip modestly sized models, specifically a 13B parameter LLaMA2-Chat model, with function-calling capabilities and DST performance comparable to ChatGPT while maintaining their chat capabilities. We have made the code publicly available at https://github.com/facebookresearch/FnCTOD.
In this paper we address the task of summarizing television shows, which touches key areas in AI research: complex reasoning, multiple modalities, and long narratives. We present a modular approach where separate components perform specialized sub-tasks which we argue affords greater flexibility compared to end-to-end methods. Our modules involve detecting scene boundaries, reordering scenes so as to minimize the number of cuts between different events, converting visual information to text, summarizing the dialogue in each scene, and fusing the scene summaries into a final summary for the entire episode. We also present a new metric, PRISMA (**P**recision and **R**ecall Evaluat**i**on of **s**ummary F**a**cts), to measure both precision and recall of generated summaries, which we decompose into atomic facts. Tested on the recently released SummScreen3D dataset (Papalampidi & Lapata, 2023), our method produces higher quality summaries than comparison models, as measured with ROUGE and our new fact-based metric.
The unparalleled performance of closed-sourced ChatGPT has sparked efforts towards its democratization, with notable strides made by leveraging real user and ChatGPT dialogues, as evidenced by Vicuna. However, due to challenges in gathering dialogues involving human participation, current endeavors like Baize and UltraChat rely on ChatGPT conducting roleplay to simulate humans based on instructions, resulting in overdependence on seeds, diminished human-likeness, limited topic diversity, and an absence of genuine multi-round conversational dynamics. To address the above issues, we propose a paradigm to simulate human behavior better and explore the benefits of incorporating more human-like questions in multi-turn conversations. Specifically, we directly target human questions extracted from genuine human-machine conversations as a learning goal and provide a novel user simulator called ‘Socratic‘. The experimental results show our response model, ‘PlatoLM‘, achieves SoTA performance among LLaMA-based 7B models in MT-Bench. Our findings further demonstrate that our method introduces highly human-like questioning patterns and rich topic structures, which can teach the response model better than previous works in multi-round conversations.
Large Language Models are Superpositions of All Characters: Attaining Arbitrary Role-play via Self-Alignment
PDF ↗Considerable efforts have been invested in augmenting the role-playing proficiency of open-source large language models (LLMs) by emulating proprietary counterparts. Nevertheless, we posit that LLMs inherently harbor role-play capabilities, owing to the extensive knowledge of characters and potential dialogues ingrained in their vast training corpora. Thus, we introduce Ditto, the first self-alignment method for role-play, which encourages an instruction-following LLM to simulate role-play dialogues as a variant of reading comprehension, and creates a role-play training set comprising 4000 characters, surpassing the scale of currently available datasets by tenfold regarding the number of roles. Subsequently, we fine-tune the LLM using this self-generated dataset to augment its role-playing capabilities. Upon evaluating our meticulously constructed role-play benchmark and the roleplay subset of MT-Bench, Ditto, in various parameter scales, consistently maintains a consistent role identity and provides accurate role-specific knowledge in multi-turn role-play conversations, outperforming all open-source role-play baselines. Furthermore, we present the first cross-supervision role-play experiment, revealing that the role-play styles can be easily acquired, while the intrinsic capabilities of LLMs confine the knowledge within role-play.
Stickers, while widely recognized for enhancing empathetic communication in online interactions, remain underexplored in current empathetic dialogue research, notably due to the challenge of a lack of comprehensive datasets. In this paper, we introduce the Agent for STICKERCONV (Agent4SC), which uses collaborative agent interactions to realistically simulate human behavior with sticker usage, thereby enhancing multimodal empathetic communication. Building on this foundation, we develop a multimodal empathetic dialogue dataset, STICKERCONV, comprising 12.9K dialogue sessions, 5.8K unique stickers, and 2K diverse conversational scenarios. This dataset serves as a benchmark for multimodal empathetic generation. To advance further, we propose PErceive and Generate Stickers (PEGS), a multimodal empathetic response generation framework, complemented by a comprehensive set of empathy evaluation metrics based on LLM. Our experiments demonstrate PEGS’s effectiveness in generating contextually relevant and emotionally resonant multimodal empathetic responses, contributing to the advancement of more nuanced and engaging empathetic dialogue systems.
Entity Alignment (EA) is vital for integrating diverse knowledge graph (KG) data, playing a crucial role in data-driven AI applications. Traditional EA methods primarily rely on comparing entity embeddings, but their effectiveness is constrained by the limited input KG data and the capabilities of the representation learning techniques. Against this backdrop, we introduce ChatEA, an innovative framework that incorporates large language models (LLMs) to improve EA. To address the constraints of limited input KG data, ChatEA introduces a KG-code translation module that translates KG structures into a format understandable by LLMs, thereby allowing LLMs to utilize their extensive background knowledge to improve EA accuracy. To overcome the over-reliance on entity embedding comparisons, ChatEA implements a two-stage EA strategy that capitalizes on LLMs’ capability for multi-step reasoning in a dialogue format, thereby enhancing accuracy while preserving efficiency. Our experimental results affirm ChatEA’s superior performance, highlighting LLMs’ potential in facilitating EA tasks.The source code is available at https://anonymous.4open.science/r/ChatEA/.
MT-Bench-101: A Fine-Grained Benchmark for Evaluating Large Language Models in Multi-Turn Dialogues
PDF ↗The advent of Large Language Models (LLMs) has drastically enhanced dialogue systems. However, comprehensively evaluating the dialogue abilities of LLMs remains a challenge. Previous benchmarks have primarily focused on single-turn dialogues or provided coarse-grained and incomplete assessments of multi-turn dialogues, overlooking the complexity and fine-grained nuances of real-life dialogues. To address this issue, we introduce MT-Bench-101, specifically designed to evaluate the fine-grained abilities of LLMs in multi-turn dialogues. By conducting a detailed analysis of real multi-turn dialogue data, we construct a three-tier hierarchical ability taxonomy comprising 4208 turns across 1388 multi-turn dialogues in 13 distinct tasks. We then evaluate 21 popular LLMs based on MT-Bench-101, conducting comprehensive analyses from both ability and task perspectives and observing differing trends in LLMs performance across dialogue turns within various tasks. Further analysis indicates that neither utilizing common alignment techniques nor chat-specific designs has led to obvious enhancements in the multi-turn abilities of LLMs. Extensive case studies suggest that our designed tasks accurately assess the corresponding multi-turn abilities. The data and code are available at https://github.com/mtbench101/mt-bench-101.
We introduce Semantic Parsing in Contextual Environments (SPICE), a task designed to enhance artificial agents’ contextual awareness by integrating multimodal inputs with prior contexts. SPICE goes beyond traditional semantic parsing by offering a structured, interpretable framework for dynamically updating an agent’s knowledge with new information, mirroring the complexity of human communication. We develop the VG-SPICE dataset, crafted to challenge agents with visual scene graph construction from spoken conversational exchanges, highlighting speech and visual data integration. We also present the Audio-Vision Dialogue Scene Parser (AViD-SP) developed for use on VG-SPICE. These innovations aim to improve multimodal information processing and integration. Both the VG-SPICE dataset and the AViD-SP model are publicly available.
ChiMed-GPT: A Chinese Medical Large Language Model with Full Training Regime and Better Alignment to Human Preferences
PDF ↗Recently, the increasing demand for superior medical services has highlighted the discrepancies in the medical infrastructure. With big data, especially texts, forming the foundation of medical services, there is an exigent need for effective natural language processing (NLP) solutions tailored to the healthcare domain. Conventional approaches leveraging pre-trained models present promising results in this domain and current large language models (LLMs) offer advanced foundation for medical text processing. However, most medical LLMs are trained only with supervised fine-tuning (SFT), even though it efficiently empowers LLMs to understand and respond to medical instructions but is ineffective in learning domain knowledge and aligning with human preference. In this work, we propose ChiMed-GPT, a new benchmark LLM designed explicitly for Chinese medical domain, and undergoes a comprehensive training regime with pre-training, SFT, and RLHF. Evaluations on tasks including information extraction, question answering, and dialogue generation demonstrate ChiMed-GPT’s superior performance over general domain LLMs. Furthermore, we analyze possible biases through prompting ChiMed-GPT to perform attitude scales regarding discrimination of patients, so as to contribute to further responsible development of LLMs in the medical domain.