The curse of multilinguality phenomenon is a fundamental problem of multilingual Large Language Models (LLMs), where the competition between massive languages results in inferior performance. It mainly comes from limited capacity and negative transfer between dissimilar languages. To address this issue, we propose a method to dynamically group and scale up the parameters of multilingual LLM while boosting positive transfer among similar languages. Specifically, the model is first tuned on monolingual corpus to determine the parameter deviation in each layer and quantify the similarity between languages. Layers with more deviations are extended to mixture-of-experts layers to reduce competition between languages, where one expert module serves one group of similar languages. Experimental results on 18 to 128 languages show that our method reduces the negative transfer between languages and significantly boosts multilingual performance with fewer parameters. Such language group specialization on experts benefits the new language adaptation and reduces the inference on the previous multilingual knowledge learned.
论文检索
输入标题、作者或关键词，从 3,086 篇学术成果中精准定位
Retrieval-augmented methods have achieved remarkable advancements in alleviating the hallucination of large language models.Nevertheless, the introduction of external knowledge does not always lead to the expected improvement in model performance, as irrelevant or harmful information present in the retrieved knowledge can compromise the prediction process.To address these challenges, we propose a novel framework aimed at improving model performance by incorporating knowledge filtering and prediction fusion mechanisms.In particular, our approach first employs a perplexity-based annotation method to collect training data.Then, we design four distinct strategies to filter out harmful retrieved knowledge.Finally, we integrate the filtered knowledge to generate the final result via batch-wise predictions.We conduct extensive experiments across multiple discriminative task datasets to evaluate the proposed framework.The results demonstrate that our framework can significantly enhance the performance of models on discriminative tasks.
Handling unanswerable questions (UAQ) is crucial for LLMs, as it helps prevent misleading responses in complex situations. While previous studies have built several datasets to assess LLMs’ performance on UAQ, these datasets lack factual knowledge support, which limits the evaluation of LLMs’ ability to utilize their factual knowledge when handling UAQ. To address the limitation, we introduce a new unanswerable question dataset UAQFact, a bilingual dataset with auxiliary factual knowledge created from a Knowledge Graph. Based on UAQFact, we further define two new tasks to measure LLMs’ ability to utilize internal and external factual knowledge, respectively. Our experimental results across multiple LLM series show that UAQFact presents significant challenges, as LLMs do not consistently perform well even when they have factual knowledge stored. Additionally, we find that incorporating external knowledge may enhance performance, but LLMs still cannot make full use of the knowledge which may result in incorrect responses. Our code and dataset are available at https://github.com/cytan17726/UAQ_Fact.
Grounding the reasoning ability of large language models (LLMs) for embodied tasks is challenging due to the complexity of the physical world. Especially, LLM planning for multi-agent collaboration requires communication of agents or credit assignment as the feedback to re-adjust the proposed plans and achieve effective coordination. However, existing methods that overly rely on physical verification or self-reflection suffer from excessive and inefficient querying of LLMs. In this paper, we propose a novel framework for multi-agent collaboration that introduces Reinforced Advantage feedback (ReAd) for efficient self-refinement of plans. Specifically, we perform critic regression to learn a sequential advantage function from LLM-planned data, and then treat the LLM planner as an optimizer to generate actions that maximize the advantage function. It endows the LLM with the foresight to discern whether the action contributes to accomplishing the final task. We provide theoretical analysis by extending advantage-weighted regression in reinforcement learning to multi-agent systems. Experiments on Overcooked-AI and a difficult variant of RoCoBench show that ReAd surpasses baselines in success rate, and also significantly decreases the interaction steps of agents and query rounds of LLMs, demonstrating its high efficiency for grounding LLMs. More results are given at https://read-llm.github.io/.
Despite achieving remarkable performance, machine translation (MT) research remains underexplored in terms of translating cultural elements in languages, such as idioms, proverbs, and colloquial expressions. This paper investigates the capability of state-of-the-art neural machine translation (NMT) and large language models (LLMs) in translating proverbs, which are deeply rooted in cultural contexts. We construct a translation dataset of standalone proverbs and proverbs in conversation for four language pairs. Our experiments show that the studied models can achieve good translation between languages with similar cultural backgrounds, and LLMs generally outperform NMT models in proverb translation. Furthermore, we find that current automatic evaluation metrics such as BLEU, CHRF++ and COMET are inadequate for reliably assessing the quality of proverb translation, highlighting the need for more culturally aware evaluation metrics.
Creative story generation has long been a goal of NLP research. While existing methodologies have aimed to generate long and coherent stories, they fall significantly short of human capabilities in terms of diversity and character depth. To address this, we introduce a novel story generation framework called CCI (Character-centric Creative story generation via Imagination). CCI features two modules for creative story generation: IG (Image-Guided Imagination) and MW (Multi-Writer model). In the IG module, we utilize a text-to-image model to create visual representations of key story elements, such as characters, backgrounds, and main plots, in a more novel and concrete manner than text-only approaches. The MW module uses these story elements to generate multiple persona-description candidates and selects the best one to insert into the story, thereby enhancing the richness and depth of the narrative. We compared the stories generated by CCI and baseline models through statistical analysis, as well as human and LLM evaluations. The results showed that the IG and MW modules significantly improve various aspects of the stories’ creativity. Furthermore, our framework enables interactive multi-modal story generation with users, opening up new possibilities for human-LLM integration in cultural development. Project page : https://www.2024cci.p-e.kr/
The increasing prevalence of large language models (LLMs) such as GPT-4 in various applications has led to a surge in the size of prompts required for optimal performance, leading to challenges in computational efficiency. Prompt compression aims to reduce the inference cost by minimizing input tokens without compromising on the task performance. However, existing prompt compression techniques either rely on sub-optimal metrics such as information entropy or model it as a task-agnostic token classification problem that fails to capture task-specific information.To address these issues, we propose a novel and efficient reinforcement learning (RL) based task-aware prompt compression method. To ensure low latency requirements, we leverage existing Transformer encoder-based token classification model while guiding the learning process with task-specific reward signals using lightweight REINFORCE algorithm. We evaluate the performance of our method on three diverse and challenging tasks including text summarization, question answering and code summarization. We demonstrate that our RL-guided compression method improves the task performance by 8% - 189% across these three scenarios over state-of-the-art compression techniques while satisfying the same compression rate and latency requirements.
LLM-based Multi-agent frameworks have shown a great potential in solving real-world software development tasks, where the agents of different roles can communicate much more efficiently than humans. Despite their efficiency, LLM-based agents can hardly fully understand each other, which frequently causes errors during the development process. Moreover, the accumulation of errors could easily lead to the failure of the whole project. In order to reduce such errors, we introduce an intention aligned multi-agent framework RTADev, which utilizes a self-correction mechanism to ensure that all agents work based on a consensus. RTADev mimics human teams where individuals are free to start meetings anytime for reaching agreement. Specifically, RTADev integrates an alignment checking phase and a conditional ad hoc group review phase, so that the errors can be effectively reduced with minimum agent communications. Our experiments on various software development tasks show that RTADev significantly improves the quality of generated software code in terms of executability, structural and functional completeness. The code of our project is available at https://github.com/codeagent-rl/RTADev.
Anomaly detection (AD) is an important machine learning task with many real-world uses, including fraud detection, medical diagnosis, and industrial monitoring. Within natural language processing (NLP), AD helps detect issues like spam, misinformation, and unusual user activity. Although large language models (LLMs) have had a strong impact on tasks such as text generation and summarization, their potential in AD has not been studied enough. This paper introduces AD-LLM, the first benchmark that evaluates how LLMs can help with NLP anomaly detection. We examine three key tasks: (i) zero-shot detection, using LLMs’ pre-trained knowledge to perform AD without tasks-specific training; (ii) data augmentation, generating synthetic data and category descriptions to improve AD models; and (iii) model selection, using LLMs to suggest unsupervised AD models. Through experiments with different datasets, we find that LLMs can work well in zero-shot AD, that carefully designed augmentation methods are useful, and that explaining model selection for specific datasets remains challenging. Based on these results, we outline six future research directions on LLMs for AD.
The structural properties of naturally arising social graphs are extensively studied to understand their evolution. Prior approaches for modeling network dynamics typically rely on rule-based models, which lack realism and generalizability, or deep learning-based models, which require large-scale training datasets. As abstract graph representations of entity-wise interactions, social graphs present an opportunity to explore network evolution mechanisms through realistic simulations of human-item interactions. Leveraging the pre-trained social consensus knowledge embedded in large language models (LLMs), we present GraphAgent-Generator (GAG), a novel simulation-based framework for dynamic, text-attributed social graph generation. GAG simulates the temporal node and edge generation processes for zero-shot social graph generation. The resulting graphs adhere to seven key macroscopic network properties, achieving an 11% improvement in microscopic graph structure metrics. Through the node classification benchmarking task, we validate that GAG effectively captures the intricate text-structure correlations in graph generation. Furthermore, GAG supports generating graphs with up to nearly 100,000 nodes or 10 million edges through large-scale LLM-based agent simulation with parallel acceleration, achieving a minimum speed-up of 90.4%. The source code is available at https://github.com/Ji-Cather/GraphAgent.
Large Language Models (LLMs) have demonstrated notable capabilities across various tasks, showcasing complex problem-solving abilities. Understanding and executing complex rules, along with multi-step planning, are fundamental to logical reasoning and critical for practical LLM agents and decision-making systems. However, evaluating LLMs as effective rule-based executors and planners remains underexplored. In this paper, we introduce LogicGame, a novel benchmark designed to evaluate the comprehensive rule understanding, execution, and planning capabilities of LLMs. Unlike traditional benchmarks, LogicGame provides diverse games that contain a series of rules with an initial state, requiring models to comprehend and apply predefined regulations to solve problems. We create simulated scenarios in which models execute or plan operations to achieve specific outcomes. These game scenarios are specifically designed to distinguish logical reasoning from mere knowledge by relying exclusively on predefined rules. This separation allows for a pure assessment of rule-based reasoning capabilities. The evaluation considers not only final outcomes but also intermediate steps, providing a comprehensive assessment of model performance. Moreover, these intermediate steps are deterministic and can be automatically verified. LogicGame defines game scenarios with varying difficulty levels, from simple rule applications to complex reasoning chains, in order to offer a precise evaluation of model performance on rule understanding and multi-step execution. Utilizing LogicGame, we test various LLMs and identify notable shortcomings in their rule-based logical reasoning abilities.
Most existing GUI agents typically depend on non-vision inputs like HTML source code or accessibility trees, limiting flexibility across diverse software environments and platforms. Current multimodal large language models (MLLMs), though excel at using vision to ground real-world objects, often struggle with accurately localizing GUI elements – a critical requirement for effective GUI automation – due to the semantic gap between real-world objects and GUI elements. In this work, we introduce Ponder & Press, a divide-and-conquer framework for general computer control that uses only visual input. Our approach combines a general-purpose MLLM as an ‘interpreter’, responsible for translating high-level user instructions into detailed action descriptions, with a GUI-specific MLLM as a ‘locator’ that precisely locates GUI elements for action placement. By leveraging a purely visual input, our agent offers a versatile, human-like interaction paradigm applicable to various applications. Ponder & Press locator outperforms existing models by +22.5% on the ScreenSpot GUI grounding benchmark. More offline and interactive agent benchmarks across various GUI environments – including web pages, desktop software, and mobile UIs – demonstrate that the Ponder & Press framework achieves state-of-the-art performance, highlighting the potential of visual GUI agents.
Leveraging Unit Language Guidance to Advance Speech Modeling in Textless Speech-to-Speech Translation
PDF ↗The success of building textless speech-to-speech translation (S2ST) models has attracted much attention. However, S2ST still faces two main challenges: 1) extracting linguistic features for various speech signals, called cross-modal (CM), and 2) learning alignment of difference languages in long sequences, called cross-lingual (CL). We propose the unit language to overcome the two modeling challenges. The unit language can be considered a text-like representation format, constructed using n-gram language modeling. We implement multi-task learning to utilize the unit language in guiding the speech modeling process. Our initial results reveal a conflict when applying source and target unit languages simultaneously. We propose task prompt modeling to mitigate this conflict. We conduct experiments on four languages of the Voxpupil dataset. Our method demonstrates significant improvements over a strong baseline and achieves performance comparable to models trained with text.
LLMs have become prevalent tools for information seeking across various fields, including law. However, their generated responses often suffer from hallucinations, hindering their widespread adoption in high stakes domains such as law, which can potentially mislead experts and propagate societal harms. To enhance trustworthiness in these systems, one promising approach is to attribute the answer to an actual source, thereby improving the factuality and verifiability of the response. In pursuit of advancing attributed legal question answering, we introduce AQuAECHR, a benchmark comprising information-seeking questions from ECHR jurisprudence along with attributions to relevant judgments. We present strategies to automatically curate this dataset from ECHR case law guides and utilize an LLM-based filtering pipeline to improve dataset quality, as validated by legal experts. Additionally, we assess several LLMs, including those trained on legal corpora, on this dataset to underscore significant challenges with the current models and strategies dealing with attributed QA, both quantitatively and qualitatively.
Multimodal Large Language Models (MLLMs) have gained increasing popularity as a promising framework for leveraging the strong language reasoning capabilities in the vision-language domain. Given a wide range of MLLMs, model merging potentially offers a cheap way to aggregate their diverse knowledge into a single MLLM. However, directly plug-in existing model merging approaches often leads to suboptimal performance due to (1) inclusion of harmful models that have over-confident predictions in the target task; (2) the lack of specialized designs for vision-language inputs. To tackle these pain points, we conduct pioneering investigations to dissect the merging procedures and propose an uncertainty-guided MLLM merging algorithm, \textit{i.e.}, \texttt{UQ-Merge}, which i) identifies beneficial candidates for merging, ii) determines the merging order and the number of helpful candidates, and iii) performs appropriate merging. Within our framework, we consider uncertainty quantification on both text and vision inputs to examine the MLLM prediction confidence, and then decide whether and when a MLLM needs to be included. It is worth mentioning that our vision-language uncertainty quantification does not require access to sample labels, making it more practical in various scenarios. Extensive experiments consistently demonstrate the superior MLLM merging performance of \texttt{UQ-Merge} in both held-in and held-out vision-language benchmarks. For example, compared to existing state-of-the-art merging methods, \texttt{UQ-Merge} brings substantial performance improvements of up to 44.3% on average accuracy in 12 datasets. Codes are available at https://anonymous.4open.science/r/UQ-Merge-7CD7.
Large Language Models (LLMs) often struggle with tasks requiring external knowledge, such as knowledge-intensive Multiple Choice Question Answering (MCQA). Integrating Knowledge Graphs (KGs) can enhance reasoning; however, existing methods typically demand costly fine-tuning or retrieve noisy KG information. Recent approaches leverage Graph Neural Networks (GNNs) to generate KG-based input embedding prefixes as soft prompts for LLMs but fail to account for question relevance, resulting in noisy prompts. Moreover, in MCQA tasks, the absence of relevant KG knowledge for certain answer options remains a significant challenge. To address these issues, we propose Question-Aware Knowledge Graph Prompting (QAP), which incorporates question embeddings into GNN aggregation to dynamically assess KG relevance. QAP employs global attention to capture inter-option relationships, enriching soft prompts with inferred knowledge. Experimental results demonstrate that QAP outperforms state-of-the-art methods across multiple datasets, highlighting its effectiveness.
Data-Centric Improvements for Enhancing Multi-Modal Understanding in Spoken Conversation Modeling
PDF ↗Conversational assistants are increasingly popular across diverse real-world applications, highlighting the need for advanced multimodal speech modeling. Speech, as a natural mode of communication, encodes rich user-specific characteristics such as speaking rate and pitch, making it critical for effective interaction. Our work introduces a data-centric customization approach for efficiently enhancing multimodal understanding in conversational speech modeling. Central to our contributions is a novel multi-task learning paradigm that involves designing auxiliary tasks to utilize a small amount of speech data. Our approach achieves state-of-the-art performance on the Spoken-SQuAD benchmark, using only 10% of the training data with open-weight models, establishing a robust and efficient framework for audio-centric conversational modeling. We also introduce ASK-QA, the first dataset for multi-turn spoken dialogue with ambiguous user requests and dynamic evaluation inputs.
In the era of Large Language Models (LLMs), establishing effective evaluation methods and standards for diverse human-AI interaction systems is increasingly challenging. To encourage more transparent documentation and facilitate discussion on human-AI system evaluation design options, we present an evaluation card SPHERE, which encompasses five key dimensions: 1) What is being evaluated?; 2) How is the evaluation conducted?; 3) Who is participating in the evaluation?; 4) When is evaluation conducted?; 5) How is evaluation validated? We conduct a review of 39 human-AI systems using SPHERE, outlining current evaluation practices and areas for improvement. We provide three recommendations for improving the validity and rigor of evaluation practices.
In recent years, the rapid advancement of large language models (LLMs) has significantly reshaped the landscape of scientific research. While LLMs have achieved notable success across various domains, their application in specialized fields such as lunar exploration remains underdeveloped, and their full potential in this domain has yet to be fully realized. To address this gap, we introduce Lunar Twins, the first LLMs designed specifically for lunar exploration, along with a collaborative framework that combines both large and small models. Additionally, we present Lunar GenData, a multi-agent collaborative workflow for generating lunar instructions, and establish the first specialized lunar dataset, which integrates real data from the Chang’e lunar missions. Lastly, we developed Lunar Eval, the first comprehensive evaluation suite for assessing the capabilities of LLMs in lunar exploration tasks. Experimental validation demonstrates that our approach not only enhances domain expertise in lunar exploration but also reveals preliminary indications of embodied intelligence potential.
With the rapid development of Large Language Models (LLMs), Parameter-Efficient Fine-Tuning (PEFT) methods have gained significant attention, which aims to achieve efficient fine-tuning of LLMs with fewer parameters. As a representative PEFT method, Low-Rank Adaptation (LoRA) introduces low-rank matrices to approximate the incremental tuning parameters and achieves impressive performance over multiple scenarios. After that, plenty of improvements have been proposed for further improvement. However, these methods either focus on single-task scenarios or separately train multiple LoRA modules for multi-task scenarios, limiting the efficiency and effectiveness of LoRA in multi-task scenarios. To better adapt to multi-task fine-tuning, in this paper, we propose a novel Mixture of Low-Rank Experts (MoRE) for multi-task PEFT. Specifically, instead of using an individual LoRA for each task, we align different ranks of LoRA module with different tasks, which we named low-rank experts. Moreover, we design a novel adaptive rank selector to select the appropriate expert for each task. By jointly training low-rank experts, MoRE can enhance the adaptability and efficiency of LoRA in multi-task scenarios. Finally, we conduct extensive experiments over multiple multi-task benchmarks along with different LLMs to verify model performance. Experimental results demonstrate that compared to traditional LoRA and its variants, MoRE significantly improves the performance of LLMs in multi-task scenarios and incurs no additional inference cost. We also release the model and code to facilitate the community.