We introduce BiMediX2, a bilingual (Arabic-English) Bio-Medical EXpert Large Multimodal Model that supports text-based and image-based medical interactions. It enables multi-turn conversation in Arabic and English and supports diverse medical imaging modalities, including radiology, CT, and histology. To train BiMediX2, we curate BiMed-V, an extensive Arabic-English bilingual healthcare dataset consisting of 1.6M samples of diverse medical interactions. This dataset supports a range of medical Large Language Model (LLM) and Large Multimodal Model (LMM) tasks, including multi-turn medical conversations, report generation, and visual question answering (VQA). We also introduce BiMed-MBench, the first Arabic-English medical LMM evaluation benchmark, verified by medical experts. BiMediX2 demonstrates excellent performance across multiple medical LLM and LMM benchmarks, achieving state-of-the-art results compared to other open-sourced models. On BiMed-MBench, BiMediX2 outperforms existing methods by over 9% in English and more than 20% in Arabic evaluations. Additionally, it surpasses GPT-4 by approximately 9% in UPHILL factual accuracy evaluations and excels in various medical VQA, report generation, and report summarization tasks. Our trained models, instruction set, and source code are available at - https://github.com/mbzuai-oryx/BiMediX2
论文检索
输入标题、作者或关键词，从 12,319 篇学术成果中精准定位
Understanding humor is a core aspect of social intelligence, yet it remains a significant challenge for Large Multimodal Models (LMMs). We introduce PixelHumor, a benchmark dataset of 2,800 annotated multi-panel comics designed to evaluate LMMs’ ability to interpret multimodal humor and recognize narrative sequences. Experiments with state-of-the-art LMMs reveal substantial gaps: for instance, top models achieve only 61% accuracy in panel sequencing, far below human performance. This underscores critical limitations in current models’ integration of visual and textual cues for coherent narrative and humor understanding. By providing a rigorous framework for evaluating multimodal contextual and narrative reasoning, PixelHumor aims to drive the development of LMMs that better engage in natural, socially aware interactions.
Zero-shot image captioning, which aims to generate image descriptions without relying on annotated data, has recently attracted increasing research interest. Pre-trained text-to-image generation models enable the creation of synthetic pairs solely from text data, while existing methods fall short in mitigating the discrepancy caused by the inability of synthetic images to fully capture the semantics of the textual input, resulting in unreliable cross-modal correspondences. To address this, we propose a retrieval-based framework that leverages only existing synthetic image-text pairs as its search corpus to systematically bridge the gap when using synthetic data for captioning. For the semantic gap between a synthetic image and its input text, our framework retrieves supplementary visual features from similar synthetic examples and integrates them to refine the image embedding. Then, it extracts image-related textual descriptions to mitigate the modality gap during decoding. Moreover, we introduce a plug-and-play visual semantic module that detects visual entities, further facilitating the construction of semantic correspondences between images and text. Experimental results on benchmark datasets demonstrate that our method obtains state-of-the-art results.
SKRAG: A Retrieval-Augmented Generation Framework Guided by Reasoning Skeletons over Knowledge Graphs
PDF ↗In specialized domains such as space science and utilization, question answering (QA) systems are required to perform complex multi-fact reasoning over sparse knowledge graphs (KGs). Existing KG-based retrieval-augmented generation (RAG) frameworks often face challenges such as inefficient subgraph retrieval, limited reasoning capabilities, and high computational costs. These issues limit their effectiveness in specialized domains. In this paper, we propose SKRAG, a novel Skeleton-guided RAG framework for knowledge graph question answering (KGQA). SKRAG leverages a lightweight language model enhanced with the Finite State Machine (FSM) constraint to produce structurally grounded reasoning skeletons, which guide accurate subgraph retrieval. The retrieved subgraph is then used to prompt a general large language model (LLM) for answer generation. We also introduce SSUQA, a KGQA dataset in the space science and utilization domain. Experiments show that SKRAG outperforms strong baselines on SSUQA and two general-domain benchmarks, demonstrating its adaptability and practical effectiveness.
Few-shot intent detection (FSID) targets the classification of user queries into in-scope intent categories or detecting them as out-of-scope, with only a few or even zero labeled examples per class. Existing PLM-based methods struggle in low-resource situations; while LLM-based methods face high inference cost and label interference. To harness their complementary strengths, we propose the \textbf{FCSLM}, a framework that collaborates a small prediction model with a large language model for the FSID task. During training, we leverage LLMs for data augmentation in self-supervised pretraining and supervised fine-tuning a task-specific prediction model. During inference, a multi-round reasoning process first applies the small prediction model to output candidate intents with uncertainty estimations, then invokes an LLM with enriched intent descriptions for refined prediction and OOS detection. Extensive experiments on three benchmark datasets demonstrate that our FCSLM outperforms strong competitors, achieving the new state-of-the-art performance in both intent classification and OOS detection. Our code is available at: https://github.com/hustchenpeng/FCSLM
Large Language Model (LLM) agents are transforming education by automating complex pedagogical tasks and enhancing both teaching and learning processes. In this survey, we present a systematic review of recent advances in applying LLM agents to address key challenges in educational settings, such as feedback comment generation, curriculum design, etc. We analyze the technologies enabling these agents, including representative datasets, benchmarks, and algorithmic frameworks. Additionally, we highlight key challenges in deploying LLM agents in educational settings, including ethical issues, hallucination and overreliance, and integration with existing educational ecosystems. Beyond the core technical focus, we include in Appendix A a comprehensive overview of domain-specific educational agents, covering areas such as science learning, language learning, and professional development.
Universal multimodal embedding models play a critical role in tasks such as interleaved image-text retrieval, multimodal RAG, and multimodal clustering. However, our empirical results indicate that existing LMM-based embedding models trained with the standard InfoNCE loss exhibit a high degree of overlap in similarity distribution between positive and negative pairs, making it challenging to distinguish hard negative pairs effectively. To deal with this issue, we propose a simple yet effective framework that dynamically improves the embedding model’s representation learning for negative pairs based on their discriminative difficulty. Within this framework, we train a series of models, named LLaVE, and evaluate them on the MMEB benchmark, which covers 4 meta-tasks and 36 datasets. Experimental results show that LLaVE establishes stronger baselines that achieve state-of-the-art (SOTA) performance while demonstrating strong scalability and efficiency. Specifically, LLaVE-2B surpasses the previous SOTA 7B models, while LLaVE-7B achieves a further performance improvement of 6.2 points. Although LLaVE is trained on image-text data, it can generalize to text-video retrieval tasks in a zero-shot manner and achieve strong performance, demonstrating its remarkable potential for transfer to other embedding tasks.
Recent advancements in Large Language Models (LLMs) have shown outstanding potential for role-playing applications. Evaluating these capabilities is becoming crucial yet remains challenging. Existing benchmarks mostly adopt a character-centric approach, simplify user-character interactions to isolated Q&A tasks, and fail to reflect real-world applications. To address this limitation, we introduce RMTBench, a comprehensive user-centric bilingual role-playing benchmark featuring 80 diverse characters and over 8,000 dialogue rounds. RMTBench includes custom characters with detailed backgrounds and abstract characters defined by simple traits, enabling evaluation across various user scenarios. Our benchmark constructs dialogues based on explicit user motivations rather than character descriptions, ensuring alignment with practical user applications. Furthermore, we construct an authentic multi-turn dialogue simulation mechanism. With carefully selected evaluation dimensions and LLM-based scoring, this mechanism captures the complex intention of conversations between the user and the character. By shifting focus from character background to user intention fulfillment, RMTBench bridges the gap between academic evaluation and practical deployment requirements, offering a more effective framework for assessing role-playing capabilities in LLMs. All code and datasets will be released soon.
Despite the rapid development of large language models (LLMs), existing benchmark datasets often focus on low-level cognitive tasks, such as factual recall and basic comprehension, while providing limited coverage of higher-level reasoning skills, including analysis, evaluation, and creation. In this work, we systematically assess the cognitive depth of popular LLM benchmarks using Bloom’s Taxonomy to evaluate both the cognitive and knowledge dimensions.Our analysis reveals a pronounced imbalance: most datasets concentrate on “Remembering” and “Understanding”, with metacognitive and creative reasoning largely underrepresented. We also find that incorporating higher-level cognitive instructions into the current instruction fine-tuning process improves model performance. These findings highlight the importance of future benchmarks incorporating metacognitive evaluations to more accurately assess and enhance model performance.
Permitted Knowledge Boundary: Evaluating the Knowledge-Constrained Responsiveness of Large Language Models
PDF ↗With the advancement of large language models (LLMs), recent research has raised concerns about their controllability.. In this paper, we argue for the importance of Knowledge-Constrained Responsiveness (KCR), ensuring that LLMs comply with human-defined constraints. However, KCR is an implicit and unobservable capability of LLMs, functioning as a black box that currently eludes quantitative assessment. To address this issue, we first introduce the definition of “permitted boundary” and define the “boundary bias” to depict KCR. We propose six metrics to quantify the boundary bias of LLMs and subsequently assess the KCR. Furthermore, we establish a benchmark with two new datasets, KCR-SimpleQA and KCR-WebNLG, to evaluate the performance of LLMs. Our extensive experiments show that several tested LLMs still struggle to varying degrees when adhering to constraints, especially without the corresponding knowledge.
2Columns1Row: A Russian Benchmark for Textual and Multimodal Table Understanding and Reasoning
PDF ↗Table understanding is a crucial task in document processing and is commonly encountered in practical applications. We introduce 2Columns1Row, the first open-source benchmark for the table question answering task in Russian. This benchmark evaluates the ability of models to reason about the relationships between rows and columns in tables, employing both textual and multimodal inputs. 2Columns1Row consists of six datasets, 28,800 tables, that vary in the complexity of the text within the table contents and the consistency of the values in the cells. We evaluate the models using text-only and multimodal approaches and analyze their performance. Through extensive evaluation, we demonstrate the limitations of current multimodal models on this task and prove the feasibility of a dynamic text-based system utilizing our benchmark. Our results highlight significant opportunities for advancing table understanding and reasoning, providing a solid foundation for future research in this domain.
Active Domain Knowledge Acquisition with 100-Dollar Budget: Enhancing LLMs via Cost-Efficient, Expert-Involved Interaction in Sensitive Domains
PDF ↗Large Language Models (LLMs) have demonstrated an impressive level of general knowledge. However, they often struggle in highly specialized and sensitive domains such as drug discovery and rare disease research due to the lack of expert knowledge, which is often costly to obtain. In this paper, we propose a novel framework (PU-ADKA) designed to efficiently enhance domain-specific LLMs by actively engaging domain experts within a fixed budget. Unlike traditional fine-tuning approaches, PU-ADKA proactively identifies and queries the most appropriate expert from a team, taking into account each expert’s availability, competency, knowledge boundaries, and consultation cost. We train PU-ADKA using simulations on PubMed publication data and validate it through domain expert interactions, showing promising improvements in LLM domain knowledge acquisition. Furthermore, our experiments with a real-world drug development team validate that PU-ADKA can significantly enhance LLM performance in specialized domains while adhering to strict budget constraints. In addition to outlining our methodological innovations and experimental results, we release a new benchmark dataset, CKAD, for cost-effective LLM domain knowledge acquisition to foster further research in this challenging area.
Marathi, the third most widely spoken language in India with over 83 million native speakers, remains significantly underrepresented in Natural Language Processing (NLP) research. While sentiment analysis has achieved substantial progress in high-resource languages such as English, Chinese, and Hindi, available Marathi datasets are limited to coarse sentiment labels and lack fine-grained emotional categorization or interpretability through explanations. To address this gap, we present a new annotated dataset of 10,762 Marathi sentences, each labeled with sentiment (positive, negative, or neutral), emotion (joy, anger, surprise, disgust, sadness, fear, or neutral), and a corresponding natural language justification. Justifications are written in English and generated using GPT-4 under a human-in-the-loop framework to ensure label fidelity and contextual alignment. Extensive experiments with both classical and transformer-based models demonstrate the effectiveness of the dataset for interpretable affective computing in a low-resource language setting, offering a benchmark for future research in multilingual and explainable NLP.
Modern language models are evaluated on large benchmarks, which are difficult to make sense of, especially for model selection.Looking at the raw evaluation numbers themselves using a model-centric lens, we propose SimBA, a three phase framework to Simplify Benchmark Analysis. The three phases of SimBA are: stalk, where we conduct dataset & model comparisons, prowl, where we discover a representative subset, and pounce, where we use the representative subset to predict performance on a held-out set of models. Applying SimBA to three popular LM benchmarks: HELM, MMLU, and BigBenchLite reveals that across all three benchmarks, datasets and models relate strongly to one another (stalk). We develop an representative set discovery algorithm which covers a benchmark using raw evaluation scores alone. Using our algorithm, we find that with 6.25% (1/16), 1.7% (1/58), and 28.4% (21/74) of the datasets for HELM, MMLU, and BigBenchLite respectively, we achieve coverage levels of at least 95% (prowl). Additionally, using just these representative subsets, we can both preserve model ranks and predict performance on a held-out set of models with near zero mean-squared error (pounce). Taken together, SimBA can help model developers improve efficiency during model training and dataset creators validate whether their newly created dataset differs from existing datasets in a benchmark. Our code is open source, available at https://github.com/nishantsubramani/simba.
Instruction tuning has underscored the significant potential of large language models (LLMs) in producing more human controllable and effective outputs in various domains. In this work, we focus on the data selection problem for task-specific instruction tuning of LLMs. Prevailing methods primarily rely on the crafted similarity metrics to select training data that aligns with the test data distribution. The goal is to minimize instruction tuning loss on the test data, ultimately improving performance on the target task. However, it has been widely observed that instruction tuning loss (i.e., cross-entropy loss for next token prediction) in LLMs often fails to exhibit a monotonic relationship with actual task performance. This misalignment undermines the effectiveness of current data selection methods for task-specific instruction tuning. To address this issue, we introduce ROSE, a novel Reward-Oriented inStruction data sElection method which leverages pairwise preference loss as a reward signal to optimize data selection for task-specific instruction tuning. Specifically, ROSE adapts an influence formulation to approximate the influence of training data points relative to a few-shot preference validation set to select the most task-related training data points. Experimental results show that by selecting just 5% of the training data using ROSE, our approach can achieve competitive results compared to fine-tuning with the full training dataset, and it surpasses other state-of-the-art data selection methods for task-specific instruction tuning. Our qualitative analysis further confirms the robust generalizability of our method across multiple benchmark datasets and diverse model architectures.
The rapid advancement of unsupervised representation learning and large-scale pre-trained vision-language models has significantly improved cross-modal retrieval tasks. However, existing multi-modal information retrieval (MMIR) studies lack a comprehensive exploration of document-level retrieval and suffer from the absence of cross-domain datasets at this granularity. To address this limitation, we introduce DocMMIR, a novel multi-modal document retrieval framework designed explicitly to unify diverse document formats and domains—including Wikipedia articles, scientific papers (arXiv), and presentation slides—within a comprehensive retrieval scenario. We construct a large-scale cross-domain multimodal dataset, comprising 450K training, 19.2K validation, and 19.2K test documents, serving as both a benchmark to reveal the shortcomings of existing MMIR models and a training set for further improvement. The dataset systematically integrates textual and visual information. Our comprehensive experimental analysis reveals substantial limitations in current state-of-the-art MLLMs (CLIP, BLIP2, SigLIP-2, ALIGN) when applied to our tasks, with only CLIP (ViT-L/14) demonstrating reasonable zero-shot performance. Through systematic investigation of cross-modal fusion strategies and loss function selection on the CLIP (ViT-L/14) model, we develop an optimised approach that achieves a +31% improvement in MRR@10 metrics from zero-shot baseline to fine-tuned model. Our findings offer crucial insights and practical guidance for future development in unified multimodal document retrieval tasks.
The existing assessments of planning capabilities of large language models (LLMs) remain largely limited to single-language or specific representation formats. To address this gap, we introduce the Multi-Plan benchmark comprising 204 multilingual and multi-format travel planning scenarios. In experimental results obtained with state-of-the-art LLMs, the Multi-Plan benchmark effectively highlights the performance disparities among models, notably showing superior results for reasoning-specialized models. Interestingly, language differences exhibited minimal impact, whereas mathematically structured representations significantly improved planning accuracy for most models, underscoring the crucial role of the input format. These findings enhance our understanding of planning abilities of LLMs, offer valuable insights for future research, and emphasize the need for more sophisticated AI evaluation methods. This dataset is publicly available at http://huggingface.co/datasets/Bllossom/Multi-Plan.
Training LLMs for Optimization Modeling via Iterative Data Synthesis and Structured Validation
PDF ↗Large Language Models (LLMs) have revolutionized various domains but encounter substantial challenges in tackling optimization modeling tasks for Operations Research (OR), particularly when dealing with complex problem. In this work, we propose Step-Opt-Instruct, a framework that augments existing datasets and generates high-quality fine-tuning data tailored to optimization modeling. Step-Opt-Instruct employs iterative problem generation to systematically increase problem complexity and stepwise validation to rigorously verify data, preventing error propagation and ensuring the quality of the generated dataset. Leveraging this framework, we fine-tune open-source LLMs, including LLaMA-3-8B and Mistral-7B, to develop Step-Opt—a model that achieves state-of-the-art performance on benchmarks such as NL4OPT, MAMO, and IndustryOR. Extensive experiments demonstrate the superior performance of Step-Opt, especially in addressing complex OR tasks, with a notable 17.01% improvement in micro average accuracy on difficult problems. These findings highlight the effectiveness of combining structured validation with gradual problem refinement to advance the automation of decision-making processes using LLMs. The code and dataset are available at https://github.com/samwu-learn/Step.
Two Challenges, One Solution: Robust Multimodal Learning through Dynamic Modality Recognition and Enhancement
PDF ↗Multimodal machine learning is often hindered by two critical challenges: modality missingness and modality imbalance. These challenges significantly degrade the performance of multimodal models. The majority of existing methods either require the availability of full-modality data during the training phase or necessitate explicit annotations to detect missing modalities. These dependencies severely limit the models’ applicability in the real world. To tackle these problems, we propose a Dynamic modality Recognition and Enhancement for Adaptive Multimodal fusion framework *DREAM*. Within DREAM, we innovatively employ a sample-level dynamic modality assessment mechanism to direct selective reconstruction of missing or underperforming modalities. Additionally, we introduce a soft masking fusion strategy that adaptively integrates different modalities according to their estimated contributions, enabling more accurate and robust predictions. Experimental results on three benchmark datasets consistently demonstrate that DREAM outperforms several representative baseline and state-of-the-art models, marking its robustness against modality missingness and imbalanced modality.
Over the years, online scams have grown dramatically,with nearly 50% of global consumersencountering scam attempts each week.These scams cause not only significant financiallosses to individuals and businesses, butalso lasting psychological trauma, largely dueto scammers’ strategic employment of psychologicaltechniques (PTs) to manipulate victims.Meanwhile, scammers continually evolve theirtactics by leveraging advances in Large LanguageModels (LLMs) to generate diverse scamvariants that easily bypass existing defenses.To address this pressing problem, we introducePsyScam, a benchmark designed to systematicallycapture the PTs employed in real-worldscam reports, and investigate how LLMs canbe utilized to generate variants of scams basedon the PTs and the contexts provided by thesescams. Specifically, we collect a wide range ofscam reports and ground its annotations of employedPTs in well-established cognitive andpsychological theories. We further demonstrateLLMs’ capabilities in generating through twodownstream tasks: scam completion, and scamaugmentation. Experimental results show thatPsyScam presents significant challenges toexisting models in both detecting and generatingscam content based on the PTs used byreal-world scammers. Our code and dataset areavailable.