F2RVLM: Boosting Fine-grained Fragment Retrieval for Multi-Modal Long-form Dialogue with Vision Language Model

Traditional dialogue retrieval aims to select the most appropriate utterance or image from recent dialogue history. However, they often fail to meet users' actual needs for revisiting semantically coherent content scattered across long-form conversations. To fill this gap, we define the Fine-grained Fragment Retrieval (FFR) task, requiring models to locate query-relevant fragments, comprising both utterances and images, from multimodal long-form dialogues. As a foundation for FFR, we construct MLDR, the longest-turn multimodal dialogue retrieval dataset to date, averaging 25.45 turns per dialogue, with each naturally spanning three distinct topics. To evaluate generalization in real-world scenarios, we curate and annotate a WeChat-based test set comprising real-world multimodal dialogues with an average of 75.38 turns. Building on these resources, we explore existing generation-based Vision-Language Models (VLMs) on FFR and observe that they often retrieve incoherent utterance-image fragments. While optimized for generating responses from visual-textual inputs, these models lack explicit supervision to ensure semantic coherence within retrieved fragments. To this end, we propose F2RVLM, a generative retrieval model trained in a two-stage paradigm: (1) supervised fine-tuning to inject fragment-level retrieval knowledge, and (2) GRPO-based reinforcement learning with multi-objective rewards promoting semantic precision, relevance, and contextual coherence. To handle varying intra-fragment complexity, from locally dense to sparsely distributed, we introduce difficulty-aware curriculum sampling that ranks training instances by model-predicted difficulty and gradually exposes the model to harder samples. This boosts reasoning ability in long, multi-turn contexts. F2RVLM outperforms popular VLMs in both in-domain and real-domain settings, demonstrating superior retrieval performance.

MEENA (PersianMMMU): Multimodal-Multilingual Educational Exams for N-level Assessment

Recent advancements in large vision-language models (VLMs) have primarily focused on English, with limited attention given to other languages. To address this gap, we introduce MEENA (also known as PersianMMMU), the first dataset designed to evaluate Persian VLMs across scientific, reasoning, and human-level understanding tasks. Our dataset comprises approximately 7,500 Persian and 3,000 English questions, covering a wide range of topics such as reasoning, mathematics, physics, diagrams, charts, and Persian art and literature. Key features of MEENA include: (1) diverse subject coverage spanning various educational levels, from primary to upper secondary school, (2) rich metadata, including difficulty levels and descriptive answers, (3) original Persian data that preserves cultural nuances, (4) a bilingual structure to assess cross-linguistic performance, and (5) a series of diverse experiments assessing various capabilities, including overall performance, the model's ability to attend to images, and its tendency to generate hallucinations. We hope this benchmark contributes to enhancing VLM capabilities beyond English.

Why Synthetic Isn't Real Yet: A Diagnostic Framework for Contact Center Dialogue Generation

Synthetic transcript generation is critical in contact center domains, where privacy and data scarcity limit model training and evaluation. Unlike prior synthetic dialogue generation work on open-domain or medical dialogues, contact center conversations are goal-oriented, role-asymmetric, and behaviorally complex, featuring disfluencies, ASR noise, and compliance-driven agent actions. In deployments where transcripts are unavailable, standard pipelines still yield derived call attributes such as Intent Summaries, Topic Flow, and QA Evaluation Forms. We leverage these as supervision signals to guide generation. To assess the quality of such outputs, we introduce a diagnostic framework of 18 linguistically and behaviorally grounded metrics for comparing real and synthetic transcripts. We benchmark four language-agnostic generation strategies, from simple prompting to characteristic-aware multi-stage approaches, alongside reference-free baselines. Results reveal persistent challenges: no method excels across all traits, with notable deficits in disfluency, sentiment, and behavioral realism. Our diagnostic tool exposes these gaps, enabling fine-grained evaluation and stress testing of synthetic dialogue across languages.

Federal Reserve Communication and the COVID-19 Pandemic

In this study, we examine the Federal Reserve's communication strategies during the COVID-19 pandemic, comparing them with communication during previous periods of economic stress. Using specialized dictionaries tailored to COVID-19, unconventional monetary policy (UMP), and financial stability, combined with sentiment analysis and topic modeling techniques, we identify a distinct focus in Fed communication during the pandemic on financial stability, market volatility, social welfare, and UMP, characterized by notable contextual uncertainty. Through comparative analysis, we juxtapose the Fed's communication during the COVID-19 crisis with its responses during the dot-com and global financial crises, examining content, sentiment, and timing dimensions. Our findings reveal that Fed communication and policy actions were more reactive to the COVID-19 crisis than to previous crises. Additionally, declining sentiment related to financial stability in interest rate announcements and minutes anticipated subsequent accommodative monetary policy decisions. We further document that communicating about UMP has become the "new normal" for the Fed's Federal Open Market Committee meeting minutes and Chairman's speeches since the Global Financial Crisis, reflecting an institutional adaptation in communication strategy following periods of economic distress. These findings contribute to our understanding of how central bank communication evolves during crises and how communication strategies adapt to exceptional economic circumstances.

Diverse And Private Synthetic Datasets Generation for RAG evaluation: A multi-agent framework

Retrieval-augmented generation (RAG) systems improve large language model outputs by incorporating external knowledge, enabling more informed and context-aware responses. However, the effectiveness and trustworthiness of these systems critically depends on how they are evaluated, particularly on whether the evaluation process captures real-world constraints like protecting sensitive information. While current evaluation efforts for RAG systems have primarily focused on the development of performance metrics, far less attention has been given to the design and quality of the underlying evaluation datasets, despite their pivotal role in enabling meaningful, reliable assessments. In this work, we introduce a novel multi-agent framework for generating synthetic QA datasets for RAG evaluation that prioritize semantic diversity and privacy preservation. Our approach involves: (1) a Diversity agent leveraging clustering techniques to maximize topical coverage and semantic variability, (2) a Privacy Agent that detects and mask sensitive information across multiple domains and (3) a QA curation agent that synthesizes private and diverse QA pairs suitable as ground truth for RAG evaluation. Extensive experiments demonstrate that our evaluation sets outperform baseline methods in diversity and achieve robust privacy masking on domain-specific datasets. This work offers a practical and ethically aligned pathway toward safer, more comprehensive RAG system evaluation, laying the foundation for future enhancements aligned with evolving AI regulations and compliance standards.

Long-Term Variability in Physiological-Arousal Relationships for Robust Emotion Estimation

Estimating emotional states from physiological signals is a central topic in affective computing and psychophysiology. While many emotion estimation systems implicitly assume a stable relationship between physiological features and subjective affect, this assumption has rarely been tested over long timeframes. This study investigates whether such relationships remain consistent across several months within individuals. We developed a custom measurement system and constructed a longitudinal dataset by collecting physiological signals -- including blood volume pulse, electrodermal activity (EDA), skin temperature, and acceleration--along with self-reported emotional states from 24 participants over two three-month periods. Data were collected in naturalistic working environments, allowing analysis of the relationship between physiological features and subjective arousal in everyday contexts. We examined how physiological-arousal relationships evolve over time by using Explainable Boosting Machines (EBMs) to ensure model interpretability. A model trained on 1st-period data showed a 5\% decrease in accuracy when tested on 2nd-period data, indicating long-term variability in physiological-arousal associations. EBM-based comparisons further revealed that while heart rate remained a relatively stable predictor, minimum EDA exhibited substantial individual-level fluctuations between periods. While the number of participants is limited, these findings highlight the need to account for temporal variability in physiological-arousal relationships and suggest that emotion estimation models should be periodically updated -- e.g., every five months -- based on observed shift trends to maintain robust performance over time.

Combining Language and Topic Models for Hierarchical Text Classification

Hierarchical text classification (HTC) is a natural language processing task which has the objective of categorising text documents into a set of classes from a predefined structured class hierarchy. Recent HTC approaches use various techniques to incorporate the hierarchical class structure information with the natural language understanding capabilities of pre-trained language models (PLMs) to improve classification performance. Furthermore, using topic models along with PLMs to extract features from text documents has been shown to be an effective approach for multi-label text classification tasks. The rationale behind the combination of these feature extractor models is that the PLM captures the finer-grained contextual and semantic information while the topic model obtains high-level representations which consider the corpus of documents as a whole. In this paper, we use a HTC approach which uses a PLM and a topic model to extract features from text documents which are used to train a classification model. Our objective is to determine whether the combination of the features extracted from the two models is beneficial to HTC performance in general. In our approach, the extracted features are passed through separate convolutional layers whose outputs are combined and passed to a label-wise attention mechanisms which obtains label-specific document representations by weighing the most important features for each class separately. We perform comprehensive experiments on three HTC benchmark datasets and show that using the features extracted from the topic model generally decreases classification performance compared to only using the features obtained by the PLM. In contrast to previous work, this shows that the incorporation of features extracted from topic models for text classification tasks should not be assumed beneficial.

Taking the Garbage Out of Data-Driven Prediction Across Climate Timescales

Artificial intelligence (AI) -- and specifically machine learning (ML) -- applications for climate prediction across timescales are proliferating quickly. The emergence of these methods prompts a revisit to the impact of data preprocessing, a topic familiar to the climate community, as more traditional statistical models work with relatively small sample sizes. Indeed, the skill and confidence in the forecasts produced by data-driven models are directly influenced by the quality of the datasets and how they are treated during model development, thus yielding the colloquialism "garbage in, garbage out." As such, this article establishes protocols for the proper preprocessing of input data for AI/ML models designed for climate prediction (i.e., subseasonal to decadal and longer). The three aims are to: (1) educate researchers, developers, and end users on the effects that preprocessing has on climate predictions; (2) provide recommended practices for data preprocessing for such applications; and (3) empower end users to decipher whether the models they are using are properly designed for their objectives. Specific topics covered in this article include the creation of (standardized) anomalies, dealing with non-stationarity and the spatiotemporally correlated nature of climate data, and handling of extreme values and variables with potentially complex distributions. Case studies will illustrate how using different preprocessing techniques can produce different predictions from the same model, which can create confusion and decrease confidence in the overall process. Ultimately, implementing the recommended practices set forth in this article will enhance the robustness and transparency of AI/ML in climate prediction studies.

Snap-Snap: Taking Two Images to Reconstruct 3D Human Gaussians in Milliseconds

Reconstructing 3D human bodies from sparse views has been an appealing topic, which is crucial to broader the related applications. In this paper, we propose a quite challenging but valuable task to reconstruct the human body from only two images, i.e., the front and back view, which can largely lower the barrier for users to create their own 3D digital humans. The main challenges lie in the difficulty of building 3D consistency and recovering missing information from the highly sparse input. We redesign a geometry reconstruction model based on foundation reconstruction models to predict consistent point clouds even input images have scarce overlaps with extensive human data training. Furthermore, an enhancement algorithm is applied to supplement the missing color information, and then the complete human point clouds with colors can be obtained, which are directly transformed into 3D Gaussians for better rendering quality. Experiments show that our method can reconstruct the entire human in 190 ms on a single NVIDIA RTX 4090, with two images at a resolution of 1024x1024, demonstrating state-of-the-art performance on the THuman2.0 and cross-domain datasets. Additionally, our method can complete human reconstruction even with images captured by low-cost mobile devices, reducing the requirements for data collection. Demos and code are available at https://hustvl.github.io/Snap-Snap/.

A Hybrid AI Methodology for Generating Ontologies of Research Topics from Scientific Paper Corpora

Taxonomies and ontologies of research topics (e.g., MeSH, UMLS, CSO, NLM) play a central role in providing the primary framework through which intelligent systems can explore and interpret the literature. However, these resources have traditionally been manually curated, a process that is time-consuming, prone to obsolescence, and limited in granularity. This paper presents Sci-OG, a semi-auto\-mated methodology for generating research topic ontologies, employing a multi-step approach: 1) Topic Discovery, extracting potential topics from research papers; 2) Relationship Classification, determining semantic relationships between topic pairs; and 3) Ontology Construction, refining and organizing topics into a structured ontology. The relationship classification component, which constitutes the core of the system, integrates an encoder-based language model with features describing topic occurrence in the scientific literature. We evaluate this approach against a range of alternative solutions using a dataset of 21,649 manually annotated semantic triples. Our method achieves the highest F1 score (0.951), surpassing various competing approaches, including a fine-tuned SciBERT model and several LLM baselines, such as the fine-tuned GPT4-mini. Our work is corroborated by a use case which illustrates the practical application of our system to extend the CSO ontology in the area of cybersecurity. The presented solution is designed to improve the accessibility, organization, and analysis of scientific knowledge, thereby supporting advancements in AI-enabled literature management and research exploration.