Navigation instruction generation for visually impaired (VI) individuals (NIG-VI) is critical yet relatively underexplored. This study, hence, focuses on producing precise, in-situ, step-by-step navigation instructions that are practically usable by VI users. Concretely, we propose LaF-GRPO (LLM-as-Follower GRPO), where an LLM simulates VI user responses to generate rewards guiding the Vision-Language Model (VLM) post-training. This enhances instruction usability while reducing costly real-world data needs. To facilitate training and testing, we introduce NIG4VI, a 27k-sample open-sourced benchmark. It provides diverse navigation scenarios with accurate spatial coordinates, supporting detailed, open-ended in-situ instruction generation. Experiments on NIG4VI show the effectiveness of LaF-GRPO by quantitative metrics (e.g., Zero-(LaF-GRPO) boosts BLEU +14\%; SFT+(LaF-GRPO) METEOR 0.542 vs. GPT-4o's 0.323) and yields more intuitive, safer instructions. Code and benchmark are available at \href{https://github.com/YiyiyiZhao/NIG4VI}{https://github.com/YiyiyiZhao/NIG4VI}.
AI for Industrial Asset Lifecycle Management aims to automate complex operational workflows -- such as condition monitoring, maintenance planning, and intervention scheduling -- to reduce human workload and minimize system downtime. Traditional AI/ML approaches have primarily tackled these problems in isolation, solving narrow tasks within the broader operational pipeline. In contrast, the emergence of AI agents and large language models (LLMs) introduces a next-generation opportunity: enabling end-to-end automation across the entire asset lifecycle. This paper envisions a future where AI agents autonomously manage tasks that previously required distinct expertise and manual coordination. To this end, we introduce AssetOpsBench -- a unified framework and environment designed to guide the development, orchestration, and evaluation of domain-specific agents tailored for Industry 4.0 applications. We outline the key requirements for such holistic systems and provide actionable insights into building agents that integrate perception, reasoning, and control for real-world industrial operations. The software is available at https://github.com/IBM/AssetOpsBench.
Multimodal foundation models have demonstrated impressive capabilities across diverse tasks. However, their potential as plug-and-play solutions for missing modality prediction remains underexplored. To investigate this, we categorize existing approaches into three representative paradigms, encompassing a total of 42 model variants, and conduct a comprehensive evaluation in terms of prediction accuracy and adaptability to downstream tasks. Our analysis reveals that current foundation models often fall short in two critical aspects: (i) fine-grained semantic extraction from the available modalities, and (ii) robust validation of generated modalities. These limitations lead to suboptimal and, at times, misaligned predictions. To address these challenges, we propose an agentic framework tailored for missing modality prediction. This framework dynamically formulates modality-aware mining strategies based on the input context, facilitating the extraction of richer and more discriminative semantic features. In addition, we introduce a \textit{self-refinement mechanism}, which iteratively verifies and enhances the quality of generated modalities through internal feedback. Experimental results show that our method reduces FID for missing image prediction by at least 14% and MER for missing text prediction by at least 10% compared to baselines.
In this work, we introduce the task of singing voice deepfake source attribution (SVDSA). We hypothesize that multimodal foundation models (MMFMs) such as ImageBind, LanguageBind will be most effective for SVDSA as they are better equipped for capturing subtle source-specific characteristics-such as unique timbre, pitch manipulation, or synthesis artifacts of each singing voice deepfake source due to their cross-modality pre-training. Our experiments with MMFMs, speech foundation models and music foundation models verify the hypothesis that MMFMs are the most effective for SVDSA. Furthermore, inspired from related research, we also explore fusion of foundation models (FMs) for improved SVDSA. To this end, we propose a novel framework, COFFE which employs Chernoff Distance as novel loss function for effective fusion of FMs. Through COFFE with the symphony of MMFMs, we attain the topmost performance in comparison to all the individual FMs and baseline fusion methods.
This paper presents ThinkTank, a comprehensive and scalable framework designed to transform specialized AI agent systems into versatile collaborative intelligence platforms capable of supporting complex problem-solving across diverse domains. ThinkTank systematically generalizes agent roles, meeting structures, and knowledge integration mechanisms by adapting proven scientific collaboration methodologies. Through role abstraction, generalization of meeting types for iterative collaboration, and the integration of Retrieval-Augmented Generation with advanced knowledge storage, the framework facilitates expertise creation and robust knowledge sharing. ThinkTank enables organizations to leverage collaborative AI for knowledge-intensive tasks while ensuring data privacy and security through local deployment, utilizing frameworks like Ollama with models such as Llama3.1. The ThinkTank framework is designed to deliver significant advantages in cost-effectiveness, data security, scalability, and competitive positioning compared to cloud-based alternatives, establishing it as a universal platform for AI-driven collaborative problem-solving. The ThinkTank code is available at https://github.com/taugroup/ThinkTank
3D Gaussian Splatting and its extension to 4D dynamic scenes enable photorealistic, real-time rendering from real-world captures, positioning Gaussian Splats (GS) as a promising format for next-generation immersive media. However, their high storage requirements pose significant challenges for practical use in sharing, transmission, and storage. Despite various studies exploring GS compression from different perspectives, these efforts remain scattered across separate repositories, complicating benchmarking and the integration of best practices. To address this gap, we present GSCodec Studio, a unified and modular framework for GS reconstruction, compression, and rendering. The framework incorporates a diverse set of 3D/4D GS reconstruction methods and GS compression techniques as modular components, facilitating flexible combinations and comprehensive comparisons. By integrating best practices from community research and our own explorations, GSCodec Studio supports the development of compact representation and compression solutions for static and dynamic Gaussian Splats, namely our Static and Dynamic GSCodec, achieving competitive rate-distortion performance in static and dynamic GS compression. The code for our framework is publicly available at https://github.com/JasonLSC/GSCodec_Studio , to advance the research on Gaussian Splats compression.
Understanding the interaction between different drugs (drug-drug interaction or DDI) is critical for ensuring patient safety and optimizing therapeutic outcomes. Existing DDI datasets primarily focus on textual information, overlooking multimodal data that reflect complex drug mechanisms. In this paper, we (1) introduce MUDI, a large-scale Multimodal biomedical dataset for Understanding pharmacodynamic Drug-drug Interactions, and (2) benchmark learning methods to study it. In brief, MUDI provides a comprehensive multimodal representation of drugs by combining pharmacological text, chemical formulas, molecular structure graphs, and images across 310,532 annotated drug pairs labeled as Synergism, Antagonism, or New Effect. Crucially, to effectively evaluate machine-learning based generalization, MUDI consists of unseen drug pairs in the test set. We evaluate benchmark models using both late fusion voting and intermediate fusion strategies. All data, annotations, evaluation scripts, and baselines are released under an open research license.
We introduce Sorrel (https://github.com/social-ai-uoft/sorrel), a simple Python interface for generating and testing new multi-agent reinforcement learning environments. This interface places a high degree of emphasis on simplicity and accessibility, and uses a more psychologically intuitive structure for the basic agent-environment loop, making it a useful tool for social scientists to investigate how learning and social interaction leads to the development and change of group dynamics. In this short paper, we outline the basic design philosophy and features of Sorrel.
We propose a method that enables large language models (LLMs) to control embodied agents by directly mapping continuous observation vectors to continuous action vectors. Initially, the LLMs generate a control strategy based on a textual description of the agent, its environment, and the intended goal. This strategy is then iteratively refined through a learning process in which the LLMs are repeatedly prompted to improve the current strategy, using performance feedback and sensory-motor data collected during its evaluation. The method is validated on classic control tasks from the Gymnasium library and the inverted pendulum task from the MuJoCo library. In most cases, it successfully identifies optimal or high-performing solutions by integrating symbolic knowledge derived through reasoning with sub-symbolic sensory-motor data gathered as the agent interacts with its environment.
Logical reasoning is a core capability for many applications of large language models (LLMs), yet existing benchmarks often rely solely on final-answer accuracy, failing to capture the quality and structure of the reasoning process. We propose FineLogic, a fine-grained evaluation framework that assesses logical reasoning across three dimensions: overall benchmark accuracy, stepwise soundness, and representation-level alignment. In addition, to better understand how reasoning capabilities emerge, we conduct a comprehensive study on the effects of supervision format during fine-tuning. We construct four supervision styles (one natural language and three symbolic variants) and train LLMs under each. Our findings reveal that natural language supervision yields strong generalization even on out-of-distribution and long-context tasks, while symbolic reasoning styles promote more structurally sound and atomic inference chains. Further, our representation-level probing shows that fine-tuning primarily improves reasoning behaviors through step-by-step generation, rather than enhancing shortcut prediction or internalized correctness. Together, our framework and analysis provide a more rigorous and interpretable lens for evaluating and improving logical reasoning in LLMs.
Visuals are valuable tools for teaching math word problems (MWPs), helping young learners interpret textual descriptions into mathematical expressions before solving them. However, creating such visuals is labor-intensive and there is a lack of automated methods to support this process. In this paper, we present Math2Visual, an automatic framework for generating pedagogically meaningful visuals from MWP text descriptions. Math2Visual leverages a pre-defined visual language and a design space grounded in interviews with math teachers, to illustrate the core mathematical relationships in MWPs. Using Math2Visual, we construct an annotated dataset of 1,903 visuals and evaluate Text-to-Image (TTI) models for their ability to generate visuals that align with our design. We further fine-tune several TTI models with our dataset, demonstrating improvements in educational visual generation. Our work establishes a new benchmark for automated generation of pedagogically meaningful visuals and offers insights into key challenges in producing multimodal educational content, such as the misrepresentation of mathematical relationships and the omission of essential visual elements.
Recent studies comparing AI-generated and human-authored literary texts have produced conflicting results: some suggest AI already surpasses human quality, while others argue it still falls short. We start from the hypothesis that such divergences can be largely explained by genuine differences in how readers interpret and value literature, rather than by an intrinsic quality of the texts evaluated. Using five public datasets (1,471 stories, 101 annotators including critics, students, and lay readers), we (i) extract 17 reference-less textual features (e.g., coherence, emotional variance, average sentence length...); (ii) model individual reader preferences, deriving feature importance vectors that reflect their textual priorities; and (iii) analyze these vectors in a shared "preference space". Reader vectors cluster into two profiles: 'surface-focused readers' (mainly non-experts), who prioritize readability and textual richness; and 'holistic readers' (mainly experts), who value thematic development, rhetorical variety, and sentiment dynamics. Our results quantitatively explain how measurements of literary quality are a function of how text features align with each reader's preferences. These findings advocate for reader-sensitive evaluation frameworks in the field of creative text generation.
We introduce the attention-indexed model (AIM), a theoretical framework for analyzing learning in deep attention layers. Inspired by multi-index models, AIM captures how token-level outputs emerge from layered bilinear interactions over high-dimensional embeddings. Unlike prior tractable attention models, AIM allows full-width key and query matrices, aligning more closely with practical transformers. Using tools from statistical mechanics and random matrix theory, we derive closed-form predictions for Bayes-optimal generalization error and identify sharp phase transitions as a function of sample complexity, model width, and sequence length. We propose a matching approximate message passing algorithm and show that gradient descent can reach optimal performance. AIM offers a solvable playground for understanding learning in modern attention architectures.
Directly training Large Language Models (LLMs) for Multi-Agent Systems (MAS) remains challenging due to intricate reward modeling, dynamic agent interactions, and demanding generalization requirements. This paper explores whether post-training techniques, specifically Supervised Fine-Tuning (SFT) and Reinforcement Learning with Verifiable Rewards (RLVR), can effectively $\textit{generalize}$ to multi-agent scenarios. We use economic reasoning as a testbed, leveraging its strong foundations in mathematics and game theory, its demand for structured analytical reasoning, and its relevance to real-world applications such as market design, resource allocation, and policy analysis. We introduce $\textbf{Recon}$ ($\textbf{R}$easoning like an $\textbf{ECON}$omist), a 7B-parameter open-source LLM post-trained on a hand-curated dataset of 2,100 high-quality economic reasoning problems. Comprehensive evaluation on economic reasoning benchmarks and multi-agent games reveals clear improvements in structured reasoning and economic rationality. These results underscore the promise of domain-aligned post-training for enhancing reasoning and agent alignment, shedding light on the roles of SFT and RL in shaping model behavior. Code is available at https://github.com/MasterZhou1/Recon .
Dysarthria, a motor speech disorder, affects intelligibility and requires targeted interventions for effective communication. In this work, we investigate automated mispronunciation feedback by collecting a dysarthric speech dataset from six speakers reading two passages, annotated by a speech therapist with temporal markers and mispronunciation descriptions. We design a three-stage framework for explainable mispronunciation evaluation: (1) overall clarity scoring, (2) mispronunciation localization, and (3) mispronunciation type classification. We systematically analyze pretrained Automatic Speech Recognition (ASR) models in each stage, assessing their effectiveness in dysarthric speech evaluation (Code available at: https://github.com/augmented-human-lab/interspeech25_speechtherapy, Supplementary webpage: https://apps.ahlab.org/interspeech25_speechtherapy/). Our findings offer clinically relevant insights for automating actionable feedback for pronunciation assessment, which could enable independent practice for patients and help therapists deliver more effective interventions.
Communication between agents often constitutes a major computational bottleneck in distributed learning. One of the most common mitigation strategies is to compress the information exchanged, thereby reducing communication overhead. To counteract the degradation in convergence associated with compressed communication, error feedback schemes -- most notably $\mathrm{EF}$ and $\mathrm{EF}^{21}$ -- were introduced. In this work, we provide a tight analysis of both of these methods. Specifically, we find the Lyapunov function that yields the best possible convergence rate for each method -- with matching lower bounds. This principled approach yields sharp performance guarantees and enables a rigorous, apples-to-apples comparison between $\mathrm{EF}$, $\mathrm{EF}^{21}$, and compressed gradient descent. Our analysis is carried out in a simplified yet representative setting, which allows for clean theoretical insights and fair comparison of the underlying mechanisms.
We develop a theoretical model that addresses the economic trade-off between cost per token versus serial token generation speed when deploying LLMs for inference at scale. Our model takes into account arithmetic, memory bandwidth, network bandwidth and latency constraints; and optimizes over different parallelism setups and batch sizes to find the ones that optimize serial inference speed at a given cost per token. We use the model to compute Pareto frontiers of serial speed versus cost per token for popular language models.
Quantified Integer Programming (QIP) bridges multiple domains by extending Quantified Boolean Formulas (QBF) to incorporate general integer variables and linear constraints while also generalizing Integer Programming through variable quantification. As a special case of Quantified Constraint Satisfaction Problems (QCSP), QIP provides a versatile framework for addressing complex decision-making scenarios. Additionally, the inclusion of a linear objective function enables QIP to effectively model multistage robust discrete linear optimization problems, making it a powerful tool for tackling uncertainty in optimization. While two primary solution paradigms exist for QBF -- search-based and expansion-based approaches -- only search-based methods have been explored for QIP and QCSP. We introduce an expansion-based approach for QIP using Counterexample-Guided Abstraction Refinement (CEGAR), adapting techniques from QBF. We extend this methodology to tackle multistage robust discrete optimization problems with linear constraints and further embed it in an optimization framework, enhancing its applicability. Our experimental results highlight the advantages of this approach, demonstrating superior performance over existing search-based solvers for QIP in specific instances. Furthermore, the ability to model problems using linear constraints enables notable performance gains over state-of-the-art expansion-based solvers for QBF.
We propose a novel diffusion-based framework for automatic colorization of Anime-style facial sketches. Our method preserves the structural fidelity of the input sketch while effectively transferring stylistic attributes from a reference image. Unlike traditional approaches that rely on predefined noise schedules - which often compromise perceptual consistency -- our framework builds on continuous-time diffusion models and introduces SSIMBaD (Sigma Scaling with SSIM-Guided Balanced Diffusion). SSIMBaD applies a sigma-space transformation that aligns perceptual degradation, as measured by structural similarity (SSIM), in a linear manner. This scaling ensures uniform visual difficulty across timesteps, enabling more balanced and faithful reconstructions. Experiments on a large-scale Anime face dataset demonstrate that our method outperforms state-of-the-art models in both pixel accuracy and perceptual quality, while generalizing to diverse styles. Code is available at github.com/Giventicket/SSIMBaD-Sigma-Scaling-with-SSIM-Guided-Balanced-Diffusion-for-AnimeFace-Colorization
As AI evolves, collaboration among heterogeneous models helps overcome data scarcity by enabling knowledge transfer across institutions and devices. Traditional Federated Learning (FL) only supports homogeneous models, limiting collaboration among clients with heterogeneous model architectures. To address this, Heterogeneous Federated Learning (HtFL) methods are developed to enable collaboration across diverse heterogeneous models while tackling the data heterogeneity issue at the same time. However, a comprehensive benchmark for standardized evaluation and analysis of the rapidly growing HtFL methods is lacking. Firstly, the highly varied datasets, model heterogeneity scenarios, and different method implementations become hurdles to making easy and fair comparisons among HtFL methods. Secondly, the effectiveness and robustness of HtFL methods are under-explored in various scenarios, such as the medical domain and sensor signal modality. To fill this gap, we introduce the first Heterogeneous Federated Learning Library (HtFLlib), an easy-to-use and extensible framework that integrates multiple datasets and model heterogeneity scenarios, offering a robust benchmark for research and practical applications. Specifically, HtFLlib integrates (1) 12 datasets spanning various domains, modalities, and data heterogeneity scenarios; (2) 40 model architectures, ranging from small to large, across three modalities; (3) a modularized and easy-to-extend HtFL codebase with implementations of 10 representative HtFL methods; and (4) systematic evaluations in terms of accuracy, convergence, computation costs, and communication costs. We emphasize the advantages and potential of state-of-the-art HtFL methods and hope that HtFLlib will catalyze advancing HtFL research and enable its broader applications. The code is released at https://github.com/TsingZ0/HtFLlib.