Ismini Lourentzou

Gaurang Karwande, Amarachi Mbakawe, Joy T. Wu et al. · ibm-research

Despite the progress in utilizing deep learning to automate chest radiograph interpretation and disease diagnosis tasks, change between sequential Chest X-rays (CXRs) has received limited attention. Monitoring the progression of pathologies that are visualized through chest imaging poses several challenges in anatomical motion estimation and image registration, i.e., spatially aligning the two images and modeling temporal dynamics in change detection. In this work, we propose CheXRelNet, a neural model that can track longitudinal pathology change relations between two CXRs. CheXRelNet incorporates local and global visual features, utilizes inter-image and intra-image anatomical information, and learns dependencies between anatomical region attributes, to accurately predict disease change for a pair of CXRs. Experimental results on the Chest ImaGenome dataset show increased downstream performance compared to baselines. Code is available at https://github.com/PLAN-Lab/ChexRelNet

Md Mahim Anjum Haque, Wasi Uddin Ahmad, Ismini Lourentzou et al.

The complexity of modern software has led to a drastic increase in the time and cost associated with detecting and rectifying software bugs. In response, researchers have explored various methods to automatically generate fixes for buggy code. However, due to the large combinatorial space of possible fixes for any given bug, few tools and datasets are available to evaluate model-generated fixes effectively. To address this issue, we introduce FixEval, a benchmark comprising of buggy code submissions to competitive programming problems and their corresponding fixes. FixEval offers an extensive collection of unit tests to evaluate the correctness of model-generated program fixes and assess further information regarding time, memory constraints, and acceptance based on a verdict. We consider two Transformer language models pretrained on programming languages as our baseline and compare them using match-based and execution-based evaluation metrics. Our experiments show that match-based metrics do not reflect model-generated program fixes accurately. At the same time, execution-based methods evaluate programs through all cases and scenarios designed explicitly for that solution. Therefore, we believe FixEval provides a step towards real-world automatic bug fixing and model-generated code evaluation. The dataset and models are open-sourced at https://github.com/mahimanzum/FixEval.

Boyi Li, Yifan Shen, Yuanzhe Liu et al.

Multimodal Large Language Models (MLLMs) have achieved remarkable success in open-vocabulary perceptual tasks, yet their ability to solve complex cognitive problems remains limited, especially when visual details are abstract and require visual memory. Current approaches primarily scale Chain-of-Thought (CoT) reasoning in the text space, even when language alone is insufficient for clear and structured reasoning, and largely neglect visual reasoning mechanisms analogous to the human visuospatial sketchpad and visual imagery. To mitigate this deficiency, we introduce Cognitive Supersensing, a novel training paradigm that endows MLLMs with human-like visual imagery capabilities by integrating a Latent Visual Imagery Prediction (LVIP) head that jointly learns sequences of visual cognitive latent embeddings and aligns them with the answer, thereby forming vision-based internal reasoning chains. We further introduce a reinforcement learning stage that optimizes text reasoning paths based on this grounded visual latent. To evaluate the cognitive capabilities of MLLMs, we present CogSense-Bench, a comprehensive visual question answering (VQA) benchmark assessing five cognitive dimensions. Extensive experiments demonstrate that MLLMs trained with Cognitive Supersensing significantly outperform state-of-the-art baselines on CogSense-Bench and exhibit superior generalization on out-of-domain mathematics and science VQA benchmarks, suggesting that internal visual imagery is potentially key to bridging the gap between perceptual recognition and cognitive understanding. We will open-source the CogSense-Bench and our model weights.

Xiaona Zhou, Muntasir Wahed, Tianjiao Yu et al.

Recent advances in Vision-Language Models (VLMs) have achieved impressive performance across many tasks, yet prior studies report unsatisfactory performance when applying large language or multimodal models to finding abnormal patterns in sequential data. Public anomaly detection benchmarks typically provide interval annotations but not natural-language rationales, making it difficult to fine-tune VLMs to produce grounded, interpretable decisions. To address this gap, we construct VisAnomBench, a curated benchmark built from public time-series datasets and augmented with high-quality anomaly explanations selected from multiple large VLMs using fine-grained, task-specific rewards. Through fine-tuning on this benchmark, we develop VisAnomReasoner, a parameter-efficient VLM for time-series anomaly detection. Experimental results on VisAnomBench show that VisAnomReasoner achieves more accurate anomaly localization and consistently outperforms all baselines, with improvements of at least 21.23 and 23.87 percentage points in precision and F1, respectively. Additional experiments on the TSB-AD-U benchmark demonstrate strong cross-benchmark generalization, with VisAnomReasoner improving precision and F1 by 9.57 and 13.39 percentage points, respectively.

Onkar Susladkar, Tushar Prakash, Adheesh Juvekar et al.

Discrete video VAEs underpin modern text-to-video generation and video understanding systems, yet existing tokenizers typically learn visual codebooks at a single scale with limited vocabularies and shallow language supervision, leading to poor cross-modal alignment and zero-shot transfer. We introduce PyraTok, a language-aligned pyramidal tokenizer that learns semantically structured discrete latents across multiple spatiotemporal resolutions. PyraTok builds on a pretrained video VAE and a novel Language aligned Pyramidal Quantization (LaPQ) module that discretizes encoder features at several depths using a shared large binary codebook, yielding compact yet expressive video token sequences. To tightly couple visual tokens with language, PyraTok jointly optimizes multi-scale text-guided quantization and a global autoregressive objective over the token hierarchy. Across ten benchmarks, PyraTok delivers state-of-the-art (SOTA) video reconstruction, consistently improves text-to-video quality, and sets new SOTA zero-shot performance on video segmentation, temporal action localization, and video understanding, scaling robustly to up to 4K/8K resolutions.

Onkar Susladkar, Tushar Prakash, Gayatri Deshmukh et al.

We propose UniDFlow, a unified discrete flow-matching framework for multimodal understanding, generation, and editing. It decouples understanding and generation via task-specific low-rank adapters, avoiding objective interference and representation entanglement, while a novel reference-based multimodal preference alignment optimizes relative outcomes under identical conditioning, improving faithfulness and controllability without large-scale retraining. UniDFlpw achieves SOTA performance across eight benchmarks and exhibits strong zero-shot generalization to tasks including inpainting, in-context image generation, reference-based editing, and compositional generation, despite no explicit task-specific training.

Sai Gurrapu, Ajay Kulkarni, Lifu Huang et al.

Recent advances in deep learning have improved the performance of many Natural Language Processing (NLP) tasks such as translation, question-answering, and text classification. However, this improvement comes at the expense of model explainability. Black-box models make it difficult to understand the internals of a system and the process it takes to arrive at an output. Numerical (LIME, Shapley) and visualization (saliency heatmap) explainability techniques are helpful; however, they are insufficient because they require specialized knowledge. These factors led rationalization to emerge as a more accessible explainable technique in NLP. Rationalization justifies a model's output by providing a natural language explanation (rationale). Recent improvements in natural language generation have made rationalization an attractive technique because it is intuitive, human-comprehensible, and accessible to non-technical users. Since rationalization is a relatively new field, it is disorganized. As the first survey, rationalization literature in NLP from 2007-2022 is analyzed. This survey presents available methods, explainable evaluations, code, and datasets used across various NLP tasks that use rationalization. Further, a new subfield in Explainable AI (XAI), namely, Rational AI (RAI), is introduced to advance the current state of rationalization. A discussion on observed insights, challenges, and future directions is provided to point to promising research opportunities.

Afrina Tabassum, Muntasir Wahed, Hoda Eldardiry et al.

One of the challenges in contrastive learning is the selection of appropriate \textit{hard negative} examples, in the absence of label information. Random sampling or importance sampling methods based on feature similarity often lead to sub-optimal performance. In this work, we introduce UnReMix, a hard negative sampling strategy that takes into account anchor similarity, model uncertainty and representativeness. Experimental results on several benchmarks show that UnReMix improves negative sample selection, and subsequently downstream performance when compared to state-of-the-art contrastive learning methods.

Yuanzhe Liu, Jingyuan Zhu, Yuchen Mo et al.

Recent advancements in vision-language-action (VLA) models have shown promise in robotic manipulation, yet they continue to struggle with long-horizon, multi-step tasks. Existing methods lack internal reasoning mechanisms that can identify task-relevant interaction cues or track progress within a subtask, leading to critical execution errors such as repeated actions, missed steps, and premature termination. To address these challenges, we introduce PALM, a VLA framework that structures policy learning around interaction-centric affordance reasoning and subtask progress cues. PALM distills complementary affordance representations that capture object relevance, contact geometry, spatial placements, and motion dynamics, and serve as task-relevant anchors for visuomotor control. To further stabilize long-horizon execution, PALM predicts continuous within-subtask progress, enabling seamless subtask transitions. Across extensive simulation and real-world experiments, PALM consistently outperforms baselines, achieving a 91.8% success rate on LIBERO-LONG, a 12.5% improvement in average length on CALVIN ABC->D, and a 2x improvement over real-world baselines across three long-horizon generalization settings.

Haoran Yuan, Weigang Yi, Zhenyu Zhang et al.

Video-Action Models (VAMs) have emerged as a promising framework for embodied intelligence, learning implicit world dynamics from raw video streams to produce temporally consistent action predictions. Although such models demonstrate strong performance on long-horizon tasks through visual reasoning, they remain limited in contact-rich scenarios where critical interaction states are only partially observable from vision alone. In particular, fine-grained force modulation and contact transitions are not reliably encoded in visual tokens, leading to unstable or imprecise behaviors. To bridge this gap, we introduce the Video-Tactile Action Model (VTAM), a multimodal world modeling framework that incorporates tactile perception as a complementary grounding signal. VTAM augments a pretrained video transformer with tactile streams via a lightweight modality transfer finetuning, enabling efficient cross-modal representation learning without tactile-language paired data or independent tactile pretraining. To stabilize multimodal fusion, we introduce a tactile regularization loss that enforces balanced cross-modal attention, preventing visual latent dominance in the action model. VTAM demonstrates superior performance in contact-rich manipulation, maintaining a robust success rate of 90 percent on average. In challenging scenarios such as potato chip pick-and-place requiring high-fidelity force awareness, VTAM outperforms the pi 0.5 baseline by 80 percent. Our findings demonstrate that integrating tactile feedback is essential for correcting visual estimation errors in world action models, providing a scalable approach to physically grounded embodied foundation models.

Muntasir Wahed, Afrina Tabassum, Ismini Lourentzou

Contrastive learning has gained popularity as an effective self-supervised representation learning technique. Several research directions improve traditional contrastive approaches, e.g., prototypical contrastive methods better capture the semantic similarity among instances and reduce the computational burden by considering cluster prototypes or cluster assignments, while adversarial instance-wise contrastive methods improve robustness against a variety of attacks. To the best of our knowledge, no prior work jointly considers robustness, cluster-wise semantic similarity and computational efficiency. In this work, we propose SwARo, an adversarial contrastive framework that incorporates cluster assignment permutations to generate representative adversarial samples. We evaluate SwARo on multiple benchmark datasets and against various white-box and black-box attacks, obtaining consistent improvements over state-of-the-art baselines.

Makanjuola Ogunleye, Eman Abdelrahman, Ismini Lourentzou

Large multimodal models are increasingly used as the reasoning core of embodied agents operating in 3D environments, yet they remain prone to hallucinations that can produce unsafe and ungrounded decisions. Existing inference-time hallucination mitigation methods largely target 2D vision-language settings and do not transfer to embodied 3D reasoning, where failures arise from object presence, spatial layout, and geometric grounding rather than pixel-level inconsistencies. We introduce 3D-VCD, the first inference-time visual contrastive decoding framework for hallucination mitigation in 3D embodied agents. 3D-VCD constructs a distorted 3D scene graph by applying semantic and geometric perturbations to object-centric representations, such as category substitutions and coordinate or extent corruption. By contrasting predictions under the original and distorted 3D contexts, our method suppresses tokens that are insensitive to grounded scene evidence and are therefore likely driven by language priors. We evaluate 3D-VCD on the 3D-POPE and HEAL benchmarks and show that it consistently improves grounded reasoning without any retraining, establishing inference-time contrastive decoding over structured 3D representations as an effective and practical route to more reliable embodied intelligence.

Ying Shen, Daniel Bis, Cynthia Lu et al.

The research community has shown increasing interest in designing intelligent embodied agents that can assist humans in accomplishing tasks. Although there have been significant advancements in related vision-language benchmarks, most prior work has focused on building agents that follow instructions rather than endowing agents the ability to ask questions to actively resolve ambiguities arising naturally in embodied environments. To address this gap, we propose an Embodied Learning-By-Asking (ELBA) model that learns when and what questions to ask to dynamically acquire additional information for completing the task. We evaluate ELBA on the TEACh vision-dialog navigation and task completion dataset. Experimental results show that the proposed method achieves improved task performance compared to baseline models without question-answering capabilities.

Yifan Shen, Jiateng Liu, Xinzhuo Li et al.

Generative world models have shown promise for simulating dynamic environments, yet egocentric video remains challenging due to rapid viewpoint changes, frequent hand-object interactions, and goal-directed procedures whose evolution depends on latent human intent. Existing approaches either focus on hand-centric instructional synthesis with limited scene evolution, perform static view translation without modeling action dynamics, or rely on dense supervision, such as camera trajectories, long video prefixes, synchronized multicamera capture, etc. In this work, we introduce EgoForge, an egocentric goal-directed world simulator that generates coherent, first-person video rollouts from minimal static inputs: a single egocentric image, a high-level instruction, and an optional auxiliary exocentric view. To improve intent alignment and temporal consistency, we propose VideoDiffusionNFT, a trajectory-level reward-guided refinement that optimizes goal completion, temporal causality, scene consistency, and perceptual fidelity during diffusion sampling. Extensive experiments show EgoForge achieves consistent gains in semantic alignment, geometric stability, and motion fidelity over strong baselines, and robust performance in real-world smart-glasses experiments.

Tianjiao Yu, Sukrit Venkatagiri, Ismini Lourentzou et al.

Social media platforms have enabled extremists to organize violent events, such as the 2021 U.S. Capitol Attack. Simultaneously, these platforms enable professional investigators and amateur sleuths to collaboratively collect and identify imagery of suspects with the goal of holding them accountable for their actions. Through a case study of Sedition Hunters, a Twitter community whose goal is to identify individuals who participated in the 2021 U.S. Capitol Attack, we explore what are the main topics or targets of the community, who participates in the community, and how. Using topic modeling, we find that information sharing is the main focus of the community. We also note an increase in awareness of privacy concerns. Furthermore, using social network analysis, we show how some participants played important roles in the community. Finally, we discuss implications for the content and structure of online crowdsourced investigations.

Joy T Wu, Daniel Beckmann, Sarah Miller et al.

[18F]FDG-PET/CT is a cornerstone imaging modality for tumor staging and treatment response assessment across many cancer types, yet expert reader shortages necessitate more efficient diagnostic aids. While standalone AI models for automatic lesion segmentation exist, clinical translation remains hindered by concerns about interpretability, explainability, reliability, and workflow integration. We present GazeXPErT, a 4D eye-tracking dataset capturing expert search patterns during tumor detection and measurement on 346 FDG-PET/CT scans. Each study was read by a trainee and a board-certified nuclear medicine or radiology specialist using an eye-tracking-enabled annotation platform that simulates routine clinical reads. From 3,948 minutes of raw 60Hz eye-tracking data, 9,030 unique gaze-to-lesion trajectories were extracted, synchronized with PET/CT image slices, and rendered in COCO-style format for multiple machine learning applications. Baseline validation experiments demonstrate that a 3D nnUNet tumor segmentation model achieved superior performance when incorporating expert gaze patterns versus without (DICE score 0.6819 versus 0.6008), and that vision transformers trained on sequential gaze and PET/CT images can improve dynamic lesion localization (74.95% predicted gaze point closer to tumor) and expert intention prediction (Accuracy 67.53% and AUROC 0.747). GazeXPErT is a valuable resource designed to explore multiple machine learning problems beyond these baseline experiments, which include and are not limited to, visual grounding or causal reasoning, clinically explainable feature augmentation, human-computer interaction, human intention prediction or understanding, and expert gaze-rewarded modeling approaches to AI in oncologic FDG-PET/CT imaging.

Yifei Wang, Yueqi Wang, Zhenrui Yue et al.

The deployment of Large Language Models (LLMs) faces a critical bottleneck when handling lengthy inputs: the prohibitive memory footprint of the Key Value (KV) cache. To address this bottleneck, the token pruning paradigm leverages attention sparsity to selectively retain a small, critical subset of tokens. However, existing approaches fall short, with static methods risking irreversible information loss and dynamic strategies employing heuristics that insufficiently capture the query-dependent nature of token importance. We propose FASA, a novel framework that achieves query-aware token eviction by dynamically predicting token importance. FASA stems from a novel insight into RoPE: the discovery of functional sparsity at the frequency-chunk (FC) level. Our key finding is that a small, identifiable subset of "dominant" FCs consistently exhibits high contextual agreement with the full attention head. This provides a robust and computationally free proxy for identifying salient tokens. %making them a powerful and efficient proxy for token importance. Building on this insight, FASA first identifies a critical set of tokens using dominant FCs, and then performs focused attention computation solely on this pruned subset. % Since accessing only a small fraction of the KV cache, FASA drastically lowers memory bandwidth requirements and computational cost. Across a spectrum of long-context tasks, from sequence modeling to complex CoT reasoning, FASA consistently outperforms all token-eviction baselines and achieves near-oracle accuracy, demonstrating remarkable robustness even under constraint budgets. Notably, on LongBench-V1, FASA reaches nearly 100\% of full-KV performance when only keeping 256 tokens, and achieves 2.56$\times$ speedup using just 18.9\% of the cache on AIME24.

Yifan Shen, Jiawen Zhang, Jian Xu et al.

While agentic AI and its core multimodal large language models (MLLMs) have demonstrated remarkable promise in language and visual reasoning across domains ranging from daily life to advanced scientific research, a profound gap remains between artificial and human intelligence. Despite the integration of powerful tools and advanced MLLMs, state-of-the-art AI agents frequently fail at foundational, seemingly simple tasks that a child can resolve with ease. Inspired by the Wechsler Intelligence Scale for Children (WISC), we introduce ChildAgentEval, the first psychometrically grounded interactive benchmark for evaluating cognitive age alignment in MLLM-based agents. ChildAgentEval systematically compares the reasoning performance of various MLLM-based interactive agents against age-specific human developmental stages, exposing where current agentic AI systems can and cannot simulate age-specific cognitive behavior.

Tianjiao Yu, Xinzhuo Li, Muntasir Wahed et al.

Understanding and generating 3D objects as compositions of meaningful parts is fundamental to human perception and reasoning. However, most text-to-3D methods overlook the semantic and functional structure of parts. While recent part-aware approaches introduce decomposition, they remain largely geometry-focused, lacking semantic grounding and failing to model how parts align with textual descriptions or their inter-part relations. We propose DreamPartGen, a framework for semantically grounded, part-aware text-to-3D generation. DreamPartGen introduces Duplex Part Latents (DPLs) that jointly model each part's geometry and appearance, and Relational Semantic Latents (RSLs) that capture inter-part dependencies derived from language. A synchronized co-denoising process enforces mutual geometric and semantic consistency, enabling coherent, interpretable, and text-aligned 3D synthesis. Across multiple benchmarks, DreamPartGen delivers state-of-the-art performance in geometric fidelity and text-shape alignment.

Afrina Tabassum, Dung Tran, Trung Dang et al.

Masked Autoencoders (MAEs) learn rich low-level representations from unlabeled data but require substantial labeled data to effectively adapt to downstream tasks. Conversely, Instance Discrimination (ID) emphasizes high-level semantics, offering a potential solution to alleviate annotation requirements in MAEs. Although combining these two approaches can address downstream tasks with limited labeled data, naively integrating ID into MAEs leads to extended training times and high computational costs. To address this challenge, we introduce uaMix-MAE, an efficient ID tuning strategy that leverages unsupervised audio mixtures. Utilizing contrastive tuning, uaMix-MAE aligns the representations of pretrained MAEs, thereby facilitating effective adaptation to task-specific semantics. To optimize the model with small amounts of unlabeled data, we propose an audio mixing technique that manipulates audio samples in both input and virtual label spaces. Experiments in low/few-shot settings demonstrate that \modelname achieves 4-6% accuracy improvements over various benchmarks when tuned with limited unlabeled data, such as AudioSet-20K. Code is available at https://github.com/PLAN-Lab/uamix-MAE

Muntasir Wahed, Kiet A. Nguyen, Adheesh Sunil Juvekar et al.

Despite significant advancements in Large Vision-Language Models (LVLMs), existing pixel-grounding models operate on single-image settings, limiting their ability to perform detailed, fine-grained comparisons across multiple images. Conversely, current multi-image understanding models lack pixel-level grounding. Our work addresses this gap by introducing the task of multi-image pixel-grounded reasoning segmentation, and PRIMA, a novel LVLM that integrates pixel-level grounding with robust multi-image reasoning capabilities to produce contextually rich, pixel-grounded explanations. Central to PRIMA is an efficient vision module that queries fine-grained visual representations across multiple images, reducing TFLOPs by $25.3\%$. To support training and evaluation, we curate $M^4Seg$, a new reasoning segmentation benchmark consisting of $\sim$224K question-answer pairs that require fine-grained visual understanding across multiple images. Experimental results demonstrate PRIMA outperforms state-of-the-art baselines.

Ying Shen, Jerry Xiong, Tianjiao Yu et al.

Recent advances in generative video modeling, driven by large-scale datasets and powerful architectures, have yielded remarkable visual realism. However, emerging evidence suggests that simply scaling data and model size does not endow these systems with an understanding of the underlying physical laws that govern real-world dynamics. Existing approaches often fail to capture or enforce such physical consistency, resulting in unrealistic motion and dynamics. In his work, we investigate whether integrating the inference of latent physical properties directly into the video generation process can equip models with the ability to produce physically plausible videos. To this end, we propose Phantom, a Physics-Infused Video Generation model that jointly models the visual content and latent physical dynamics. Conditioned on observed video frames and inferred physical states, Phantom jointly predicts latent physical dynamics and generates future video frames. Phantom leverages a physics-aware video representation that serves as an abstract yet informaive embedding of the underlying physics, facilitating the joint prediction of physical dynamics alongside video content without requiring an explicit specification of a complex set of physical dynamics and properties. By integrating the inference of physical-aware video representation directly into the video generation process, Phantom produces video sequences that are both visually realistic and physically consistent. Quantitative and qualitative results on both standard video generation and physics-aware benchmarks demonstrate that Phantom not only outperforms existing methods in terms of adherence to physical dynamics but also delivers competitive perceptual fidelity.

Meghana Holla, Ismini Lourentzou

Zero-shot Natural Language-Video Localization (NLVL) methods have exhibited promising results in training NLVL models exclusively with raw video data by dynamically generating video segments and pseudo-query annotations. However, existing pseudo-queries often lack grounding in the source video, resulting in unstructured and disjointed content. In this paper, we investigate the effectiveness of commonsense reasoning in zero-shot NLVL. Specifically, we present CORONET, a zero-shot NLVL framework that leverages commonsense to bridge the gap between videos and generated pseudo-queries via a commonsense enhancement module. CORONET employs Graph Convolution Networks (GCN) to encode commonsense information extracted from a knowledge graph, conditioned on the video, and cross-attention mechanisms to enhance the encoded video and pseudo-query representations prior to localization. Through empirical evaluations on two benchmark datasets, we demonstrate that CORONET surpasses both zero-shot and weakly supervised baselines, achieving improvements up to 32.13% across various recall thresholds and up to 6.33% in mIoU. These results underscore the significance of leveraging commonsense reasoning for zero-shot NLVL.

Jiawei Liu, Nirav Diwan, Zhe Wang et al.

We introduce PurpCode, the first post-training recipe for training safe code reasoning models towards generating secure code and defending against malicious cyberactivities. PurpCode trains a reasoning model in two stages: (i) Rule Learning, which explicitly teaches the model to reference cybersafety rules to generate vulnerability-free code and to avoid facilitating malicious cyberactivities; and (ii) Reinforcement Learning, which optimizes model safety and preserves model utility through diverse, multi-objective reward mechanisms. To empower the training pipelines with comprehensive cybersafety data, we conduct internal red-teaming to synthesize comprehensive and high-coverage prompts based on real-world tasks for inducing unsafe cyberactivities in the model. Based on PurpCode, we develop a reasoning-based coding model, namely PurpCode-32B, which demonstrates state-of-the-art cybersafety, outperforming various frontier models. Meanwhile, our alignment method decreases the model overrefusal rates in both general and cybersafety-specific scenarios, while preserving model utility in both code generation and common security knowledge.

Onkar Susladkar, Dong-Hwan Jang, Tushar Prakash et al.

We introduce RewardFlow, an inversion-free framework that steers pretrained diffusion and flow-matching models at inference time through multi-reward Langevin dynamics. RewardFlow unifies complementary differentiable rewards for semantic alignment, perceptual fidelity, localized grounding, object consistency, and human preference, and further introduces a differentiable VQA-based reward that provides fine-grained semantic supervision through language-vision reasoning. To coordinate these heterogeneous objectives, we design a prompt-aware adaptive policy that extracts semantic primitives from the instruction, infers edit intent, and dynamically modulates reward weights and step sizes throughout sampling. Across several image editing and compositional generation benchmarks, RewardFlow delivers state-of-the-art edit fidelity and compositional alignment.

Kiet A. Nguyen, Adheesh Juvekar, Tianjiao Yu et al.

Recent advances in Large Vision-Language Models (LVLMs) have enabled general-purpose vision tasks through visual instruction tuning. While existing LVLMs can generate segmentation masks from text prompts for single images, they struggle with segmentation-grounded reasoning across images, especially at finer granularities such as object parts. In this paper, we introduce the new task of part-focused semantic co-segmentation, which involves identifying and segmenting common objects, as well as common and unique object parts across images. To address this task, we present CALICO, the first LVLM designed for multi-image part-level reasoning segmentation. CALICO features two key components, a novel Correspondence Extraction Module that identifies semantic part-level correspondences, and Correspondence Adaptation Modules that embed this information into the LVLM to facilitate multi-image understanding in a parameter-efficient manner. To support training and evaluation, we curate MixedParts, a large-scale multi-image segmentation dataset containing $\sim$2.4M samples across $\sim$44K images spanning diverse object and part categories. Experimental results demonstrate that CALICO, with just 0.3% of its parameters finetuned, achieves strong performance on this challenging task.

Kavana Venkatesh, Yusuf Dalva, Ismini Lourentzou et al.

We introduce a novel approach to enhance the capabilities of text-to-image models by incorporating a graph-based RAG. Our system dynamically retrieves detailed character information and relational data from the knowledge graph, enabling the generation of visually accurate and contextually rich images. This capability significantly improves upon the limitations of existing T2I models, which often struggle with the accurate depiction of complex or culturally specific subjects due to dataset constraints. Furthermore, we propose a novel self-correcting mechanism for text-to-image models to ensure consistency and fidelity in visual outputs, leveraging the rich context from the graph to guide corrections. Our qualitative and quantitative experiments demonstrate that Context Canvas significantly enhances the capabilities of popular models such as Flux, Stable Diffusion, and DALL-E, and improves the functionality of ControlNet for fine-grained image editing tasks. To our knowledge, Context Canvas represents the first application of graph-based RAG in enhancing T2I models, representing a significant advancement for producing high-fidelity, context-aware multi-faceted images.

Muntasir Wahed, Xiaona Zhou, Kiet A. Nguyen et al.

Recent advancements in Large Language Models (LLMs) have significantly enhanced their code generation capabilities. However, their robustness against adversarial misuse, particularly through multi-turn malicious coding prompts, remains underexplored. In this work, we introduce code decomposition attacks, where a malicious coding task is broken down into a series of seemingly benign subtasks across multiple conversational turns to evade safety filters. To facilitate systematic evaluation, we introduce \benchmarkname{}, a large-scale benchmark designed to evaluate the robustness of code LLMs against both single-turn and multi-turn malicious prompts. Empirical results across open- and closed-source models reveal persistent vulnerabilities, especially under multi-turn scenarios. Fine-tuning on MOCHA improves rejection rates while preserving coding ability, and importantly, enhances robustness on external adversarial datasets with up to 32.4% increase in rejection rates without any additional supervision.

Yifan Shen, Yuanzhe Liu, Jingyuan Zhu et al.

Current Vision-Language Models (VLMs) struggle with fine-grained spatial reasoning, particularly when multi-step logic and precise spatial alignment are required. In this work, we introduce SpatialReasoner-R1, a vision-language reasoning model designed to address these limitations. To construct high-quality supervision for spatial reasoning, we design a Multi-Model Monte Carlo Tree Search (M3CTS) method that generates diverse, logically consistent Long Chain-of-Thought (LongCoT) reasoning trajectories. In addition, we propose fine-grained Direct Preference Optimization (fDPO), which introduces segment-specific preference granularity for descriptive grounding and logical reasoning, guided by a spatial reward mechanism that evaluates candidate responses based on visual consistency, spatial grounding, and logical coherence. Experimental results demonstrate that fDPO achieves an average improvement of 4.1% over standard DPO across spatial quality tasks, and a 9.0% gain in spatial quantity tasks. SpatialReasoner-R1, trained with fDPO, sets a new SoTA on SPATIALRGPT-Bench, outperforming the strongest baseline by 9.8% in average accuracy, while maintaining competitive performance on general vision-language tasks.

Xiaona Zhou, Constantin Brif, Ismini Lourentzou

Multivariate time series anomaly detection (MTS-AD) is critical in domains like healthcare, cybersecurity, and industrial monitoring, yet remains challenging due to complex inter-variable dependencies, temporal dynamics, and sparse anomaly labels. We introduce mTSBench, the largest benchmark to date for MTS-AD and unsupervised model selection, spanning 344 labeled time series across 19 datasets and 12 diverse application domains. mTSBench evaluates 24 anomaly detection methods, including large language model (LLM)-based detectors for multivariate time series, and systematically benchmarks unsupervised model selection techniques under standardized conditions. Consistent with prior findings, our results confirm that no single detector excels across datasets, underscoring the importance of model selection. However, even state-of-the-art selection methods remain far from optimal, revealing critical gaps. mTSBench provides a unified evaluation suite to enable rigorous, reproducible comparisons and catalyze future advances in adaptive anomaly detection and robust model selection.

Ying Shen, Zhiyang Xu, Jiuhai Chen et al.

Recent advances in multimodal foundation models unifying image understanding and generation have opened exciting avenues for tackling a wide range of vision-language tasks within a single framework. Despite progress, existing unified models typically require extensive pretraining and struggle to achieve the same level of performance compared to models dedicated to each task. Additionally, many of these models suffer from slow image generation speeds, limiting their practical deployment in real-time or resource-constrained settings. In this work, we propose Layerwise Timestep-Expert Flow-based Transformer (LaTtE-Flow), a novel and efficient architecture that unifies image understanding and generation within a single multimodal model. LaTtE-Flow builds upon powerful pretrained Vision-Language Models (VLMs) to inherit strong multimodal understanding capabilities, and extends them with a novel Layerwise Timestep Experts flow-based architecture for efficient image generation. LaTtE-Flow distributes the flow-matching process across specialized groups of Transformer layers, each responsible for a distinct subset of timesteps. This design significantly improves sampling efficiency by activating only a small subset of layers at each sampling timestep. To further enhance performance, we propose a Timestep-Conditioned Residual Attention mechanism for efficient information reuse across layers. Experiments demonstrate that LaTtE-Flow achieves strong performance on multimodal understanding tasks, while achieving competitive image generation quality with around 6x faster inference speed compared to recent unified multimodal models.

Yingyan Zeng, Ismini Lourentzou, Xinwei Deng et al.

Artificial intelligence (AI) systems have been increasingly adopted in the Manufacturing Industrial Internet (MII). Investigating and enabling the AI resilience is very important to alleviate profound impact of AI system failures in manufacturing and Industrial Internet of Things (IIoT) operations, leading to critical decision making. However, there is a wide knowledge gap in defining the resilience of AI systems and analyzing potential root causes and corresponding mitigation strategies. In this work, we propose a novel framework for investigating the resilience of AI performance over time under hazard factors in data quality, AI pipelines, and the cyber-physical layer. The proposed method can facilitate effective diagnosis and mitigation strategies to recover AI performance based on a multimodal multi-head self latent attention model. The merits of the proposed method are elaborated using an MII testbed of connected Aerosol Jet Printing (AJP) machines, fog nodes, and Cloud with inference tasks via AI pipelines.

Xiaona Zhou, Yingyan Zeng, Ran Jin et al.

The success of modern machine learning hinges on access to high-quality training data. In many real-world scenarios, such as acquiring data from public repositories or sharing across institutions, data is naturally organized into discrete datasets that vary in relevance, quality, and utility. Selecting which repositories or institutions to search for useful datasets, and which datasets to incorporate into model training are therefore critical decisions, yet most existing methods select individual samples and treat all data as equally relevant, ignoring differences between datasets and their sources. In this work, we formalize the task of dataset selection: selecting entire datasets from a large, heterogeneous pool to improve downstream performance under resource constraints. We propose Dataset Selection via Hierarchies (DaSH), a dataset selection method that models utility at both dataset and group (e.g., collections, institutions) levels, enabling efficient generalization from limited observations. Across two public benchmarks (Digit-Five and DomainNet), DaSH outperforms state-of-the-art data selection baselines by up to 26.2% in accuracy, while requiring significantly fewer exploration steps. Ablations show DaSH is robust to low-resource settings and lack of relevant datasets, making it suitable for scalable and adaptive dataset selection in practical multi-source learning workflows.

Tianjiao Yu, Xinzhuo Li, Yifan Shen et al.

Recent advances in large multimodal models suggest that explicit reasoning mechanisms play a critical role in improving model reliability, interpretability, and cross-modal alignment. While such reasoning-centric approaches have been proven effective in language and vision tasks, their extension to 3D remains underdeveloped. CoRe3D introduces a unified 3D understanding and generation reasoning framework that jointly operates over semantic and spatial abstractions, enabling high-level intent inferred from language to directly guide low-level 3D content formation. Central to this design is a spatially grounded reasoning representation that decomposes 3D latent space into localized regions, allowing the model to reason over geometry in a compositional and procedural manner. By tightly coupling semantic chain-of-thought inference with structured spatial reasoning, CoRe3D produces 3D outputs that exhibit strong local consistency and faithful alignment with linguistic descriptions.

Afrina Tabassum, Bin Guo, Xiyao Ma et al.

Multimodal Procedural Planning (MPP) aims to generate step-by-step instructions that combine text and images, with the central challenge of preserving object-state consistency across modalities while producing informative plans. Existing approaches often leverage large language models (LLMs) to refine textual steps; however, visual object-state alignment and systematic evaluation are largely underexplored. We present MMPlanner, a zero-shot MPP framework that introduces Object State Reasoning Chain-of-Thought (OSR-CoT) prompting to explicitly model object-state transitions and generate accurate multimodal plans. To assess plan quality, we design LLM-as-a-judge protocols for planning accuracy and cross-modal alignment, and further propose a visual step-reordering task to measure temporal coherence. Experiments on RECIPEPLAN and WIKIPLAN show that MMPlanner achieves state-of-the-art performance, improving textual planning by +6.8%, cross-modal alignment by +11.9%, and visual step ordering by +26.7%

Xinzhuo Li, Adheesh Juvekar, Xingyou Liu et al.

Recent progress in vision-language segmentation has significantly advanced grounded visual understanding. However, these models often exhibit hallucinations by producing segmentation masks for objects not grounded in the image content or by incorrectly labeling irrelevant regions. Existing evaluation protocols for segmentation hallucination primarily focus on label or textual hallucinations without manipulating the visual context, limiting their capacity to diagnose critical failures. In response, we introduce HalluSegBench, the first benchmark specifically designed to evaluate hallucinations in visual grounding through the lens of counterfactual visual reasoning. Our benchmark consists of a novel dataset of 1340 counterfactual instance pairs spanning 281 unique object classes, and a set of newly introduced metrics that quantify hallucination sensitivity under visually coherent scene edits. Experiments on HalluSegBench with state-of-the-art vision-language segmentation models reveal that vision-driven hallucinations are significantly more prevalent than label-driven ones, with models often persisting in false segmentation, highlighting the need for counterfactual reasoning to diagnose grounding fidelity.

Tianjiao Yu, Vedant Shah, Muntasir Wahed et al.

Articulated objects are common in the real world, yet modeling their structure and motion remains a challenging task for 3D reconstruction methods. In this work, we introduce Part$^{2}$GS, a novel framework for modeling articulated digital twins of multi-part objects with high-fidelity geometry and physically consistent articulation. Part$^{2}$GS leverages a part-aware 3D Gaussian representation that encodes articulated components with learnable attributes, enabling structured, disentangled transformations that preserve high-fidelity geometry. To ensure physically consistent motion, we propose a motion-aware canonical representation guided by physics-based constraints, including contact enforcement, velocity consistency, and vector-field alignment. Furthermore, we introduce a field of repel points to prevent part collisions and maintain stable articulation paths, significantly improving motion coherence over baselines. Extensive evaluations on both synthetic and real-world datasets show that Part$^{2}$GS consistently outperforms state-of-the-art methods by up to 10$\times$ in Chamfer Distance for movable parts.

Amartya Dutta, Kazi Sajeed Mehrab, Medha Sawhney et al.

Scene-Graph Generation (SGG) seeks to recognize objects in an image and distill their salient pairwise relationships. Most methods depend on dataset-specific supervision to learn the variety of interactions, restricting their usefulness in open-world settings, involving novel objects and/or relations. Even methods that leverage large Vision Language Models (VLMs) typically require benchmark-specific fine-tuning. We introduce Open-World SGG, a training-free, efficient, model-agnostic framework that taps directly into the pretrained knowledge of VLMs to produce scene graphs with zero additional learning. Casting SGG as a zero-shot structured-reasoning problem, our method combines multimodal prompting, embedding alignment, and a lightweight pair-refinement strategy, enabling inference over unseen object vocabularies and relation sets. To assess this setting, we formalize an Open-World evaluation protocol that measures performance when no SGG-specific data have been observed either in terms of objects and relations. Experiments on Visual Genome, Open Images V6, and the Panoptic Scene Graph (PSG) dataset demonstrate the capacity of pretrained VLMs to perform relational understanding without task-level training.

Tianjiao Yu, Vedant Shah, Muntasir Wahed et al. · allen-ai

Expressing confidence is challenging for embodied agents navigating dynamic multimodal environments, where uncertainty arises from both perception and decision-making processes. We present the first work investigating embodied confidence elicitation in open-ended multimodal environments. We introduce Elicitation Policies, which structure confidence assessment across inductive, deductive, and abductive reasoning, along with Execution Policies, which enhance confidence calibration through scenario reinterpretation, action sampling, and hypothetical reasoning. Evaluating agents in calibration and failure prediction tasks within the Minecraft environment, we show that structured reasoning approaches, such as Chain-of-Thoughts, improve confidence calibration. However, our findings also reveal persistent challenges in distinguishing uncertainty, particularly under abductive settings, underscoring the need for more sophisticated embodied confidence elicitation methods.

Muntasir Wahed, Daniel Gruhl, Alfredo Alba et al.

Recent advances in text representation have shown that training on large amounts of text is crucial for natural language understanding. However, models trained without predefined notions of topical interest typically require careful fine-tuning when transferred to specialized domains. When a sufficient amount of within-domain text may not be available, expanding a seed corpus of relevant documents from large-scale web data poses several challenges. First, corpus expansion requires scoring and ranking each document in the collection, an operation that can quickly become computationally expensive as the web corpora size grows. Relying on dense vector spaces and pairwise similarity adds to the computational expense. Secondly, as the domain concept becomes more nuanced, capturing the long tail of domain-specific rare terms becomes non-trivial, especially under limited seed corpora scenarios. In this paper, we consider the problem of fast approximate corpus expansion given a small seed corpus with a few relevant documents as a query, with the goal of capturing the long tail of a domain-specific set of concept terms. To efficiently collect large-scale domain-specific corpora with limited relevance feedback, we propose a novel truncated sparse document bit-vector representation, termed Signature Assisted Unsupervised Corpus Expansion (SAUCE). Experimental results show that SAUCE can reduce the computational burden while ensuring high within-domain lexical coverage.

Joy T. Wu, Nkechinyere N. Agu, Ismini Lourentzou et al.

Despite the progress in automatic detection of radiologic findings from chest X-ray (CXR) images in recent years, a quantitative evaluation of the explainability of these models is hampered by the lack of locally labeled datasets for different findings. With the exception of a few expert-labeled small-scale datasets for specific findings, such as pneumonia and pneumothorax, most of the CXR deep learning models to date are trained on global "weak" labels extracted from text reports, or trained via a joint image and unstructured text learning strategy. Inspired by the Visual Genome effort in the computer vision community, we constructed the first Chest ImaGenome dataset with a scene graph data structure to describe $242,072$ images. Local annotations are automatically produced using a joint rule-based natural language processing (NLP) and atlas-based bounding box detection pipeline. Through a radiologist constructed CXR ontology, the annotations for each CXR are connected as an anatomy-centered scene graph, useful for image-level reasoning and multimodal fusion applications. Overall, we provide: i) $1,256$ combinations of relation annotations between $29$ CXR anatomical locations (objects with bounding box coordinates) and their attributes, structured as a scene graph per image, ii) over $670,000$ localized comparison relations (for improved, worsened, or no change) between the anatomical locations across sequential exams, as well as ii) a manually annotated gold standard scene graph dataset from $500$ unique patients.

Nkechinyere N. Agu, Joy T. Wu, Hanqing Chao et al.

Radiologists usually observe anatomical regions of chest X-ray images as well as the overall image before making a decision. However, most existing deep learning models only look at the entire X-ray image for classification, failing to utilize important anatomical information. In this paper, we propose a novel multi-label chest X-ray classification model that accurately classifies the image finding and also localizes the findings to their correct anatomical regions. Specifically, our model consists of two modules, the detection module and the anatomical dependency module. The latter utilizes graph convolutional networks, which enable our model to learn not only the label dependency but also the relationship between the anatomical regions in the chest X-ray. We further utilize a method to efficiently create an adjacency matrix for the anatomical regions using the correlation of the label across the different regions. Detailed experiments and analysis of our results show the effectiveness of our method when compared to the current state-of-the-art multi-label chest X-ray image classification methods while also providing accurate location information.

Safa Messaoud, Ismini Lourentzou, Assma Boughoula et al.

The recent growth of web video sharing platforms has increased the demand for systems that can efficiently browse, retrieve and summarize video content. Query-aware multi-video summarization is a promising technique that caters to this demand. In this work, we introduce a novel Query-Aware Hierarchical Pointer Network for Multi-Video Summarization, termed DeepQAMVS, that jointly optimizes multiple criteria: (1) conciseness, (2) representativeness of important query-relevant events and (3) chronological soundness. We design a hierarchical attention model that factorizes over three distributions, each collecting evidence from a different modality, followed by a pointer network that selects frames to include in the summary. DeepQAMVS is trained with reinforcement learning, incorporating rewards that capture representativeness, diversity, query-adaptability and temporal coherence. We achieve state-of-the-art results on the MVS1K dataset, with inference time scaling linearly with the number of input video frames.

Arkin Dharawat, Ismini Lourentzou, Alex Morales et al.

Given the widespread dissemination of inaccurate medical advice related to the 2019 coronavirus pandemic (COVID-19), such as fake remedies, treatments and prevention suggestions, misinformation detection has emerged as an open problem of high importance and interest for the research community. Several works study health misinformation detection, yet little attention has been given to the perceived severity of misinformation posts. In this work, we frame health misinformation as a risk assessment task. More specifically, we study the severity of each misinformation story and how readers perceive this severity, i.e., how harmful a message believed by the audience can be and what type of signals can be used to recognize potentially malicious fake news and detect refuted claims. To address our research questions, we introduce a new benchmark dataset, accompanied by detailed data analysis. We evaluate several traditional and state-of-the-art models and show there is a significant gap in performance when applying traditional misinformation classification models to this task. We conclude with open challenges and future directions.

Alexandros Karargyris, Satyananda Kashyap, Ismini Lourentzou et al.

We developed a rich dataset of Chest X-Ray (CXR) images to assist investigators in artificial intelligence. The data were collected using an eye tracking system while a radiologist reviewed and reported on 1,083 CXR images. The dataset contains the following aligned data: CXR image, transcribed radiology report text, radiologist's dictation audio and eye gaze coordinates data. We hope this dataset can contribute to various areas of research particularly towards explainable and multimodal deep learning / machine learning methods. Furthermore, investigators in disease classification and localization, automated radiology report generation, and human-machine interaction can benefit from these data. We report deep learning experiments that utilize the attention maps produced by eye gaze dataset to show the potential utility of this data.

Ismini Lourentzou, Kabir Manghnani, ChengXiang Zhai

Social media offer an abundant source of valuable raw data, however informal writing can quickly become a bottleneck for many natural language processing (NLP) tasks. Off-the-shelf tools are usually trained on formal text and cannot explicitly handle noise found in short online posts. Moreover, the variety of frequently occurring linguistic variations presents several challenges, even for humans who might not be able to comprehend the meaning of such posts, especially when they contain slang and abbreviations. Text Normalization aims to transform online user-generated text to a canonical form. Current text normalization systems rely on string or phonetic similarity and classification models that work on a local fashion. We argue that processing contextual information is crucial for this task and introduce a social media text normalization hybrid word-character attention-based encoder-decoder model that can serve as a pre-processing step for NLP applications to adapt to noisy text in social media. Our character-based component is trained on synthetic adversarial examples that are designed to capture errors commonly found in online user-generated text. Experiments show that our model surpasses neural architectures designed for text normalization and achieves comparable performance with state-of-the-art related work.