In this study, we evaluate the ability of OpenAI's gpt-4o model to classify chest X-ray images as either NORMAL or PNEUMONIA in a zero-shot setting, without any prior fine-tuning. A balanced test set of 400 images (200 from each class) was used to assess performance across four distinct prompt designs, ranging from minimal instructions to detailed, reasoning-based prompts. The results indicate that concise, feature-focused prompts achieved the highest classification accuracy of 74\%, whereas reasoning-oriented prompts resulted in lower performance. These findings highlight that while ChatGPT exhibits emerging potential for medical image interpretation, its diagnostic reliability remains limited. Continued advances in visual reasoning and domain-specific adaptation are required before such models can be safely applied in clinical practice.

Learning a set of tasks over time, also known as continual learning (CL), is one of the most challenging problems in artificial intelligence due to catastrophic forgetting. Large language models (LLMs) are often impractical to frequent re-training and continual learning , due to high cost of computational resources for training. Moreover, LLM are not suitable for continual learning as updating these models over time for acquiring new knowledge leads to overwrites existing knowledge leading to common phenomenon know as \textit{catastrophic forgetting}. In this paper, we aim to address these concerns using a novel framework , COLA that employs an autoencoder to learn capture low-dimensional embeddings of the weights associated with various tasks. Our approach facilitates the transfer of knowledge to new tasks while preventing catastrophic forgetting, all without using data replay or a substantial set of task-specific parameters. Our approach, COLA, makes the LLM efficiently learn new tasks with minimal training, insignificant performance degradation on previous tasks, and eliminates the need for retaining earlier training data. Empirical evaluation on different datasets ranging from task oriented dialouge system to intent classsfication datasets showcases that our method not only overcomes catastrophic forgetting but also achieves significant reduction in parameter usage and memory size, across multiple tasks and outperforming the existing state of the art methods across multiple datasets.

Pruning is a widely used technique to reduce the size and inference cost of large language models (LLMs), but it often causes performance degradation. To mitigate this, existing restoration methods typically employ parameter-efficient fine-tuning (PEFT), such as LoRA, to recover the pruned model's performance. However, most PEFT methods are designed for dense models and overlook the distinct properties of pruned models, often resulting in suboptimal recovery. In this work, we propose a targeted restoration strategy for pruned models that restores performance while preserving their low cost and high efficiency. We observe that pruning-induced information loss is reflected in attention activations, and selectively reintroducing components of this information can significantly recover model performance. Based on this insight, we introduce RestoreLCC (Restoring Pruned LLMs via Lost Component Compensation), a plug-and-play method that contrastively probes critical attention heads via activation editing, extracts lost components from activation differences, and finally injects them back into the corresponding pruned heads for compensation and recovery. RestoreLCC is compatible with structured, semi-structured, and unstructured pruning schemes. Extensive experiments demonstrate that RestoreLCC consistently outperforms state-of-the-art baselines in both general and task-specific performance recovery, without compromising the sparsity or inference efficiency of pruned models.

Effectively reasoning about abstractive image inputs poses an elevated challenge for MLLMs, as it demands not only basic object recognition but also a deeper understanding and interpretation of the complex information encapsulated within these human-defined abstractive visual forms. Among the diverse array of abstractive images, an important area remains underexplored: ST ructured and A bstractive R easoning (ST AR) on images with M ulti-M odal R elational K nowledge (MMRK). As illustrated in Figure 1, MMRK consists of multiple multi-modal entities and concepts that are interconnected by abstract relational edges, representing well-organized and structured factual knowledge. Unlike natural or other abstractive images, MMRK offers a flexible and structured format for encoding complex semantic relations, with broad application potential (An et al., 2025). The relational links act as higher-order human-defined abstractions, modeling intricate connections among entities, and thus place greater demands on MLLM's reasoning capabilities. To accurately perform ST AR, MLLMs must understand both the entities and the underlying relational structure. However, ST AR remains largely unaddressed, with only a few studies (Zhang et al., 2024a; 2025d) briefly investigating this capability, which still face two critical challenges: (i) Lack of large-scale data synthesis method for ST AR. From the data perspective, there is a shortage of high-quality MMRK images and corresponding multi-modal instruction data. Automated pipelines for generating diverse and scalable MMRK datasets are missing, along with reliable chain-of-thought (CoT) reasoning annotations needed to improve MLLM's complex thinking and generalization ability.

This study examines the prompt fidelity of ChatGPT4o / DALL - E3 text - to - image visualisations by analysing whether anullributes explicitly specified in autogenously generated prompts are correctly rendered in the resulting images. Using two public - domain datasets comprising 200 visualisations of women working in the cultural and creative industries and 230 visualisations of museum curators, the study assessed accuracy across personal anullributes (age, hair), appearance (anullire, glasses), and paraphernalia (name tags, clipboards). While correctly rendered in most cases, DALL - E3 deviated from prompt specifications in 15.6% of all anullributes (n=710). Errors were lowest for paraphernalia, moderate for personal appearance, and highest for depictions of the person themselves, particularly age. These findings demonstrate measurable prompt - to - image fidelity gaps with implications for bias detection and model evaluation.

Earth Observation (EO) Foundation Modelling (FM) holds great promise for simplifying and improving the use of EO data for diverse real-world tasks. However, most existing models require additional adaptation before they can be used and are structured rigidly around particular data sources or training approaches. To address this, we take inspiration from large language models, where diverse tasks, both pre-training and downstream, are implicitly captured through next-token prediction over unified token sequences, leveraging the structure and diversity of the training data. We introduce SITS-DECO (Satellite Image Time Series-DECoder Only), a proof-of-concept generative model that applies this unified-sequence framing to EO data. Using a simple GPT-style decoder-only architecture, and demonstrate its ability to perform useful EO tasks (pixel-wise, multi-temporal, multi-modal crop-type classification) in a purely generative framework. Through symbolic prompting, we show that the model can perform multiple supervised and self-supervised tasks within a single unified architecture, without task- or modality-specific adaptation. Despite its simplicity and lack of spatial context, SITS-DECO outperforms much larger EO foundation models on crop-type classification (PASTIS-R) demonstrating that dense temporal sequence modelling is a critical missing ingredient in the current paradigm. This work exemplifies a data-centric modelling paradigm in which capability arises from the diversity and structure of the training data rather than from architectural complexity. SITS-DECO provides a lightweight, practical route to multi-modal, multi-task EO modelling, and a conceptual bridge toward future generative EO foundation models.

Recent breakthroughs in reasoning models have markedly advanced the reasoning capabilities of large language models, particularly via training on tasks with verifiable rewards. Y et, a significant gap persists in their adaptation to real-world mul-timodal scenarios, most notably, vision-language tasks, due to a heavy focus on single-modal language settings. While efforts to transplant reinforcement learning techniques from NLP to Visual Language Models (VLMs) have emerged, these approaches often remain confined to perception-centric tasks or reduce images to textual summaries, failing to fully exploit visual context and commonsense knowledge, ultimately constraining the generalization of reasoning capabilities across diverse multimodal environments. To address this limitation, we introduce a novel fine-tuning task, Masked Prediction via Context and Commonsense (MPCC), which forces models to integrate visual context and commonsense reasoning by reconstructing semantically meaningful content from occluded images, thereby laying the foundation for generalized reasoning. To systematically evaluate the model's performance in generalized reasoning, we developed a specialized evaluation benchmark, MPCC-Eval, and employed various fine-tuning strategies to guide reasoning. Among these, we introduced an innovative training method, Reinforcement Fine-Tuning with Prior Sampling, which not only enhances model performance but also improves its generalized reasoning capabilities in out-of-distribution (OOD) and cross-task scenarios. Code and data are available at yjainqdc.

The rapid advancement of time series foundation models (TSFMs) has been propelled by migrating architectures from language models. While existing TSFMs demonstrate impressive performance, their direct adoption of cross-domain architectures constrains effective capture of multiscale temporal dependencies inherent to time series data. This limitation becomes particularly pronounced during zero-shot transfer across datasets with divergent underlying patterns and sampling strategies. To address these challenges, we propose Hierarchical Interleaved Block Attention (HIBA) which employs hierarchical inter- and intra-block sparse attention to effectively capture multi-scale dependencies. Intra-block attention facilitates local information exchange, and inter-block attention operates across blocks to capture global temporal pattern interaction and dynamic evolution. Leveraging the HIBA architecture, we introduce Xihe, a scalable TSFM family spanning from an ultra-efficient 9.5M parameter configuration to high-capacity 1.5B variant. Evaluated on the comprehensive GIFT-Eval benchmark, our most compact Xihe-tiny model (9.5M) surpasses the majority of contemporary TSFMs, demonstrating remarkable parameter efficiency. More impressively, Xihe-max (1.5B) establishes new state-of-the-art zero-shot performance, surpassing previous best results by a substantial margin. This consistent performance excellence across the entire parameter spectrum provides compelling evidence for the exceptional generalization capabilities and architectural superiority of HIBA.

Vision-Language Models (VLMs) have become essential backbones of modern multimodal intelligence, yet their outputs remain prone to hallucination-plausible text misaligned with visual inputs. Existing alignment approaches often rely on expensive fine-tuning with annotated preference data or sequence-level inference strategies that provide only coarse, delayed feedback. To overcome these limitations, we present TITA (Token-level Inference-Time Alignment), a lightweight framework that freezes the base VLM and instead trains a reward model to approximate its distribution. During inference, implicit preference signals are extracted as log-probability ratios between the reward model and the target VLM, yielding dense autoregressive feedback. This formulation can be viewed as an inference-time variant of Direct Preference Optimization (DPO), providing token-level corrective signals without retraining the backbone. Extensive evaluations on LLaVA-1.5-7B and 13B show consistent gains across 12 benchmarks, with improvements of 8.6% on MMVet and 6.7% on POPE, indicating stronger general understanding and reduced hallucinations. Additional experiments on Qwen2.5-VL-7B and DeepSeek-VL2-27.5B show comparable gains, especially in hallucination reduction and VQA accuracy, while incurring negligible inference overhead.

Script event induction, which aims to predict the subsequent event based on the context, is a challenging task in NLP, achieving remarkable success in practical applications. However, human events are mostly recorded and presented in the form of videos rather than scripts, yet there is a lack of related research in the realm of vision. To address this problem, we introduce AVEP (Action-centric Video Event Prediction), a task that distinguishes itself from existing video prediction tasks through its incorporation of more complex logic and richer semantic information. We present a large structured dataset, which consists of about $35K$ annotated videos and more than $178K$ video clips of event, built upon existing video event datasets to support this task. The dataset offers more fine-grained annotations, where the atomic unit is represented as a multimodal event argument node, providing better structured representations of video events. Due to the complexity of event structures, traditional visual models that take patches or frames as input are not well-suited for AVEP. We propose EventFormer, a node-graph hierarchical attention based video event prediction model, which can capture both the relationships between events and their arguments and the coreferencial relationships between arguments. We conducted experiments using several SOTA video prediction models as well as LVLMs on AVEP, demonstrating both the complexity of the task and the value of the dataset. Our approach outperforms all these video prediction models. We will release the dataset and code for replicating the experiments and annotations.