We introduce Eagle2.5, a frontier vision-language model (VLM) for long-context multimodal learning. Our work addresses the challenges in long video comprehension and high-resolution image understanding, introducing a generalist framework for both tasks. The proposed training framework incorporates Automatic Degrade Sampling and Image Area Preservation, two techniques that preserve contextual integrity and visual details. The framework also includes numerous efficiency optimizations in the pipeline for long-context data training. Finally, we propose Eagle-Video-110K, a novel dataset that integrates both story-level and clip-level annotations, facilitating long-video understanding. Eagle2.5 demonstrates substantial improvements on long-context multimodal benchmarks, providing a robust solution to the limitations of existing VLMs.

Advancements in direct-integration fine-tuning frameworks have underscored their potential to enhance the performance of labeled scenarios and tasks. To enhance the generalization of different categories in the same dataset, some methods have added visual loss to these frameworks for unlabeled scenarios. However, the performance of these methods through visual loss does not improve significantly in domain generalization and cross-dataset generalization tasks. This may be attributed to the uncoordinated learning of the two-modalities alignment and visual loss. To mitigate this issue of uncoordinated learning, we propose a novel method called Levelling Paradigm (LePa) to improve performance for unlabeled tasks or scenarios. The proposed LePa, designed as a plug-in module, dynamically constrains and coordinates multiple objective functions, thereby improving the generalization of these baseline methods. Comprehensive experiments have shown that our design can effectively address generalized scenarios and tasks.

Although multimodal large language models (MLLMs) exhibit remarkable reasoning capabilities on complex multimodal understanding tasks, they still suffer from the notorious "hallucination" issue: generating outputs misaligned with obvious visual or factual evidence. Currently, training-based solutions, like direct preference optimization (DPO), leverage paired preference data to suppress hallucinations. However, they risk sacrificing general reasoning capabilities due to the likelihood displacement. Meanwhile, training-free solutions, like contrastive decoding, achieve this goal by subtracting the estimated hallucination pattern from a distorted input. Yet, these handcrafted perturbations (e.g., add noise to images) may poorly capture authentic hallucination patterns. To avoid these weaknesses of existing methods, and realize "robust" hallucination mitigation (i.e., maintaining general reasoning performance), we propose a novel framework: Decoupling Contrastive Decoding (DCD).

Large vision-language models (LVLMs), which integrate a vision encoder (VE) with a large language model, have achieved remarkable success across various tasks. However, there are still crucial challenges in LVLMs such as object hallucination, generating descriptions of objects that are not in the input image. Here, we argue that uncertain visual tokens within the VE is a key factor that contributes to object hallucination. Our statistical analysis found that there are positive correlations between visual tokens with high epistemic uncertainty and the occurrence of hallucinations. Furthermore, we show theoretically and empirically that visual tokens in early VE layers that exhibit large representation deviations under small adversarial perturbations indicate high epistemic uncertainty. Based on these findings, we propose a simple yet effective strategy to mitigate object hallucination by modifying the VE only. Our method comprises a proxy method with adversarial perturbations for identifying uncertain visual tokens efficiently and a method to mask these uncertain visual tokens during the self-attention process in the middle layers of the VE, suppressing their influence on visual encoding and thus alleviating hallucinations. Extensive experiments show that our method significantly reduces object hallucinations in LVLMs and can synergistically work with other prior arts.

Understanding what and how neural networks memorize during training is crucial, both from the perspective of unintentional memorization of potentially sensitive information and from the standpoint of effective knowledge acquisition for realworld, knowledge-intensive tasks. While previous studies primarily investigate memorization within a single modality, such as text memorization in large language models or image memorization in diffusion models, unified multimodal models are becoming increasingly prevalent in practical applications. In this work, we focus on the unique characteristics of cross-modality memorization and conduct a systematic study centered on vision-language models. To facilitate controlled experiments, we first introduce a synthetic persona dataset comprising diverse synthetic person images and textual descriptions. We quantify factual knowledge memorization and cross-modal transferability by training models on a single modality and evaluating their performance in the other. Our results reveal that facts learned in one modality transfer to the other, but a significant gap exists between recalling information in the "source" and "target" modalities. Furthermore, we observe that this gap exists across various scenarios, including more capable models, machine unlearning, and the multi-hop case. At the end, we propose a baseline method to mitigate this challenge. We hope our study can inspire future research on developing more robust multimodal learning techniques to enhance cross-modal transferability.

Large Vision-Language Models (VLMs) have achieved remarkable success in understanding complex real-world scenarios and supporting data-driven decisionmaking processes. However, VLMs exhibit significant vulnerability against adversarial examples, either text or image, which can lead to various adversarial outcomes, e.g., jailbreaking, hijacking, and hallucination, etc. In this work, we empirically and theoretically demonstrate that VLMs are particularly susceptible to image-based adversarial examples, where imperceptible perturbations can precisely manipulate each output token. To this end, we propose a novel attack called Visionlanguage model Manipulation Attack (VMA), which integrates first-order and second-order momentum optimization techniques with a differentiable transformation mechanism to effectively optimize the adversarial perturbation.

Test-time adaptation (TTA) methods have gained significant attention for enhancing the performance of vision-language models (VLMs) such as CLIP during inference, without requiring additional labeled data. However, current TTA researches generally suffer from major limitations such as duplication of baseline results, limited evaluation metrics, inconsistent experimental settings, and insufficient analysis. These problems hinder fair comparisons between TTA methods and make it difficult to assess their practical strengths and weaknesses. To address these challenges, we introduce TTA-VLM, a comprehensive benchmark for evaluating TTA methods on VLMs. Our benchmark implements 8 episodic TTA and 7 online TTA methods within a unified and reproducible framework, and evaluates them across 15 widely used datasets. Unlike prior studies focused solely on CLIP, we extend the evaluation to SigLIP--a model trained with a Sigmoid loss--and include training-time tuning methods such as CoOp, MaPLe, and TeCoA to assess generality. Beyond classification accuracy, TTA-VLM incorporates various evaluation metrics, including robustness, calibration, out-of-distribution detection, and stability, enabling a more holistic assessment of TTA methods. Through extensive experiments, we find that 1) existing TTA methods produce limited gains compared to the previous pioneering work; 2) current TTA methods exhibit poor collaboration with trainingtime fine-tuning methods; 3) accuracy gains frequently come at the cost of reduced model trustworthiness. We release TTA-VLM to provide fair comparison and comprehensive evaluation of TTA methods for VLMs, and we hope it encourages the community to develop more reliable and generalizable TTA strategies.

Natural language can offer a concise and human-interpretable means of specifying reinforcement learning (RL) tasks. The ability to extract rewards from a language instruction can enable the development of robotic systems that can learn from human guidance; however, it remains a challenging problem, especially in visual environments. Existing approaches that employ large, pretrained language models either rely on non-visual environment representations, require prohibitively large amounts of feedback, or generate noisy, ill-shaped reward functions. In this paper, we propose a novel method, GoalLadder, that leverages vision-language models (VLMs) to train RL agents from a single language instruction in visual environments. GoalLadder works by incrementally discovering states that bring the agent closer to completing a task specified in natural language.

The established redundancy in visual tokens within large vision-language models (LVLMs) allows for pruning to effectively reduce their substantial computational demands. Empirical evidence from previous works indicates that visual tokens in later decoder stages receive less attention than shallow layers. Then, previous methods typically employ heuristics layer-specific pruning strategies where, although the number of tokens removed may differ across decoder layers, the overall pruning schedule is fixed and applied uniformly to all input samples and tasks, failing to align token elimination with the model's holistic reasoning trajectory. Cognitive science indicates that human visual processing often begins with broad exploration to accumulate evidence before narrowing focus as the target becomes distinct. Our experiments reveal an analogous pattern in LVLMs.

Prompt learning has emerged as a promising paradigm for adapting pre-trained vision-language models (VLMs) to few-shot whole slide image (WSI) classification by aligning visual features with textual representations, thereby reducing annotation cost and enhancing model generalization. Nevertheless, existing methods typically rely on slide-level prompts and fail to capture the subtype-specific phenotypic variations of histological entities (e.g., nuclei, glands) that are critical for cancer diagnosis. To address this gap, we propose Multi-scale Attribute-enhanced Prompt Learning (MAPLE), a hierarchical framework for few-shot WSI classification that jointly integrates multi-scale visual semantics and performs prediction at both the entity and slide levels. Specifically, we first leverage large language models (LLMs) to generate entity-level prompts that can help identify multi-scale histological entities and their phenotypic attributes, as well as slide-level prompts to capture global visual descriptions. Then, an entity-guided cross-attention module is proposed to generate entity-level features, followed by aligning with their corresponding subtype-specific attributes for fine-grained entity-level prediction. To enrich entity representations, we further develop a cross-scale entity graph learning module that can update these representations by capturing their semantic correlations within and across scales. The refined representations are then aggregated into a slide-level representation and aligned with the corresponding prompts for slide-level prediction. Finally, we combine both entity-level and slide-level outputs to produce the final prediction results. Results on three cancer cohorts confirm the effectiveness of our approach in addressing few-shot pathology diagnosis tasks.