Prompt Disentanglement via Language Guidance and Representation Alignment for Domain Generalization

Domain Generalization (DG) seeks to develop a versatile model capable of performing effectively on unseen target domains. Notably, recent advances in pre-trained Visual Foundation Models (VFMs), such as CLIP, have demonstrated considerable potential in enhancing the generalization capabilities of deep learning models. Despite the increasing attention toward VFM-based domain prompt tuning within DG, the effective design of prompts capable of disentangling invariant features across diverse domains remains a critical challenge. In this paper, we propose addressing this challenge by leveraging the controllable and flexible language prompt of the VFM. Noting that the text modality of VFMs is naturally easier to disentangle, we introduce a novel framework for text feature-guided visual prompt tuning. This framework first automatically disentangles the text prompt using a large language model (LLM) and then learns domain-invariant visual representation guided by the disentangled text feature. However, relying solely on language to guide visual feature disentanglement has limitations, as visual features can sometimes be too complex or nuanced to be fully captured by descriptive text. To address this, we introduce Worst Explicit Representation Alignment (WERA), which extends text-guided visual prompts by incorporating an additional set of abstract prompts. These prompts enhance source domain diversity through stylized image augmentations, while alignment constraints ensure that visual representations remain consistent across both the original and augmented distributions. Experiments conducted on major DG datasets, including PACS, VLCS, OfficeHome, DomainNet, and TerraInc, demonstrate that our proposed method outperforms state-of-the-art DG methods.

RichControl: Structure- and Appearance-Rich Training-Free Spatial Control for Text-to-Image Generation

Text-to-image (T2I) diffusion models have shown remarkable success in generating high-quality images from text prompts. Recent efforts extend these models to incorporate conditional images (e.g., depth or pose maps) for fine-grained spatial control. Among them, feature injection methods have emerged as a training-free alternative to traditional fine-tuning approaches. However, they often suffer from structural misalignment, condition leakage, and visual artifacts, especially when the condition image diverges significantly from natural RGB distributions. By revisiting existing methods, we identify a core limitation: the synchronous injection of condition features fails to account for the trade-off between domain alignment and structural preservation during denoising. Inspired by this observation, we propose a flexible feature injection framework that decouples the injection timestep from the denoising process. At its core is a structure-rich injection module, which enables the model to better adapt to the evolving interplay between alignment and structure preservation throughout the diffusion steps, resulting in more faithful structural generation. In addition, we introduce appearance-rich prompting and a restart refinement strategy to further enhance appearance control and visual quality. Together, these designs enable training-free generation that is both structure-rich and appearance-rich. Extensive experiments show that our approach achieves state-of-the-art performance across diverse zero-shot conditioning scenarios.

SafePTR: Token-Level Jailbreak Defense in Multimodal LLMs via Prune-then-Restore Mechanism

By incorporating visual inputs, Multimodal Large Language Models (MLLMs) extend LLMs to support visual reasoning. However, this integration also introduces new vulnerabilities, making MLLMs susceptible to multimodal jailbreak attacks and hindering their safe deployment.Existing defense methods, including Image-to-Text Translation, Safe Prompting, and Multimodal Safety Tuning, attempt to address this by aligning multimodal inputs with LLMs' built-in safeguards.Yet, they fall short in uncovering root causes of multimodal vulnerabilities, particularly how harmful multimodal tokens trigger jailbreak in MLLMs? Consequently, they remain vulnerable to text-driven multimodal jailbreaks, often exhibiting overdefensive behaviors and imposing heavy training overhead.To bridge this gap, we present an comprehensive analysis of where, how and which harmful multimodal tokens bypass safeguards in MLLMs. Surprisingly, we find that less than 1% tokens in early-middle layers are responsible for inducing unsafe behaviors, highlighting the potential of precisely removing a small subset of harmful tokens, without requiring safety tuning, can still effectively improve safety against jailbreaks. Motivated by this, we propose Safe Prune-then-Restore (SafePTR), an training-free defense framework that selectively prunes harmful tokens at vulnerable layers while restoring benign features at subsequent layers.Without incurring additional computational overhead, SafePTR significantly enhances the safety of MLLMs while preserving efficiency. Extensive evaluations across three MLLMs and five benchmarks demonstrate SafePTR's state-of-the-art performance in mitigating jailbreak risks without compromising utility.

Multimodal Large Language Models for Medical Report Generation via Customized Prompt Tuning

Medical report generation from imaging data remains a challenging task in clinical practice. While large language models (LLMs) show great promise in addressing this challenge, their effective integration with medical imaging data still deserves in-depth exploration. In this paper, we present MRG-LLM, a novel multimodal large language model (MLLM) that combines a frozen LLM with a learnable visual encoder and introduces a dynamic prompt customization mechanism. Our key innovation lies in generating instance-specific prompts tailored to individual medical images through conditional affine transformations derived from visual features. We propose two implementations: prompt-wise and promptbook-wise customization, enabling precise and targeted report generation. Extensive experiments on IU X-ray and MIMIC-CXR datasets demonstrate that MRG-LLM achieves state-of-the-art performance in medical report generation. Our code will be made publicly available.

PRISM2: Unlocking Multi-Modal General Pathology AI with Clinical Dialogue

Recent pathology foundation models can provide rich tile-level representations but fall short of delivering general-purpose clinical utility without further extensive model development. These models lack whole-slide image (WSI) understanding and are not trained with large-scale diagnostic data, limiting their performance on diverse downstream tasks. We introduce PRISM2, a multi-modal slide-level foundation model trained via clinical dialogue to enable scalable, generalizable pathology AI. PRISM2 is trained on nearly 700,000 specimens (2.3 million WSIs) paired with real-world clinical diagnostic reports in a two-stage process. In Stage 1, a vision-language model is trained using contrastive and captioning objectives to align whole slide embeddings with textual clinical diagnosis. In Stage 2, the language model is unfrozen to enable diagnostic conversation and extract more clinically meaningful representations from hidden states. PRISM2 achieves strong performance on diagnostic and biomarker prediction tasks, outperforming prior slide-level models including PRISM and TITAN. It also introduces a zero-shot yes/no classification approach that surpasses CLIP-style methods without prompt tuning or class enumeration. By aligning visual features with clinical reasoning, PRISM2 improves generalization on both data-rich and low-sample tasks, offering a scalable path forward for building general pathology AI agents capable of assisting diagnostic and prognostic decisions.

NAP-Tuning: Neural Augmented Prompt Tuning for Adversarially Robust Vision-Language Models

Vision-Language Models (VLMs) such as CLIP have demonstrated remarkable capabilities in understanding relationships between visual and textual data through joint embedding spaces. Despite their effectiveness, these models remain vulnerable to adversarial attacks, particularly in the image modality, posing significant security concerns. Building upon our previous work on Adversarial Prompt Tuning (AdvPT), which introduced learnable text prompts to enhance adversarial robustness in VLMs without extensive parameter training, we present a significant extension by introducing the Neural Augmentor framework for Multi-modal Adversarial Prompt Tuning (NAP-Tuning).Our key innovations include: (1) extending AdvPT from text-only to multi-modal prompting across both text and visual modalities, (2) expanding from single-layer to multi-layer prompt architectures, and (3) proposing a novel architecture-level redesign through our Neural Augmentor approach, which implements feature purification to directly address the distortions introduced by adversarial attacks in feature space. Our NAP-Tuning approach incorporates token refiners that learn to reconstruct purified features through residual connections, allowing for modality-specific and layer-specific feature correction.Comprehensive experiments demonstrate that NAP-Tuning significantly outperforms existing methods across various datasets and attack types. Notably, our approach shows significant improvements over the strongest baselines under the challenging AutoAttack benchmark, outperforming them by 33.5% on ViT-B16 and 33.0% on ViT-B32 architectures while maintaining competitive clean accuracy.

TimeMaster: Training Time-Series Multimodal LLMs to Reason via Reinforcement Learning

Time-series reasoning remains a significant challenge in multimodal large language models (MLLMs) due to the dynamic temporal patterns, ambiguous semantics, and lack of temporal priors. In this work, we introduce TimeMaster, a reinforcement learning (RL)-based method that enables time-series MLLMs to perform structured, interpretable reasoning directly over visualized time-series inputs and task prompts. TimeMaster adopts a three-part structured output format, reasoning, classification, and domain-specific extension, and is optimized via a composite reward function that aligns format adherence, prediction accuracy, and open-ended insight quality. The model is trained using a two-stage pipeline: we first apply supervised fine-tuning (SFT) to establish a good initialization, followed by Group Relative Policy Optimization (GRPO) at the token level to enable stable and targeted reward-driven improvement in time-series reasoning. We evaluate TimeMaster on the TimerBed benchmark across six real-world classification tasks based on Qwen2.5-VL-3B-Instruct. TimeMaster achieves state-of-the-art performance, outperforming both classical time-series models and few-shot GPT-4o by over 14.6% and 7.3% performance gain, respectively. Notably, TimeMaster goes beyond time-series classification: it also exhibits expert-like reasoning behavior, generates context-aware explanations, and delivers domain-aligned insights. Our results highlight that reward-driven RL can be a scalable and promising path toward integrating temporal understanding into time-series MLLMs.

Using Vision Language Models to Detect Students' Academic Emotion through Facial Expressions

Students' academic emotions significantly influence their social behavior and learning performance. Traditional approaches to automatically and accurately analyze these emotions have predominantly relied on supervised machine learning algorithms. However, these models often struggle to generalize across different contexts, necessitating repeated cycles of data collection, annotation, and training. The emergence of Vision-Language Models (VLMs) offers a promising alternative, enabling generalization across visual recognition tasks through zero-shot prompting without requiring fine-tuning. This study investigates the potential of VLMs to analyze students' academic emotions via facial expressions in an online learning environment. We employed two VLMs, Llama-3.2-11B-Vision-Instruct and Qwen2.5-VL-7B-Instruct, to analyze 5,000 images depicting confused, distracted, happy, neutral, and tired expressions using zero-shot prompting. Preliminary results indicate that both models demonstrate moderate performance in academic facial expression recognition, with Qwen2.5-VL-7B-Instruct outperforming Llama-3.2-11B-Vision-Instruct. Notably, both models excel in identifying students' happy emotions but fail to detect distracted behavior. Additionally, Qwen2.5-VL-7B-Instruct exhibits relatively high performance in recognizing students' confused expressions, highlighting its potential for practical applications in identifying content that causes student confusion.

Quantitative Comparison of Fine-Tuning Techniques for Pretrained Latent Diffusion Models in the Generation of Unseen SAR Image Concepts

This work investigates the adaptation of large pre-trained latent diffusion models to a radically new imaging domain: Synthetic Aperture Radar (SAR). While these generative models, originally trained on natural images, demonstrate impressive capabilities in text-to-image synthesis, they are not natively adapted to represent SAR data, which involves different physics, statistical distributions, and visual characteristics. Using a sizeable SAR dataset (on the order of 100,000 to 1 million images), we address the fundamental question of fine-tuning such models for this unseen modality. We explore and compare multiple fine-tuning strategies, including full model fine-tuning and parameter-efficient approaches like Low-Rank Adaptation (LoRA), focusing separately on the UNet diffusion backbone and the text encoder components. To evaluate generative quality, we combine several metrics: statistical distance from real SAR distributions, textural similarity via GLCM descriptors, and semantic alignment assessed with a CLIP model fine-tuned on SAR data. Our results show that a hybrid tuning strategy yields the best performance: full fine-tuning of the UNet is better at capturing low-level SAR-specific patterns, while LoRA-based partial tuning of the text encoder, combined with embedding learning of the <SAR> token, suffices to preserve prompt alignment. This work provides a methodical strategy for adapting foundation models to unconventional imaging modalities beyond natural image domains.

Text to Image for Multi-Label Image Recognition with Joint Prompt-Adapter Learning

Benefited from image-text contrastive learning, pre-trained vision-language models, e.g., CLIP, allow to direct leverage texts as images (TaI) for parameter-efficient fine-tuning (PEFT). While CLIP is capable of making image features to be similar to the corresponding text features, the modality gap remains a nontrivial issue and limits image recognition performance of TaI. Using multi-label image recognition (MLR) as an example, we present a novel method, called T2I-PAL to tackle the modality gap issue when using only text captions for PEFT. The core design of T2I-PAL is to leverage pre-trained text-to-image generation models to generate photo-realistic and diverse images from text captions, thereby reducing the modality gap. To further enhance MLR, T2I-PAL incorporates a class-wise heatmap and learnable prototypes. This aggregates local similarities, making the representation of local visual features more robust and informative for multi-label recognition. For better PEFT, we further combine both prompt tuning and adapter learning to enhance classification performance. T2I-PAL offers significant advantages: it eliminates the need for fully semantically annotated training images, thereby reducing the manual annotation workload, and it preserves the intrinsic mode of the CLIP model, allowing for seamless integration with any existing CLIP framework. Extensive experiments on multiple benchmarks, including MS-COCO, VOC2007, and NUS-WIDE, show that our T2I-PAL can boost recognition performance by 3.47% in average above the top-ranked state-of-the-art methods.