China Machine Press

Text-to-Image (T2I) models typically deploy safety mechanisms to prevent the generation of sensitive images. Unfortunately, recent jailbreaking attack methods manually design instructions for the LLM to generate adversarial prompts, which effectively exposing safety vulnerabilities of T2I models. However, existing methods have two limitations: 1) relying on manually exhaustive strategies for designing adversarial prompts, lacking a unified framework, and 2) requiring numerous queries to achieve a successful attack, limiting their practical applicability. To address this issue, we propose Reason2Attack~(R2A), which aims to enhance the effectiveness and efficiency of the LLM in jailbreaking attacks. Specifically, we first use Frame Semantics theory to systematize existing manually crafted strategies and propose a unified generation framework to generate CoT adversarial prompts step by step. Following this, we propose a two-stage LLM reasoning training framework guided by the attack process. In the first stage, the LLM is fine-tuned with CoT examples generated by the unified generation framework to internalize the adversarial prompt generation process grounded in Frame Semantics. In the second stage, we incorporate the jailbreaking task into the LLM's reinforcement learning process, guided by the proposed attack process reward function that balances prompt stealthiness, effectiveness, and length, enabling the LLM to understand T2I models and safety mechanisms. Extensive experiments on various T2I models with safety mechanisms, and commercial T2I models, show the superiority and practicality of R2A.

Text-to-image generation models can create diverse, high-quality images, but they frequently encounter challenges in accurately rendering text within those images due to the insufficient representation of desired text. In this study, we introduce RealText, a method for generating scene text images that excels in producing precise and realistic scene text images in any language. We disentangle scene text images generation into three stages: background and glyph image generation, text deformation, and whole image generation. Initially, we utilize prompts to guide the creation of well-organized background images. By identifying optimal text placements on these backgrounds, we render the glyph images of target text using user-specified font, effectively eliminating incorrect characters. In the next stage, we propose scene sensing to perceive text carrier surfaces and viewpoints through 3D scene reconstruction using depth and normal map to apply text deformation, thereby enhancing the realism of generated images. The final stage involves generating complete image with the aid of background and glyph guidance. Thanks to glyph disentangling, scene sensing, and text inpainting, we can exert more precise control over scene text image generation process. We have developed a unified framework which supports major generation models. Extensive experiments illustrate the exceptional performance of our method in generating images with multilingual text. The codes will soon be available at https://github.com/cccvl/RealText.

Trajectory generation is a pivotal task in autonomous driving. Recent studies have introduced the autoregressive paradigm, leveraging the state transition model to approximate future trajectory distributions. This paradigm closely mirrors the real-world trajectory generation process and has achieved notable success. However, its potential is limited by the ineffective representation of realistic trajectories within the redundant state space. To address this limitation, we propose the Kinematic-Driven Generative Model for Realistic Agent Simulation (KiGRAS). Instead of modeling in the state space, KiGRAS factorizes the driving scene into action probability distributions at each time step, providing a compact space to represent realistic driving patterns. By establishing physical causality from actions (cause) to trajectories (effect) through the kinematic model, KiGRAS eliminates massive redundant trajectories. All states derived from actions in the causal space are constrained to be physically feasible. Furthermore, redundant trajectories representing identical action sequences are mapped to the same representation, reflecting their underlying actions. This approach significantly reduces task complexity and ensures physical feasibility. KiGRAS achieves state-of-the-art performance in Waymo's SimAgents Challenge, ranking first on the WOMD leaderboard with significantly fewer parameters than other models.

Current methods for detecting deep fakes concentrate on specific patterns of forgery like noise characteristics, local textures, or frequency statistics. These approaches assume training and test sets exhibit similar data distributions, which bring severe performance drops and further limit broader applications when migrating unseen domains. Existing works show that reconstruction learning is effective in capturing unseen forgery clues. However, 2D reconstruction is insufficient and can not handle non-frontal face reconstruction, while 3D reconstruction provides more critical details of facial structure and finds accurate forgery regions. In this paper, we propose a bi-source reconstruction based classification network (BRCNet) to incorporate 2D and 3D reconstruction as the supervisions and learn the optimal feature representation. In detail, we employ an encoder-decoder architecture to facilitate reconstruction learning, enhancing the learned representations to detect forgery patterns that are unknown. To further capture forgery evidence across multiple scales, instead of using encoder features from the reconstruction network only, we build a feature improvement network to combine feature details from encoder and decoder features in a multi-scale fashion. In addition, we use the reconstruction difference to supervise the feature aggregation, which enables detecting the subtle and trivial discrepancies between fake and real video frames. Extensive experiments are conducted to validate the performance of our proposed method on several deep fake benchmarks. The results demonstrate the efficacy of our approach, offering promising results and showcasing its potential for practical applications. The source code is available at https://github.com/cccvl/BRCNet.

While large language models (LLMs) have exhibited impressive instruction-following capabilities, it is still unclear whether and to what extent they can respond to explicit constraints that might be entailed in various instructions. As a significant aspect of LLM alignment, it is thus important to formulate such a specialized set of instructions as well as investigate the resulting behavior of LLMs. To address this vacancy, we propose a new benchmark CoDI-Eval to systematically and comprehensively evaluate LLMs' responses to instructions with various constraints. We construct a large collection of constraints-attributed instructions as a test suite focused on both generalization and coverage. Specifically, we advocate an instruction diversification process to synthesize diverse forms of constraint expression and also deliberate the candidate task taxonomy with even finer-grained sub-categories. Finally, we automate the entire evaluation process to facilitate further developments. Different from existing studies on controllable text generation, CoDI-Eval extends the scope to the prevalent instruction-following paradigm for the first time. We provide extensive evaluations of representative LLMs (e.g., ChatGPT, Vicuna) on CoDI-Eval, revealing their limitations in following instructions with specific constraints and there is still a significant gap between open-source and commercial closed-source LLMs. We believe this benchmark will facilitate research into improving the controllability of LLMs' responses to instructions. Our data and code are available at https://github.com/Xt-cyh/CoDI-Eval.