Despite the advancements in training Large Language Models (LLMs) with alignment techniques to enhance the safety of generated content, these models remain susceptible to jailbreak, an adversarial attack method that exposes security vulnerabilities in LLMs. Notably, the Greedy Coordinate Gradient (GCG) method has demonstrated the ability to automatically generate adversarial suffixes that jailbreak state-of-the-art LLMs. However, the optimization process involved in GCG is highly time-consuming, rendering the jailbreaking pipeline inefficient. In this paper, we investigate the process of GCG and identify an issue of Indirect Effect, the key bottleneck of the GCG optimization. To this end, we propose the Model Attack Gradient Index GCG (MAGIC), that addresses the Indirect Effect by exploiting the gradient information of the suffix tokens, thereby accelerating the procedure by having less computation and fewer iterations. Our experiments on AdvBench show that MAGIC achieves up to a 1.5x speedup, while maintaining Attack Success Rates (ASR) on par or even higher than other baselines. Our MAGIC achieved an ASR of 74% on the Llama-2 and an ASR of 54% when conducting transfer attacks on GPT-3.5. Code is available at https://github.com/jiah-li/magic.

Transformer models have demonstrated exceptional performance across a wide range of applications. Though forming the foundation of Transformer models, the dot-product attention does not scale well to long-context data since its time requirement grows quadratically with context length. In this work, we propose Radar, a training-free approach that accelerates inference by dynamically searching for the most important context tokens. For any pre-trained Transformer, Radar can reduce the decoding time complexity without training or heuristically evicting tokens. Moreover, we provide theoretical justification for our approach, demonstrating that Radar can reliably identify the most important tokens with high probability. We conduct extensive comparisons with the previous methods on a wide range of tasks. The results demonstrate that Radar achieves the state-of-the-art performance across different architectures with reduced time complexity, offering a practical solution for efficient long-context processing of Transformers. The code is publicly available at https://github.com/BorealisAI/radar-decoding.

The widespread application of autonomous aerial vehicles (AAVs) in various fields has driven the development of target tracking based on AAV perspectives. Among the existing tracking algorithms, tracking by detection performs well, which depends heavily on the robustness of the object detection and re-identification algorithms. The images obtained from the perspective of drones have complex backgrounds and small image sizes, making tracking tasks more difficult. We introduce the parallel space and channel attention module to optimize the object detection algorithm, exploring compelling spatial and channel dimensions features. We propose the channel-spatial fully connected information extraction module, which explores the connections between pixels from a spatial perspective and considers the relationships between feature maps from a channel perspective, compensating for the shortcomings of convolutional networks in extracting global features. To obtain more discriminative feature descriptions, we design the global-local-residual feature fusion module that divides the image into global and local parts and gets the difference between local information and other information, which helps with data correlation and label matching. Experiments show that our method improves the mAP by 0.4% and AP by 2.1% in the object detection task. In the re-identification task, top-1 accuracy is improved by 5.3% and mAP by 11.3%.

Large language models have become increasingly popular and demonstrated remarkable performance in various natural language processing (NLP) tasks. However, these models are typically computationally expensive and difficult to be deployed in resource-constrained environments. In this paper, we propose LLMR, a novel knowledge distillation (KD) method based on a reward function induced from large language models. We conducted experiments on multiple datasets in the dialogue generation and summarization tasks. Empirical results demonstrate that our LLMR approach consistently outperforms traditional KD methods in different tasks and datasets.

Despite large neural networks demonstrating remarkable abilities to complete different tasks, they require excessive memory usage to store the optimization states for training. To alleviate this, the low-rank adaptation (LoRA) is proposed to reduce the optimization states by training fewer parameters. However, LoRA restricts overall weight update matrices to be low-rank, limiting the model performance. In this work, we investigate the dynamics of LoRA and identify that it can be approximated by a random projection. Based on this observation, we propose Flora, which is able to achieve high-rank updates by resampling the projection matrices while enjoying the sublinear space complexity of optimization states. We conduct experiments across different tasks and model architectures to verify the effectiveness of our approach.

Wenxuan Wang, Wenxiang Jiao, Yongchang Hao, Xing Wang, Shuming Shi, Zhaopeng Tu, Michael Lyu. Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2022.

Reinforcement learning (RL) has been widely used in text generation to alleviate the exposure bias issue or to utilize non-parallel datasets. The reward function plays an important role in making RL training successful. However, previous reward functions are typically task-specific and sparse, restricting the use of RL. In our work, we propose a task-agnostic approach that derives a step-wise reward function directly from a model trained with teacher forcing. We additionally propose a simple modification to stabilize the RL training on non-parallel datasets with our induced reward function. Empirical results show that our method outperforms self-training and reward regression methods on several text generation tasks, confirming the effectiveness of our reward function.

Yongchang Hao, Shilin He, Wenxiang Jiao, Zhaopeng Tu, Michael Lyu, Xing Wang. Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2021.