Large language models (LLMs), with their billions of parameters, pose substantial challenges for deployment on edge devices, straining both memory capacity and computational resources. Block Floating Point (BFP) quantisation reduces memory and computational overhead by converting high-overhead floating point operations into low-bit fixed point operations. However, BFP requires aligning all data to the maximum exponent, which causes loss of small and moderate values, resulting in quantisation error and degradation in the accuracy of LLMs. To address this issue, we propose a Bidirectional Block Floating Point (BBFP) data format, which reduces the probability of selecting the maximum as shared exponent, thereby reducing quantisation error. By utilizing the features in BBFP, we present a full-stack Bidirectional Block Floating Point-Based Quantisation Accelerator for LLMs (BBAL), primarily comprising a processing element array based on BBFP, paired with proposed cost-effective nonlinear computation unit. Experimental results show BBAL achieves a 22% improvement in accuracy compared to an outlier-aware accelerator at similar efficiency, and a 40% efficiency improvement over a BFP-based accelerator at similar accuracy.

Visual reranking has become a widely-accepted method to improve traditional text-based image search results. The main principle is to exploit the visual aggregation property of relevant images among top results so as to boost ranking scores of relevant images, by explicitly or implicitly detecting the confident relevant images, and propagating ranking scores among visually similar images. However, such a visual aggregation property does not always hold, and thus these schemes may fail. In this paper, we instead propose to filter out the most probable irrelevant images using deep contexts, which is the extra information that is not limited in the current search results. The deep contexts for each image consist of sets of images that are returned by searches using the queries formed by the textual context of this image. We compare the popularity of this image in the current search results and the deep contexts to check the irrelevance score. Then the irrelevance scores are propagated to the images whose useful textual context is missed. We formulate the two schemes together to reach a Markov random field, which is effectively solved by graph cuts. The key is that our scheme does not rely on the assumption that relevant images are visually aggregated among top results and is based on the observation that an outlier under the current query is likely to be more popular under some other query. After that, we perform graph reranking over filtered results to reorder them. Experimental results on the INRIA dataset show that our proposed method achieves significant improvements over previous approaches.

Multi-core processors are becoming the dominant choice due to energy and thermal considerations, which also applies to embedded and real-time systems. While fixed-priority scheduling with task-splitting in real-time systems are widely applied, current approaches have not taken into consideration energy-aware aspects such as dynamic voltage/frequency scheduling (DVS). In this paper, we propose two strategies to apply DVS to fixed-priority scheduling algorithms with task-splitting for periodic real-time tasks on multi-core processors. We first propose a strategy that does traditional DVS for each processor after scheduling (post-DVS), which ensures all tasks meet the timing requirements on synchronization. We then propose a new strategy, which determines the frequency of each task before scheduling (pre-DVS) according to the total utilization of task-set and number of cores available, so that the system could take full advantage of all the cores. The combination of frequency pre-allocation and task-splitting makes it possible to maximize energy savings with DVS. We perform a series of simulations to compare the performance of each algorithm with two state-of-the-art scheduling algorithms with task-splitting. Simulation results show that the pre-DVS algorithm we proposed has performed satisfactorily on both schedulability and energy consumption in comparison to the previous approaches.