The precise prediction of multi-scale traffic is a ubiquitous challenge in the urbanization process for car owners, road administrators, and governments. In the case of complex road networks, current and past traffic information from both upstream and downstream roads are crucial since various road networks have different semantic information about traffic. Rationalizing the utilization of semantic information can realize short-term, long-term, and unseen road traffic prediction. As the demands of multi-scale traffic analysis increase, on-demand interactions and visualizations are expected to be available for transportation participants. We have designed a multi-scale traffic generation system, namely TrafficGPT, using three AI agents to process multi-scale traffic data, conduct multi-scale traffic analysis, and present multi-scale visualization results. TrafficGPT consists of three essential AI agents: 1) a text-to-demand agent that is employed with Question & Answer AI to interact with users and extract prediction tasks through texts; 2) a traffic prediction agent that leverages multi-scale traffic data to generate temporal features and similarity, and fuse them with limited spatial features and similarity, to achieve accurate prediction of three tasks; and 3) a suggestion and visualization agent that uses the prediction results to generate suggestions and visualizations, providing users with a comprehensive understanding of traffic conditions. Our TrafficGPT system focuses on addressing concerns about traffic prediction from transportation participants, and conducted extensive experiments on five real-world road datasets to demonstrate its superior predictive and interactive performance

We present ZSL-RPPO, an improved zero-shot learning architecture that overcomes the limitations of teacher-student neural networks and enables generating robust, reliable, and versatile locomotion for quadrupedal robots in challenging terrains. We propose a new algorithm RPPO (Recurrent Proximal Policy Optimization) that directly trains recurrent neural network in partially observable environments and results in more robust training using domain randomization. Our locomotion controller supports extensive perturbation across simulation-to-reality transfer for both intrinsic and extrinsic physical parameters without further fine-tuning. This can avoid the significant decline of student's performance during simulation-to-reality transfer and therefore enhance the robustness and generalization of the locomotion controller. We deployed our controller on the Unitree A1 and Aliengo robots in real environment and exteroceptive perception is provided by either a solid-state Lidar or a depth camera. Our locomotion controller was tested in various challenging terrains like slippery surfaces, Grassy Terrain, and stairs. Our experiment results and comparison show that our approach significantly outperforms the state-of-the-art.

With the rapid advancement of large language models (LLMs), there is a pressing need for a comprehensive evaluation suite to assess their capabilities and limitations. Existing LLM leaderboards often reference scores reported in other papers without consistent settings and prompts, which may inadvertently encourage cherry-picking favored settings and prompts for better results. Our retrospective study on OpenAI's earlier models offers valuable insights into the evolutionary path from GPT-3 to GPT-4. Currently, the community is eager to know how GPT-3 progressively improves to GPT-4, including technical details like whether adding code data improves LLM's reasoning capability, which aspects of LLM capability can be improved by SFT and RLHF, how much is the alignment tax, etc. Our analysis sheds light on many of these questions, aiming to improve the transparency of advanced LLMs. Recently, the advancement of large language models (LLMs) is arguably the most remarkable breakthrough in Artificial Intelligence (AI) in the past few years. Based on the Transformer (Vaswani et al., 2017) architecture, these LLMs are trained on massive Web-scale text corpora. Despite their straightforward method of using a self-supervised objective to predict the next token, leading LLMs demonstrate exceptional capabilities across a range of challenging tasks (Bubeck et al., 2023), even showing a potential path towards Artificial General Intelligence (AGI). With the rapid progress of LLMs, there is a growing demand for better understanding these powerful models, including the distribution of their multi-aspect capabilities, limitations and risks, and directions and priorities of their future improvement. It is critical to establish a carefully curated evaluation suite that measures LLMs in a systematic, transparent and reproducible manner. Although there already exist many LLM leaderboards and evaluation suites, some key challenges are yet to be addressed: Inconsistent settings: The evaluation settings, such as the number of in-context example "shots", whether Chain-of-Thought (CoT; Wei et al. 2022) prompting is used, methods of answer parsing and metric computation, etc., often differ across the existing LLM works. Moreover, most of the released LLMs do not disclose their prompts used for evaluation, making it difficult to reproduce the reported scores.

The click-through rate (CTR) prediction task is to predict whether a user will click on the recommended item. As mind-boggling amounts of data are produced online daily, accelerating CTR prediction model training is critical to ensuring an up-to-date model and reducing the training cost. One approach to increase the training speed is to apply large batch training. However, as shown in computer vision and natural language processing tasks, training with a large batch easily suffers from the loss of accuracy. Our experiments show that previous scaling rules fail in the training of CTR prediction neural networks. To tackle this problem, we first theoretically show that different frequencies of ids make it challenging to scale hyperparameters when scaling the batch size. To stabilize the training process in a large batch size setting, we develop the adaptive Column-wise Clipping (CowClip). It enables an easy and effective scaling rule for the embeddings, which keeps the learning rate unchanged and scales the L2 loss. We conduct extensive experiments with four CTR prediction networks on two real-world datasets and successfully scaled 128 times the original batch size without accuracy loss. In particular, for CTR prediction model DeepFM training on the Criteo dataset, our optimization framework enlarges the batch size from 1K to 128K with over 0.1% AUC improvement and reduces training time from 12 hours to 10 minutes on a single V100 GPU. Our code locates at https://github.com/bytedance/LargeBatchCTR.