Chih-Wei Chen

Gemini Team, Petko Georgiev, Ving Ian Lei et al. · deepmind, mila

In this report, we introduce the Gemini 1.5 family of models, representing the next generation of highly compute-efficient multimodal models capable of recalling and reasoning over fine-grained information from millions of tokens of context, including multiple long documents and hours of video and audio. The family includes two new models: (1) an updated Gemini 1.5 Pro, which exceeds the February version on the great majority of capabilities and benchmarks; (2) Gemini 1.5 Flash, a more lightweight variant designed for efficiency with minimal regression in quality. Gemini 1.5 models achieve near-perfect recall on long-context retrieval tasks across modalities, improve the state-of-the-art in long-document QA, long-video QA and long-context ASR, and match or surpass Gemini 1.0 Ultra's state-of-the-art performance across a broad set of benchmarks. Studying the limits of Gemini 1.5's long-context ability, we find continued improvement in next-token prediction and near-perfect retrieval (>99%) up to at least 10M tokens, a generational leap over existing models such as Claude 3.0 (200k) and GPT-4 Turbo (128k). Finally, we highlight real-world use cases, such as Gemini 1.5 collaborating with professionals on completing their tasks achieving 26 to 75% time savings across 10 different job categories, as well as surprising new capabilities of large language models at the frontier; when given a grammar manual for Kalamang, a language with fewer than 200 speakers worldwide, the model learns to translate English to Kalamang at a similar level to a person who learned from the same content.

Liren Lin, Chih-Wei Chen

We provide a new interpretation of Hessian locally linear embedding (HLLE), revealing that it is essentially a variant way to implement the same idea of locally linear embedding (LLE). Based on the new interpretation, a substantial simplification can be made, in which the idea of "Hessian" is replaced by rather arbitrary weights. Moreover, we show by numerical examples that HLLE may produce projection-like results when the dimension of the target space is larger than that of the data manifold, and hence one further modification concerning the manifold dimension is suggested. Combining all the observations, we finally achieve a new LLE-type method, which is called tangential LLE (TLLE). It is simpler and more robust than HLLE.

Chih-Wei Chen, Charles Harrison, Hsin-Hsiung Huang

In real-world application scenarios, it is crucial for marine navigators and security analysts to predict vessel movement trajectories at sea based on the Automated Identification System (AIS) data in a given time span. This article presents an unsupervised method of ship movement trajectory prediction which represents the data in a three-dimensional space which consists of time difference between points, the scaled error distance between the tested and its predicted forward and backward locations, and the space-time angle. The representation feature space reduces the search scope for the next point to a collection of candidates which fit the local path prediction well, and therefore improve the accuracy. Unlike most statistical learning or deep learning methods, the proposed clustering-based trajectory reconstruction method does not require computationally expensive model training. This makes real-time reliable and accurate prediction feasible without using a training set. Our results show that the most prediction trajectories accurately consist of the true vessel paths.