On Leakage of Code Generation Evaluation Datasets

In this paper we consider contamination by code generation test sets, in particular in their use in modern large language models. We discuss three possible sources of such contamination and show findings supporting each of them: (i) direct data leakage, (ii) indirect data leakage through the use of synthetic data and (iii) overfitting to evaluation sets during model selection. Key to our findings is a new dataset of 161 prompts with their associated python solutions, dataset which is released at https://huggingface.co/datasets/CohereForAI/lbpp .

CAR-Net: Clairvoyant Attentive Recurrent Network

We present an interpretable framework for path prediction that leverages dependencies between agents' behaviors and their spatial navigation environment. We exploit two sources of information: the past motion trajectory of the agent of interest and a wide top-view image of the navigation scene. We propose a Clairvoyant Attentive Recurrent Network (CAR-Net) that learns where to look in a large image of the scene when solving the path prediction task. Our method can attend to any area, or combination of areas, within the raw image (e.g., road intersections) when predicting the trajectory of the agent. This allows us to visualize fine-grained semantic elements of navigation scenes that influence the prediction of trajectories. To study the impact of space on agents' trajectories, we build a new dataset made of top-view images of hundreds of scenes (Formula One racing tracks) where agents' behaviors are heavily influenced by known areas in the images (e.g., upcoming turns). CAR-Net successfully attends to these salient regions. Additionally, CAR-Net reaches state-of-the-art accuracy on the standard trajectory forecasting benchmark, Stanford Drone Dataset (SDD). Finally, we show CAR-Net's ability to generalize to unseen scenes.

An Improved Training Procedure for Neural Autoregressive Data Completion

Neural autoregressive models are explicit density estimators that achieve state-of-the-art likelihoods for generative modeling. The D-dimensional data distribution is factorized into an autoregressive product of one-dimensional conditional distributions according to the chain rule. Data completion is a more involved task than data generation: the model must infer missing variables for any partially observed input vector. Previous work introduced an order-agnostic training procedure for data completion with autoregressive models. Missing variables in any partially observed input vector can be imputed efficiently by choosing an ordering where observed dimensions precede unobserved ones and by computing the autoregressive product in this order. In this paper, we provide evidence that the order-agnostic (OA) training procedure is suboptimal for data completion. We propose an alternative procedure (OA++) that reaches better performance in fewer computations. It can handle all data completion queries while training fewer one-dimensional conditional distributions than the OA procedure. In addition, these one-dimensional conditional distributions are trained proportionally to their expected usage at inference time, reducing overfitting. Finally, our OA++ procedure can exploit prior knowledge about the distribution of inference completion queries, as opposed to OA. We support these claims with quantitative experiments on standard datasets used to evaluate autoregressive generative models.

Dataiku's Solution to SPHERE's Activity Recognition Challenge

Our team won the second prize of the Safe Aging with SPHERE Challenge organized by SPHERE, in conjunction with ECML-PKDD and Driven Data. The goal of the competition was to recognize activities performed by humans, using sensor data. This paper presents our solution. It is based on a rich pre-processing and state of the art machine learning methods. From the raw train data, we generate a synthetic train set with the same statistical characteristics as the test set. We then perform feature engineering. The machine learning modeling part is based on stacking weak learners through a grid searched XGBoost algorithm. Finally, we use post-processing to smooth our predictions over time.