Abstract: Systems and methods are provided for developing/updating training datasets for traffic light detection/perception models. V2I-based information may indicate a particular traffic light state/state of transition. This information can be compared to a traffic light perception prediction. When the prediction is inconsistent with the V2I-based information, data regarding the condition(s)/traffic light(s)/etc. can be saved and uploaded to a training database to update/refine the training dataset(s) maintained therein. In this way, an existing traffic light perception model can be updated/improved and/or a better traffic light perception model can be developed.

Abstract: A novel electrical connector and method of manufacture is disclosed which provides an integral attachment and retention means for the purpose of electrically and mechanically interconnecting circuit elements in electronic devices, said circuit elements including but not limited to printed circuit boards, flexible printed circuits, rigid flex circuits, semiconductor package substrates, modules, and batteries. The electrical connector of the present invention utilizes a bonding material, disposed at least between the electrical spring contact elements on a surface of the connector, to bond and retain first and second portions of the electrical connector in an actuated state on a mating circuit element whereby stable and low resistance electrical interconnections are formed and maintained between the electrical connector and interconnection terminals on the mating circuit element. This design permits the electrical connector to be low-profile and use a reduced amount of space on a circuit member such as a PCB.

Abstract: A novel interconnection structure and method of manufacture is provided which provides an improved means of interconnecting external connector interfaces, such as Universal Serial Bus (USB) connectors, to the internal system boards of electronic devices, such as laptop computers, tablets, and mobile phones. An external connector interface used for interconnecting separate electronic devices is connected to the internal system board of the device in which it resides by being interconnected mechanically and electrically, or alternatively being integral to and of a unitary structure with, a printed circuit substrate, said printed circuit substrate having a plurality of conductive, elastic spring contacts mounted on one surface, with at least one of said electrical spring contacts electrically interconnected to the external electrical connections of the USB connector, and said electrical spring contacts providing an electrical interconnection means to a system board inside the electronic device.

Abstract: A novel, low profile connector element is disclosed for the purpose of electrically and mechanically interconnecting circuit elements in electronic devices, said circuit elements including but not limited to printed circuit boards, flexible printed circuits, rigid flex circuits, semiconductors, semiconductor package substrates, ground shields, and batteries, whereby the connector includes electrical contacts which have a unitary structure consisting of at least a distal end, a proximal end, and a middle section between the distal and proximal ends. The contacts of the present invention exhibit a contact diametric true position with respect to one another in an array of less than 0.2 millimeters. The electrical contacts are created in batch form from a high conductivity sheet of spring material such as a copper alloy.

Abstract: The subject disclosure generally describes a technology by which text and/or speech descriptions are collected by showing a stimulus such as video clips to contributors (e.g., of a crowd-sourcing service). The descriptions, which are in the language of each contributor's choice, are of the same stimulus and thus associated with one another. While each contributor may be monolingual, the technique allows for the collection of approximately bilingual data, since more than one language may be represented among the different contributors. The descriptions may be used as translation data for training a machine translation engine, and as paraphrase data (grouped by the same language) for training a machine paraphrasing system. Also described is evaluating the quality of a machine paraphrasing system via a distinctiveness metric.

Abstract: The current invention provides methods for performing a cleaning process that provides greater cleaning efficiency with less damage to device structures. After etching and photoresist stripping, a first plasma clean is performed. The first plasma clean may comprise one or more steps. Following the first plasma clean, a first HO based clean is performed. The first HO based clean may be a de-ionized water rinse, a water vapor clean, or a plasma clean, where the plasma includes hydrogen and oxygen. Following the first HO based clean, a second plasma clean is performed, which may comprise one or more steps. A second HO based clean follows the second plasma clean, and may be a de-ionized water rinse, a water vapor clean, or a plasma clean, where the plasma includes hydrogen and oxygen. For plasma processes, an RF, generated plasma, a microwave generated plasma, an inductively coupled plasma, or combination may be used.

Abstract: The current invention provides methods for performing a cleaning process that provides greater cleaning efficiency with less damage to device structures. After etching and photoresist stripping, a first plasma clean is performed. The first plasma clean may comprise one or more steps. Following the first plasma clean, a first HO based clean is performed. The first HO based clean may be a de-ionized water rinse, a water vapor clean, or a plasma clean, where the plasma includes hydrogen and oxygen. Following the first HO based clean, a second plasma clean is performed, which may comprise one or more steps. A second HO based clean follows the second plasma clean, and may be a de-ionized water rinse, a water vapor clean, or a plasma clean, where the plasma includes hydrogen and oxygen. For plasma processes, an RF generated plasma, a microwave generated plasma, an inductively coupled plasma, or combination may be used.

Abstract: The current invention provides methods for performing a cleaning process that provides greater cleaning efficiency with less damage to device structures. After etching and photoresist stripping, a first plasma clean is performed. The first plasma clean may comprise one or more steps. Following the first plasma clean, a first HO based clean is performed. The first HO based clean may be a de-ionized water rinse, a water vapor clean, or a plasma clean, where the plasma includes hydrogen and oxygen. Following the first HO based clean, a second plasma clean is performed, which may comprise one or more steps. A second HO based clean follows the second plasma clean, and may be a de-ionized water rinse, a water vapor clean, or a plasma clean, where the plasma includes hydrogen and oxygen. For plasma processes, an RF generated plasma, a microwave generated plasma, an inductively coupled plasma, or combination may be used.

Abstract: A unique photoresist strip sequence using a downstream plasma system is described. The sequence can include a RF directional plasma alone, downstream plasma alone or combine both RF plasma and downstream plasma together. The process sequence can be a single step or multiple steps, which produce high strip rates while maintaining the dielectric properties of the film. The process can be an oxidizing process carried out at low temperature and low pressure, which reduces the reactivity of the oxygen with the low-k film. Furthermore, by adding a small percentage of an additive gas, such as a fluorine-containing gas, to the plasma, the inorganic residues from the strip process are removed, leaving a clean film cleared of photoresist and residue.