Abstract: Innovations in adaptive encoding for units of a video sequence can improve coding efficiency. For example, some of the innovations relate to encoding that includes adaptive switching of color spaces between units within a video sequence. Other innovations relate encoding that includes adaptive switching of color sampling rates between units within a video sequence. Still other innovations relate encoding that includes adaptive switching of bit depths between units within a video sequence.

Abstract: Methods and systems for method for video streaming in a one-way transfer (OWT) system. An example method includes transmitting, from a source computing environment, a video stream for receipt by a destination computing device in a destination computing environment. The video stream is duplicated into a primary video stream and a secondary video stream, both of which are received by a switching device. When the switching device detects an interruption in one of the duplicate data streams, the other stream is transmitted to the destination computing device.

Abstract: Examples of the present disclosure describe systems and methods for UDP file serialization. In examples, a file received at a first device in a OWT system is separated into multiple data chunks. Each of the data chunks is further separated into multiple data segments. Metadata associated with the file is inserted into each of the data chunks and each of the data segments. Data packets that comprise the data segments and compose the data chunks are transmitted to a second device in the OWT system. The second device uses the metadata in the data chunks and the data segments to reconstruct the file. In some examples, data loss mitigation strategies are implemented to mitigate data packet loss and data packet corruption during processing and transmission.

Abstract: A data processing system implements a hybrid environment for interactions between remote and in-person users. The data processing techniques provide tools for facilitating mingling of remote and in-person users in semi-structured interaction, such as but not limited to tradeshows or conferences, and unstructured interactions, such as but not limited to social gatherings that solve the technical problems associated with enabling such systems. The data processing system implements audio porosity and map-based navigation to facilitate improved spatial awareness and awareness of the presence of other remote or in-person users nearby with whom the user can interact.

Abstract: In examples, a user activity recommendation is generated for troubleshooting and/or for improving user understanding of software and/or hardware functionality. In examples, implicit and/or explicit user signals are evaluated to identify a set of actions associated with a task being performed by the user, such that the set of actions are evaluated to determine whether one or more actions can be recommended to the user to facilitate completion of the user's task accordingly. For instance, a recommended action may resolve an issue encountered by the user and/or may enable the user to complete the task more quickly, among other examples. Thus, as a result of generating and providing a user activity recommendation, user frustration may be reduced, user productivity may be increased, technical support may be provided more quickly, and guidance may be provided even in instances where such guidance was not provided by a third-party software developer.

Abstract: Examples are disclosed that relate to front-end image preprocessing. One example provides a computing device comprising an image sensor, an image signal processor, and a front-end preprocessor disposed electrically between the image sensor and image signal processor. The computing device further comprises a storage subsystem comprising instructions executable by a logic subsystem to receive a first image from the image sensor at the preprocessor, preprocess the first image to produce a quantification of first image content, receive a second image, preprocess the second image to produce a quantification of second image content, compare the second quantification to the first quantification to determine a difference, and responsive to the difference satisfying a threshold condition, cause an operating state of a component of the computing device to be changed.

Abstract: A method, system, and storage device storing a computer program, for generating questions based on provided content, such as, for example, a document having words. The method comprises automatically estimating the probability of interesting phrases in the provided content, and generating a question in natural language based on the estimating. In one example embodiment herein, the estimating includes predicting the interesting phrases as answers, and the estimating is performed by a neural model. The method further comprises conditioning a question generation model based on the interesting phrases predicted in the predicting, the question generation model generating the question. The method also can include training the neural model. In one example, the method further comprises identifying start and end locations of the phrases in the provided content, and the identifying includes performing a dot product attention mechanism parameterizing a probability distribution.

Abstract: Methods and systems for transferring data through a one-way transfer (OWT) system using a fault-tolerant data diode. Data from a source computing environment may be converted to an optical signal that is then split into duplicate signals. The duplicate optical signals are received by two receiving devices that have optical receivers. The receiving devices operate as either a primary device or a secondary device to continue transmission of the received data. The operating states of the receiving devices may change based the health or status of the respective devices.

Abstract: In examples, a saliency metric may be determined for an instance of content and/or a location of a digital environment. Accordingly, a set of candidate content and/or locations may be ranked according to associated saliency metrics, such that content and/or an associated location with which to adapt the digital environment for a given user may be determined from a set of candidates. For example, the digital environment may be adapted to present a two-dimensional or a three-dimensional asset to a user. As another example, a game mechanic of the digital environment may be altered. In examples, content from another digital environment may be identified and used to adapt the digital environment, thereby incorporating external content. Thus, a saliency metric associated with a location or an instance of content may be used to programmatically generate a relative or absolute metric with which to evaluate and adapt aspects of the digital environment.

Abstract: An electronic device includes a substrate having a first surface and an opposite second surface; a photonic transmitter supported by the first surface of the substrate; a photonic receiver supported by the first surface of the substrate; a microfluidic volume positioned in the second surface of the substrate; a waveguide positioned to direct photonic signal from the photonic transmitter to the photonic receiver, wherein at least a portion of the waveguide is positioned between the first surface of the substrate and at least a portion of the microfluidic volume; and a working fluid in the microfluidic volume to receive heat from the waveguide.

Abstract: This document relates to distributed devices teleconferencing. Some implementations can employ adaptive microphone selection based on signal characteristics such as signal-to-noise ratios or speech quality, and/or based on a microphone affinity approach. The selected microphone signals can be synchronized and mixed to generate a playback signal that is sent to a remote device. Further implementations can perform proximity-based mixing, where microphone signals received from devices in a particular room can be omitted from playback signals transmitted to other devices in the same room. These techniques can allow enhanced call quality for teleconferencing sessions where co-located users can employ their own devices to participate in a call with other users.

Abstract: Aspects of the technology described herein provide a collaborative browsing experience in which real-time browsing activity and saved browsing activity of session collaborators in a collaborative browsing session are shared with the collaborators. A collaborative session may be initiated, which may create a tab group associated with the session and linked to a collaborator. Other collaborators may be invited to join the session, and additional tab groups for each collaborator may be created. The tab groups of the collaborators may be included in a collective tab group, which may be updated in real-time with changes made by any of the collaborators. For example, client changes may be handled locally and communicated to a service to which each client is connected. The service may sequence and broadcast the ordered changes to the clients, which may each implement the changes according to the sequence to synchronize a shared state amongst clients.

Abstract: A system and method for automatically generating content for a user based on a context includes conducting a search of data associated with the user to generate search results associated with the user, applying first models to the first search results to infer characteristics and patterns associated with the user and to generate a first set of content based on the inferred characteristics and patterns, and generating a first pseudo-document that includes a first set of pattern content, based on the first set of content generated by the one or more first models. Additionally, second models are applied to data associated with the context to infer characteristics and patterns associated with the context and to generate a second set of content based on the inferred characteristics and patterns of the context. A second pseudo-document is generated based on the second set of content.

Abstract: Systems and methods to determine when a media is a high-fidelity reproduction of an original media from a trusted entity are disclosed. In certain aspects, systems and method for generating a fragile watermark are disclosed. The fragile watermark may be inserted into digital media in a manner such that the watermark cannot be identified if the media content is significantly altered. Media content may be subsequently analyzed to determine the presence of a fragile watermark. When the fragile watermark is present, provenance of the media content can be verified and an indication of provenance is provided to the user.

Abstract: In examples, a set of actions performed by a user is identified as an action sequence. If user performance of the same action sequence or similar action sequences exceeds a predetermined threshold, a recommendation to create a macro may be generated. The macro may have one or more associated triggers, such that it may be invoked using voice input or via a user interface, among other examples. A macro may have an associated context in which it applies. In some instances, a trigger used to invoke the macro comprises an indication as to such a context. For example, the macro may be invoked in the context of a document, such that one or more document parts are processed accordingly. As another example, the macro may be invoked to process multiple documents, as may be related in subject matter or associated with the same application.

Abstract: Examples of the present disclosure describe systems and methods for implementing a software-based security abstraction engine in a one-way transfer (OWT) system. In examples, data is received at a first device in the OWT system. A first set of policies is identified based on a dataflow identifier associated with the transfer of the data. A policy engine associated with the first set of policies applies the first set of policies to the data to create digital signatures. The digital signatures are evaluated by the security abstraction engine to determine whether the set of digital signatures is valid. If the digital signatures are determined to be valid, a provenance digital signature is created for the data and a second set of policies is applied to the data. The data is then transmitted to a second device or destination in the OWT system based on the dataflow identifier.

Abstract: Timekeeping on a computing device is deterministically performed by implementing two successive calls to a time function that returns current time based on a continuously running counter that is maintained in one or more cores of the device's CPU. The same fixed time computation parameters are used in each call, with the single variable being a value that is read from the counter. For the initial call to the time function, the processor optimizes the instruction execution by predicting the function's execution path based on observed patterns. As the instructions and data are already cached, and the processor has the results of the prior execution path prediction, the subsequent call executes quickly and predictably relative to the initial call while the processor remains in a working (i.e., non-sleep) state. The series of calls provides a deterministic time computation with improved accuracy by mitigating the unpredictability of processor sleep state delays.

Abstract: Examples of the present disclosure describe systems and methods for an enterprise data container (EDC) that facilitates the secure transfer of data between data boundaries of one or more computing environments. In examples, the EDC serves as a message wrapper for transmitted data. The EDC includes metadata, identification, tracking, security attributes, authenticity, and handling caveats relevant to the operational constraints of one or more computing environments through which data is transferred. The EDC is computing environment agnostic and agnostically manages the data wrapped in the EDC.

Abstract: Aspects of the present disclosure relate to storage policy change usage estimation. In examples, a base storage utilization is generated for a storage object according to a current storage policy. The current storage policy may comprise a RAID (“redundant array of independent disks”) level and/or a fault tolerance level with which the storage object is stored. An estimated storage utilization may then be generated for the storage object based on the base storage utilization and a new storage policy accordingly. Thus, the base storage utilization may omit or otherwise account for overhead associated with the RAID level and/or fault tolerance level indicated by the current storage policy, while the resulting estimated storage utilization may indicate an estimated physical storage utilization for the storage object if the new storage policy is implemented in place of the current storage policy.

Abstract: A method for transforming customer content data to anonymized system metadata includes causing execution of an enterprise application on remote computing systems operated by users associated with multiple enterprises and logging customer content data including data samples corresponding to the users' interactions with the enterprise application. The method includes performing k-aggregation of the data samples by: (a) randomly selecting an enterprise; (b) randomly selecting a user associated with the enterprise; (c) randomly selecting a data sample of the user; (d) repeating (a), (b), and (c) k times, where (a) and (b) are performed without replacement; and aggregating the randomly-requested data samples by position.

Abstract: A laser beam scanning (“LBS”) display device is configured with an optical system that includes a laser beam emitter configured to emit a laser beam. The optical system also includes a driver configured to generate a driving signal for controlling a mirror, such as a microelectromechanical systems (“MEMS”) mirror. The optical system also includes a controller configured to generate a driving signal while rejecting a system disturbance response.

Abstract: Distributed security key management for protecting roaming data via a trusted platform module is performed by systems that include first and second processors, and first and second respective hardware security modules. The first security module encrypts a security key using a public key from the second security module, and the encrypted security key is provided to the second security module. A virtual machine (VM) executed by the first processor has a first virtual security module instance having state data that includes a storage key encrypting VM virtual disk data and that is encrypted with the security key. When a transfer condition is determined, the VM is transferred and executed by the second processor, using a second virtual security module instance, based on decrypting the security key by the second security module using a private key and decrypting the state data for the second virtual security module using the security key.

Abstract: A system includes a development tool for adding electronically-driven effects to a dynamic user-influenced media experience. The development tool is adapted to receive first user input and second user input. The first user input defines an audio trigger corresponding to one or more words or phrases appearing in a textual transcript of an audio content stream that includes audio data to be played as part of the dynamic user-influenced media experience. The second user input defines an event that is to be executed in temporal association with an audible occurrence of the audio trigger. The development tool generates metadata based on the first user input and the second user input, and an application engine interprets the metadata as an instruction to selectively trigger execution of the event in association with the audible occurrence of the audio trigger while presenting the dynamic user-influenced media experience.

Abstract: Systems and methods are provided for tracking a passive controller system using an active sensor system within a mixed-reality environment. The passive controller system includes a body configured to be held in a hand of a user, as well as a plurality of retroreflectors that collectively provides at least 180 degrees of reflecting surface for reflecting a radar signal in at least 180 degrees of spherical range when the passive controller system is positioned within a predetermined distance from a source of the radar signal and with an orientation that is within the at least 180 degrees of spherical range relative to the source of the radar signal. Signals transmitted to the passive controller and reflected back from the passive controller are used to calculate the position and orientation of the passive controller system relative to the active sensor system.

Abstract: A system for implementing object state distributions obtains a first object state distribution associated with a first image set. The first object state distribution includes a first plurality of object state vectors generated based upon (i) a first set of object type classifications associated with the first image set and (ii) a first set of image information based on the first image set. The system obtains a second object state distribution associated with a second image set. The second object state distribution comprises a second plurality of object state vectors generated based upon (i) a second set of object type classifications associated with the second image set and (ii) a second set of image information based on the second image set. The system determines a distance measure between the first object state distribution and the second object state distribution and assigns a label based upon the distance measure.

Abstract: An automated system for resolving program merges uses neural transformers with attention. In one aspect, a neural encoder transformer model is trained from developer-resolved merge conflicts to learn to predict a resolution strategy that aids a developer in constructing a merged program. In a second aspect, a neural decoder transformer model is trained on the syntax and semantics of different source code programming languages to predict a merge resolution consisting of interleaved lines of source code from programs A, B, or O, where programs A and B contain changes to code base O.

Abstract: A system includes a synthesizer unit having a fluid input to receive fluids and a communication input to receive commands to synthesize data-encoded DNA sequences and cleave the DNA. A first flexible chemistry reaction chamber module may be fluidically coupled to the synthesizer unit to receive the data-encoded DNA sequences and amplify the sequences. A deposition unit may be fluidically coupled to the first flexible chemistry reaction chamber module to receive the amplified DNA sequences and encapsulate the amplified DNA sequences into one or more wells in a storage plate for storage and retrieval to and from a plate storage unit. Retrieved DNA may be processed and read by further units.

Abstract: Systems and methods for providing status indicators for various forms of user activity that occurs across different digital contexts of a collaboration platform. A system can monitor activity that a particular user is performing within a particular digital context and provide status indicators to a different user within a different digital context when the monitored activity meets one or more criteria. For example, a system may cause a status indicator to be displayed in association with a data object within the digital context of a message thread when a specific type of user activity is occurring with respect to that data object within the digital context of an application that facilitates editing of the content of the data object. Thus, a system can deliver timely and contextually relevant status indicators about how team members are currently interacting with a data object without users having to switch between digital contexts.

Abstract: The present disclosure relates to systems, methods, and computer-readable media for utilizing a concept graphing system to determine and provide relationships between concepts within document collections or corpora. For example, the concept graphing system can generate and utilize machine-learning models, such as a sparse graph recovery machine-learning model, to identify less-obvious correlations between concepts, including positive and negative concept connections, as well as provide these connections within a visual concept graph. Additionally, the concept graphing system can provide a visual concept graph that determines and displays concept correlations based on the input of a single concept, multiple concepts, or no concepts.

Abstract: The present disclosure relates to systems and methods for automatically associating additional content with sports games. The systems and methods obtain game schedule data for a plurality of sports games, and for each scheduled game, the systems and methods select articles published within a timeframe of the game date of the scheduled game. The systems and methods identify sports articles associated with the sports game based on entities and event phrases extracted from the sports articles matching the schedule game data. The systems and methods classify the sports articles into a plurality of clusters and use the clusters to associate the sports articles to the game schedule data and store the associated sports articles with the game schedule data.

Abstract: In some examples, matrix based bot implementation may include obtaining, for a plurality of bots that are used to respond to a query, a matrix that includes entries including a plurality of scenarios, a plurality of questions corresponding to the plurality of scenarios, and a plurality of responses. Each response may correspond to a specified question. A plurality of scripts may be generated based on an analysis of the matrix. Each script may include at least one question followed by at least one response, and further followed by at least one scenario. For each script, a closest pre-existing script may be identified based on a comparison of the script to pre-existing scripts. For each script, a modification to the matrix may be generated based on a difference in the script from the closest pre-existing script. The bots may be utilized to respond to the query based on the modified matrix.

Abstract: Generally discussed herein are devices, systems, and methods for scan surface reduction in sensitive information scanning. A method can include receiving a document, determining, by an evidence checker, a keyword that indicates sensitive information of a sensitive information type, is present in the document, responsive to determining the keyword is present, determining, by a sensitive information scanner and based on a regular expression associated with the sensitive information type, that sensitive information is present in the document, and performing a sensitive information mitigation operation to mitigate the sensitive information.

Abstract: A method for electronic device identifier assignment at a host computing device includes receiving a current local identifier of a separate electronic device and a full unique identifier of the separate electronic device. A host-specific identifier is assigned to the separate electronic device by, based at least in part on determining that the current local identifier of the separate electronic device is included in a set of local identifier values available for assignment, assigning the host-specific identifier to the separate electronic device with a same local identifier value as the current local identifier. Based at least in part on determining that the current local identifier of the separate electronic device is already assigned by the host computing device to a different separate electronic device, the host-specific local identifier is selected from the set of local identifier values available for assignment and assigned to the separate electronic device.

Abstract: Examples described herein generate training data for machine learning (ML) for natural language (NL) processing (such as semantic parsing for translating NL). A formula tree is generated based on sampling both a formula grammar and NL templates. Using the formula tree, an ML training data instance pair is generated comprising a formula example and an NL example. A context example may also be used during instantiation of the formula tree. An ML model is trained with training data including the ML training data instance pair, and ML output is generated from NL input. The ML output includes, for example, a machine-interpretable formula, a database querying language command, or a general programming language instruction. Some examples support context-free grammar, probabilistic context-free grammar, and/or non-context-free production rules.

Abstract: A relevance system ranks a set of medical studies based on a relevance of each medical study in the set of medical studies to a patient profile. The relevance system includes a relevance model. The relevance model determines a relevance of each medical study to the patient profile based on a semantic relationship score, a concept relationship score, and a term-occurrence score. The semantic relationship score is a measure of a similarity in semantic meaning of a medical study and a patient profile. The concept relationship score is a measure of the closeness of medical concepts in a medical study to medical concepts in a patient profile. The term-occurrence score is a measure of occurrences of terms in a medical study that also appear in a patient profile and the statistical significances of the terms.

Abstract: A computing apparatus comprises one or more computer readable storage media, one or more processors operatively coupled with the one or more computer readable storage media, and program instructions stored on the one or more computer readable storage media. The program instructions, when executed by the one or more processors, direct the computing apparatus to at least generate an audio recording of speech, extract features from the audio recording indicative of vocal patterns in the speech, determine a register classification of the speech based at least on the features, and display an indication of the register classification in a user interface.

Abstract: The present disclosure provides methods and apparatus for optimizing a keyword spotting system. A set of utterance texts including a given keyword may be generated. A set of speech signals corresponding to the set of utterance texts may be synthesized. An acoustic model in the keyword spotting system may be optimized with at least a part of speech signals in the set of speech signals and utterance texts in the set of utterance texts corresponding to the at least a part of speech signals.

Abstract: Technologies pertaining to accounting for pulse history effects are described herein. In connection with accounting for pulse history effects, an amount of time between a first current pulse and a second current pulse that are to be transmitted to a pulsed laser is determined. Based upon such an amount of time, a determination is made as to whether a porch pulse is to be prepended to the second current pulse. When the porch pulse is to be prepended to the second current pulse, an amplitude and duration of the porch pulse are computed based upon the amount of time. The porch pulse is transmitted to the pulsed laser immediately followed by the second current pulse, wherein the porch pulse pre-charges the pulsed laser for emitting a pulse of light based upon the second current pulse.

Abstract: A device includes a controller body, at least one user input device supported by the controller body, a communication device, and a status indicator. The communication device communicates at least one property of the at least one user input device to a selected endpoint connection. The status indicator is located proximate a surface of the body, and the status indicator dynamically indicates the selected endpoint connection of the communication device, where the selected endpoint connection is selected from a plurality of endpoint connections including at least a local electronic device and an access point configured to communicate with a cloud service.

Abstract: A technique for interacting with a machine-trained language model uses dynamic prompt management. The technique includes: receiving an input query and creating prompt information that expresses the input query and targeted context information. The targeted context information is selected from candidate context information. Further, a part of the prompt information is formed by compressing source information by reducing a number of content units in the source information (where the source information includes the input query and/or the candidate context information). The method further includes submitting the prompt information to the machine-trained language model, and receiving a response from the machine-trained language model based on the prompt information. The technique has the overall effect of reducing the number of content units submitted to the language model, which, in turn, reduces the amount of resources and time required by the language model to process the input query.

Abstract: A technique for interacting with a machine-trained language model uses dynamic prompt management. The technique includes: receiving an input query; accessing a state data store that provides candidate context information; partitioning the candidate context information into plural parts; selecting targeted context information from the candidate context information by determining a semantic relevance of the input query to each of the plural parts by performing vector-based analysis; creating prompt information that includes the input query and the targeted context information; submitting the prompt information to the machine-trained language model; and receiving a response from the machine-trained language model based on the prompt information. The technique has the overall effect of reducing the number of content units submitted to the language model, which, in turn, reduces the amount of resources and time required by the language model to process the input query.

Abstract: This document relates to solving challenges associated with solving partial differential equations (PDEs) via numerical simulations relating to natural or physical systems. One example obtains input data relating to a physical system and partitions tensors of a neural network across multiple parallel processors. The example distributes the input data across multiple parallel cloud processing resources for numerical simulations involving partial differential equations to produce corresponding output data. The example trains the neural network across the tensors of the multiple parallel processors with the input data and the output data to produce a surrogate model of the partial differential equations. The example can receive subsequent input data and generate corresponding subsequent output data utilizing the surrogate model.

Abstract: A method for automated device disassembly includes, at a computerized demanufacturing system, receiving a target electronic device for disassembly. Using a set of one or more sensors of the computerized demanufacturing system, a set of sensor data is collected that quantifies one or more physical properties of the target electronic device. Based at least in part on the set of sensor data, correspondences are identified between one or more effectors of the computerized demanufacturing system and one or more interactable elements of the target electronic device. The interactable elements are automatically manipulated using the one or more effectors to at least partially disassemble the target electronic device.

Abstract: Innovations in the use of base color index map (“BCIM”) mode during encoding and/or decoding simplify implementation by reducing the number of modifications made to support BCIM mode and/or improve coding efficiency of BCIM mode. For example, some of the innovations involve reuse of a syntax structure that is adapted for transform coefficients to instead signal data for elements of an index map in BCIM mode. Other innovations relate to mapping of index values in BCIM mode or prediction of elements of an index map in BCIM mode. Still other innovations relate to handling of exception values in BCIM mode.

Abstract: Disclosed herein are innovations for bitstreams having clean random access (CRA) pictures and/or other types of random access point (RAP) pictures. New type definitions and strategic constraints on types of RAP pictures can simplify mapping of units of elementary video stream data to a container format. Such innovations can help improve the ability for video coding systems to more flexibly perform adaptive video delivery, production editing, commercial insertion, and the like.

Abstract: Disclosed herein are innovations for bitstreams having clean random access (CRA) pictures and/or other types of random access point (RAP) pictures. New type definitions and strategic constraints on types of RAP pictures can simplify mapping of units of elementary video stream data to a container format. Such innovations can help improve the ability for video coding systems to more flexibly perform adaptive video delivery, production editing, commercial insertion, and the like.

Abstract: Disclosed herein are innovations for bitstreams having clean random access (CRA) pictures and/or other types of random access point (RAP) pictures. New type definitions and strategic constraints on types of RAP pictures can simplify mapping of units of elementary video stream data to a container format. Such innovations can help improve the ability for video coding systems to more flexibly perform adaptive video delivery, production editing, commercial insertion, and the like.

Abstract: When encoding/decoding a current block of a current picture using intra block copy (“BC”) prediction, the location of a reference block is constrained so that it can be entirely within an inner search area of the current picture or entirely within an outer search area of the current picture, but cannot overlap both the inner search area and the outer search area. In some hardware-based implementations, on-chip memory buffers sample values of the inner search area, and off-chip memory buffers sample values of the outer search area. By enforcing this constraint on the location of the reference block, an encoder/decoder can avoid memory access operations that are split between on-chip memory and off-chip memory when retrieving the sample values of the reference block. At the same time, a reference block close to the current block may be used for intra BC prediction, helping compression efficiency.

Abstract: Disclosed herein are innovations for bitstreams having clean random access (CRA) pictures and/or other types of random access point (RAP) pictures. New type definitions and strategic constraints on types of RAP pictures can simplify mapping of units of elementary video stream data to a container format. Such innovations can help improve the ability for video coding systems to more flexibly perform adaptive video delivery, production editing, commercial insertion, and the like.

Abstract: Disclosed herein are innovations for bitstreams having clean random access (CRA) pictures and/or other types of random access point (RAP) pictures. New type definitions and strategic constraints on types of RAP pictures can simplify mapping of units of elementary video stream data to a container format. Such innovations can help improve the ability for video coding systems to more flexibly perform adaptive video delivery, production editing, commercial insertion, and the like.