Abstract: Systems, methods and non-transitory computer readable media for prompt based background replacement are provided. A visual content including a background portion and at least one foreground object may be accessed. Further, a textual input indicative of a desire of an individual to modify the visual content may be received. The textual input and the visual content may be analyzed to generate a modified version of the visual content. The modified version may differ from the visual content in the background portion. Further, the modified version may include a depiction of the at least one foreground object substantially identical to a depiction of the at least one foreground object in the visual content. Further, a presentation of the modified version of the visual content to the individual may be caused.

Abstract: Implementations are directed to generating an edited synthetic image, that corresponds to a real image captured using a real camera, but that is generated based on, and reflects, edit(s) to an original natural language (NL) caption for the real image. For example, the original NL caption can be used in performing an inversion on the real image to generate a diffusion trajectory for the real image. Further, the diffusion trajectory can be used to optimize a sequence of unconditional embeddings, for the real image, that are not based on the NL caption for the real image. Yet further, the edited NL caption, the unconditional embeddings, and at least part of a noise vector (of the diffusion trajectory) can be processed, using a Large-scale language-image (LLI) model, to generate the edited synthetic image.

Abstract: Systems, methods and non-transitory computer readable media for chained attribution of generated content are provided. A generative model may be accessed. The generative model may be a result of training a machine learning model using a plurality of training examples. Each training example of the plurality of training examples may be associated with a respective content. An input in a natural language may be received. At least one auxiliary content may be obtained. The generative model may be used to analyze the input and the at least one auxiliary content to generate a new content, the generated new content is based on the at least one auxiliary content and at least one of the plurality of training examples. A degree of attribution of the generated new content to the at least one auxiliary content may be determined. Information based on the degree of attribution may be provided.

Abstract: Systems, methods and non-transitory computer readable media for generating audio clips from listening habits are provided. In some examples, a generative model may be accessed. Further, a plurality of audio clips may be accessed. For each audio clip of the plurality of audio clips, the respective audio clip may be analyzed to identify a musical property of the respective audio clip. The generative model and the identified musical properties may be used to generate a new audio clip. The new audio clip may be provided.

Abstract: Systems, methods and non-transitory computer readable media for attributing generated textual contents to training examples are provided. A first textual content generated using a generative model may be received. The generative model may be a result of training a machine learning model using a plurality of training examples. Each training example of the plurality of training examples may be associated with a respective textual content. Properties of the first textual content may be determined. For each training example of the plurality of training examples, properties of the respective textual content may be determined. The properties of the first textual content and the properties of the textual contents associated with the plurality of training examples may be used to attribute the first textual content to a first subgroup of at least one but not all of the plurality of training examples.

Abstract: Some implementations are directed to editing a source image, where the source image is one generated based on processing a source natural language (NL) prompt using a Large-scale language-image (LLI) model. Those implementations edit the source image based on user interface input that indicates an edit to the source NL prompt, and optionally independent of any user interface input that specifies a mask in the source image and/or independent of any other user interface input. Some implementations of the present disclosure are additionally or alternatively directed to applying prompt-to-prompt editing techniques to editing a source image that is one generated based on a real image, and that approximates the real image.

Abstract: Some implementations are directed to editing a source image, where the source image is one generated based on processing a source natural language (NL) prompt using a Large-scale language-image (LLI) model. Those implementations edit the source image based on user interface input that indicates an edit to the source NL prompt, and optionally independent of any user interface input that specifies a mask in the source image and/or independent of any other user interface input. Some implementations of the present disclosure are additionally or alternatively directed to applying prompt-to-prompt editing techniques to editing a source image that is one generated based on a real image, and that approximates the real image.

Abstract: Systems, methods and non-transitory computer readable media for generating content using a generative model without relying on selected training examples are provided. An input indicative of a desire to generate a new content using a generative model may be received. The generative model may be a result of training a machine learning model using a plurality of training examples. Each training example of the plurality of training examples may be associated with a respective content. Further, an indication of a particular subgroup of at least one but not all of the plurality of training examples may be obtained. Based on the indication, the input and the generative model may be used to generate the new content, abstaining from basing the generation of the new content on any training example included in the particular subgroup. The new content may be provided.

Abstract: Systems, methods and non-transitory computer readable media for generating content using a generative model without relying on selected training examples are provided. An input indicative of a desire to generate a new content using a generative model may be received. The generative model may be a result of training a machine learning model using a plurality of training examples. Each training example of the plurality of training examples may be associated with a respective content. Further, an indication of a particular subgroup of at least one but not all of the plurality of training examples may be obtained. Based on the indication, the input and the generative model may be used to generate the new content, abstaining from basing the generation of the new content on any training example included in the particular subgroup. The new content may be provided.

Abstract: Systems, methods and non-transitory computer readable media for attributing aspects of generated visual contents to training examples are provided. A first visual content generated using a generative model may be received. The generative model may be a result of training a machine learning model using a plurality of training examples. Properties of an aspect of the first visual content and properties of visual contents associated with the plurality of training examples may be used to attribute the aspect of the first visual content to a subgroup of the plurality of training examples. For each source of the sources associated with the visual contents associated with the training examples of the subgroup, a data-record associated with the source may be updated based on the attribution of the aspect of the first visual content.

Abstract: Systems, methods and non-transitory computer readable media for attributing generated audio contents to training examples are provided. A first audio content generated using a generative model may be received. The generative model may be a result of training a machine learning model using training examples. Each training example may be associated with a respective audio content. Properties of the first audio content may be determined. For each training example of the training examples, the respective audio content may be analyzed to determine properties of the respective audio content. The properties of the first audio content and the properties of the audio contents associated with the training examples may be used to attribute the first audio content to a subgroup of the training examples. A respective data-record associated with a source associated with the training examples of the subgroup may be updated based on the attribution.

Abstract: Systems, methods and non-transitory computer readable media for attributing aspects of generated visual contents to training examples are provided. A first visual content generated using a generative model may be received. The generative model may be a result of training a machine learning model using a plurality of training examples. Properties of an aspect of the first visual content and properties of visual contents associated with the plurality of training examples may be used to attribute the aspect of the first visual content to a subgroup of the plurality of training examples. For each source of the sources associated with the visual contents associated with the training examples of the subgroup, a data-record associated with the source may be updated based on the attribution of the aspect of the first visual content.

Abstract: Systems, methods and non-transitory computer readable media for prompt based background replacement are provided. A visual content including a background portion and at least one foreground object may be accessed. Further, a textual input indicative of a desire of an individual to modify the visual content may be received. The textual input and the visual content may be analyzed to generate a modified version of the visual content. The modified version may differ from the visual content in the background portion. Further, the modified version may include a depiction of the at least one foreground object substantially identical to a depiction of the at least one foreground object in the visual content. Further, a presentation of the modified version of the visual content to the individual may be caused.

Abstract: Systems, methods and non-transitory computer readable media for attributing generated audio contents to training examples are provided. A first audio content generated using a generative model may be received. The generative model may be a result of training a machine learning model using training examples. Each training example may be associated with a respective audio content. Properties of the first audio content may be determined. For each training example of the training examples, the respective audio content may be analyzed to determine properties of the respective audio content. The properties of the first audio content and the properties of the audio contents associated with the training examples may be used to attribute the first audio content to a subgroup of the training examples. A respective data-record associated with a source associated with the training examples of the subgroup may be updated based on the attribution.

Abstract: Systems, methods and non-transitory computer readable media for attributing aspects of generated visual contents to training examples are provided. A first visual content generated using a generative model may be received. The generative model may be a result of training a machine learning model using a plurality of training examples. Properties of an aspect of the first visual content and properties of visual contents associated with the plurality of training examples may be used to attribute the aspect of the first visual content to a subgroup of the plurality of training examples. For each source of the sources associated with the visual contents associated with the training examples of the subgroup, a data-record associated with the source may be updated based on the attribution of the aspect of the first visual content.

Abstract: Systems, methods and non-transitory computer readable media for generating content using a generative model without relying on selected training examples are provided. An input indicative of a desire to generate a new content using a generative model may be received. The generative model may be a result of training a machine learning model using a plurality of training examples. Each training example of the plurality of training examples may be associated with a respective content. Further, an indication of a particular subgroup of at least one but not all of the plurality of training examples may be obtained. Based on the indication, the input and the generative model may be used to generate the new content, abstaining from basing the generation of the new content on any training example included in the particular subgroup. The new content may be provided.

Abstract: Some implementations are directed to editing a source image, where the source image is one generated based on processing a source natural language (NL) prompt using a Large-scale language-image (LLI) model. Those implementations edit the source image based on user interface input that indicates an edit to the source NL prompt, and optionally independent of any user interface input that specifies a mask in the source image and/or independent of any other user interface input. Some implementations of the present disclosure are additionally or alternatively directed to applying prompt-to-prompt editing techniques to editing a source image that is one generated based on a real image, and that approximates the real image.