Salesforce Com, Inc.

Salesforce, Inc.

Saleforce.com, Inc.

Caiming Xiong

Richard Socher

Yingbo Zhou

Kazuma Hashimoto

Wenhao Liu

Bryan McCann

Ran Xu

Nitish Shirish Keskar

Chu Hong Hoi

James Bradbury

Tian Xie

Victor Zhong

Jia Li

Chien-Sheng Wu

Huan Wang

Xinyi Yang

Govardana Sachithanandam Ramachandran

Jin Qu

Shashank Harinath

Mingfei Gao

Le Xue

Jiasen Lu

Juan Carlos Niebles Duque

Ehsan Hosseini-Asl

Chen Xing

Tong Niu

Feihong Wu

Stephen Merity

Zachary Alexander

Shu Zhang

Wojciech Kryscinski

Alexis Roos

Ankit Chadha

Chetan Ramaiah

Lichao Sun

Silvio Savarese

Xugang Ye

Linqing Liu

Teresa Anna Marshall

Raffaella Buschiazzo

Junnan Li

Na Cheng

Sitaram Asur

Semih Yavuz

Weiran Wang

Romain Paulus

Ka Chun Au

Xi Victoria Lin

Prafulla Kumar Choubey

Alexander Richard Trott

Alexander R Fabbri

Zeyuan Chen

Bo Pang

Wenzhuo Yang

Ning Yu

Wenpeng Yin

Erik Lennart Nijkamp

Markus Anderle

Chenxi Li

Yongjun Chen

Lily Hu

Harshavardhan Utharavalli

Congying Xia

Chih-Yao Ma

Stephen Joseph Merity

Kai Sheng Tai

Lav Raj Varshney

Luowei Zhou

Akari Asai

Evan Albright

Hung Le

Roojuta Lalani

Nazneen Rajani

Jean-Marc Soumet

Thomas Archie Cook, Jr

Shiva Kumar Pentyala

Manju Vijayakumar

Yixin Mao

Jesse Vig

Stephan Zheng

Michael Machado

Johnson Liu

Ankit Kumar

Victor Winslow Yee

Soujanya Lanka

Victor Yee

Roberto Martín-Martín

Juan Rodriguez

Hao Liu

Huishuai Zhang

Michael Shum

Simo Arajarvi

Xilai Li

Tong Che

Peng Tang

Xiang Chen

Karl Ryszard Skucha

James Douglas Harrison

Embodiments are directed to pre-training a transformer model using more parameters for sophisticated patterns (PSP++). The transformer model is divided into a held-out model and a main model. A forward pass and a backward pass are performed on the held-out model, where the forward pass determines self-attention hidden states of the held-out model and the backward pass determines loss of the held-out model. A forward pass on the main model is performed to determine a self-attention hidden states of the main model. The self-attention hidden states of the main model are concatenated with the self-attention hidden states of the held-out model. A backward pass is performed on the main model to determine a loss of the main model. The parameters of the held-out model are updated to reflect the loss of the held-out model and parameters of the main model are updated to reflect the loss of the main model.

Parameter utilization for language pre-training

Embodiments described herein provide a training framework for generative NLP models that operate on previously learnt knowledge from pretrained large language models. Specifically, to train an NLP model to generate a response to a user utterance (e.g., 'resolve login issue'), document embeddings of support IT documents encoded by a pretrained LLM are fed to an NLP decoder together with a training dialogue (e.g., a dialogue between the chat agent on how to “resolve login issue”). The NLP decoder can thus be trained by a causal language modeling loss computed based on the predicted next token and the ground-truth token from the training dialogue.

Systems and methods for training a neural network model using knowledge from pre-trained large language models

Embodiments described herein provides an intent prototypical contrastive learning framework that leverages intent similarities between users with different behavior sequences. Specifically, user behavior sequences are encoded into a plurality of user interest representations. The user interest representations are clustered into a plurality of clusters based on mutual distances among the user interest representations in a representation space. Intention prototypes are determined based on centroids of the clusters. A set of augmented views for user behavior sequences are created and encoded into a set of view representations. A contrastive loss is determined based on the set of augmented views and the plurality of intention prototypes. Model parameters are updated based at least in part on the contrastive loss.

Systems and methods for sequential recommendation

A method of training a neural network based three-dimensional (3D) encoder is provided. A first plurality of samples of a training dataset are generated using a first 3D model. An image generator with multi-view rendering is used to generate a plurality of two-dimensional (2D) images having different viewpoints of the first 3D model. A first language model is used to generate a plurality of texts corresponding to the plurality of 2D images respectively. A first text for a first image is generated by using one or more text descriptions generated by the first language model. A point cloud is generated by randomly sampling points in the 3D model. The first plurality of samples are generated using the plurality of 2D images, the corresponding plurality of texts, and the point cloud. The neural network based 3D encoder is trained using the training dataset including the first plurality of samples.

Systems and methods for multimodal pretraining for three-dimensional understanding models

Systems and methods for training a neural network based three-dimensional (3D) encoder for 3D classification are provided. A training dataset including a plurality of samples is received, wherein a first sample includes an image, a text, and a point cloud. An image encoder of a pretrained vision and language model is used to generate image representations for the image of the first sample. A text encoder of the pretrained vision and language model is used to generate text representations for the text of the first sample. The neural network based 3D encoder is used to generate 3D representations for the point cloud of the first sample. A loss objective is computed based on the image representations, text representations, and 3D representations. Parameters of the neural network based 3D encoder are updated based on the computed loss objective via backpropagation.

Systems and methods for learning unified representations of language, image, and point cloud for three-dimensional recognition

A method of training a neural network based three-dimensional (3D) encoder is provided. A training dataset is generated using a plurality of 3D models of a 3D model dataset. To generate a first sample of the training dataset, an image generator with multi-view rendering is used to generate a plurality of image candidates of a first 3D model. A word is chosen from metadata associated with the first 3D model. A language model is used to generate one or more text descriptions using the selected word and a plurality of prompts. A point cloud is generated by randomly sampling points in the 3D model. The first sample is generated to include a first image randomly selected from the plurality of image candidates, one or more text descriptions, and the point cloud is generated. The 3D encoder is trained using the training dataset including the first sample.

Embodiments described herein provide a video generation framework built on a decoupled multimodal cross-attention module to simultaneously condition the generation on both an input image and a text input. The video generation may thus be conditioned on the visual appearance of a target object reflected in the input image. In this way, zero-shot video generation may be achieved with little fine-tuning efforts.

Systems and methods for controllable video generation

A computing device may perform training of a set of machine learning models on a first data set associated with a first domain. In some examples, the training may include, for each machine learning model of the set of machine learning models, inputting, as values for a set of parameters of the respective sets of parameters and for an iteration of a set of iterations, a moving average of the set of parameters calculated over a threshold number of previous iterations. The computing device may select a set of model states that are generated during the training of the plurality of machine learning models based on a validation performance of the set of model states performed during the training. The computing device may then generate an ensembled machine learning model by aggregating the set of machine learning models corresponding to the set of selected model states.

Techniques for machine learning model selection for domain generalization

Embodiments described herein provide a two-stage model-agnostic approach for generating counterfactual explanation via counterfactual feature selection and counterfactual feature optimization. Given a query instance, counterfactual feature selection picks a subset of feature columns and values that can potentially change the prediction and then counterfactual feature optimization determines the best feature value for the selected feature as a counterfactual example.

Systems and methods for counterfactual explanation in machine learning models

Embodiments described herein provide a prompt-based transfer learning method that employs shared latent space prompt tuning). Specifically, a shared latent space is assumed, among all source and target tasks, where each vector in the space captures a basis skill to do a particular task. Given an instance (from either a source task or a target task), it is first encoded into an instance representation vector and then queries the latent space, which yields a skill vector for this instance. This vector modulates a frozen model, via soft prompts which are a simple prompt transformation (the prompt generator in FIG.) of the basis skill vector, to generate an answer for the instance. The latent space and prompt transformation are learned end-to-end in upstream pre-training on source tasks.

Systems and methods for shared latent space prompt tuning

Embodiments described herein provide a structured conversation summarization framework. A user interface may be provided which allows an agent to perform a conversation with a customer, for example regarding resolving a customer support issue. Utterances by both the agent and customer may be stored, and at the end of the conversation, the utterances may be used to generate a structured summary. The structured summary may include components such as a general summary, an issue summary, and a resolution summary. Using neural network models and heuristics, each component of the summary may be automatically generated.

Systems and methods for structured conversation summarization

Embodiments described herein provide a system for three-dimensional (3D) object detection. The system includes an input interface configured to obtain 3D point data describing spatial information of a plurality of points, and a memory storing a neural network based 3D object detection model having an encoder and a decoder. The system also includes processors to perform operations including: encoding, by the encoder, a first set of coordinates into a first set of point features and a set of object features; sampling a second set of point features from the first set of point features; generating, by attention layers at the decoder, a set of attention weights by applying cross-attention over at least the set of object features and the second set of point feature, and generate, by the decoder, a predicted bounding box among the plurality of points based on at least in part on the set of attention weights.

Systems and methods for attention mechanism in three-dimensional object detection

Some embodiments of the current disclosure disclose methods and systems for training for training a natural language processing intent classification model to perform few-shot classification tasks. In some embodiments, a pair of an utterance and a first semantic label labeling the utterance may be generated and a neural network that is configured to perform natural language inference tasks may be utilized to determine the existence of an entailment relationship between the utterance and the semantic label. The semantic label may be predicted as the intent class of the utterance based on the entailment relationship and the pair may be used to train the natural language processing intent classification model to perform few-shot classification tasks.

Systems and methods for few-shot intent classifier models

Embodiments are directed to a machine learning recommendation system. The system receives a user query for generating a recommendation for one or more items with an explanation associated with recommending the one or more items. The system obtains first features of at least one user and second features of a set of items. The system provides the first features and the second features to a first machine learning network for determining a predicted score for an item. The system provides a portion of the first features and a portion of the second features to second machine learning networks for determining explainability scores for an item and generating corresponding explanation narratives. The system provides the recommendation for one or more items and corresponding explanation narratives based on ranking predicted scores and explainability scores for the items.

Diversity and explainability parameters for recommendation accuracy in machine learning recommendation systems

Embodiments described herein provide systems and methods for a partially supervised training model for online action detection. Specifically, the online action detection framework may include two modules that are trained jointly—a Temporal Proposal Generator (TPG) and an Online Action Recognizer (OAR). In the training phase, OAR performs both online per-frame action recognition and start point detection. At the same time, TPG generates class-wise temporal action proposals serving as noisy supervisions for OAR. TPG is then optimized with the video-level annotations. In this way, the online action detection framework can be trained with video-category labels only without pre-annotated segment-level boundary labels.

Systems and methods for partially supervised online action detection in untrimmed videos

Growing community of inventors

Palo Alto, CA, United States of America

Caiming Xiong