Search Results for author:
Found 16 papers, 10 papers with code
A Self-supervised Joint Training Framework for Document Reranking
Pretrained language models such as BERT have been successfully applied to a wide range of natural language processing tasks and also achieved impressive performance in document reranking tasks.
StableToolBench: Towards Stable Large-Scale Benchmarking on Tool Learning of Large Language Models
The virtual API server contains a caching system and API simulators which are complementary to alleviate the change in API status.
DecisionNCE: Embodied Multimodal Representations via Implicit Preference Learning
Multimodal pretraining has emerged as an effective strategy for the trinity of goals of representation learning in autonomous robots: 1) extracting both local and global task progression information; 2) enforcing temporal consistency of visual representation; 3) capturing trajectory-level language grounding.
DEEM: Dynamic Experienced Expert Modeling for Stance Detection
Recent work has made a preliminary attempt to use large language models (LLMs) to solve the stance detection task, showing promising results.
Speak It Out: Solving Symbol-Related Problems with Symbol-to-Language Conversion for Language Models
We propose symbol-to-language (S2L), a tuning-free method that enables large language models to solve symbol-related problems with information expressed in natural language.
EgoThink: Evaluating First-Person Perspective Thinking Capability of Vision-Language Models
However, the capability of VLMs to "think" from a first-person perspective, a crucial attribute for advancing autonomous agents and robotics, remains largely unexplored.
OpenChat: Advancing Open-source Language Models with Mixed-Quality Data
Specifically, we consider the general SFT training data, consisting of a small amount of expert data mixed with a large proportion of sub-optimal data, without any preference labels.
Ranked #22 on
Code Generation
on HumanEval
Prompt-Guided Retrieval Augmentation for Non-Knowledge-Intensive Tasks
There are two main challenges to leveraging retrieval-augmented methods for NKI tasks: 1) the demand for diverse relevance score functions and 2) the dilemma between training cost and task performance.
Modeling Adversarial Attack on Pre-trained Language Models as Sequential Decision Making
Pre-trained language models (PLMs) have been widely used to underpin various downstream tasks.
Say What You Mean! Large Language Models Speak Too Positively about Negative Commonsense Knowledge
Large language models (LLMs) have been widely studied for their ability to store and utilize positive knowledge.
Unsupervised Explanation Generation via Correct Instantiations
The major difficulty is finding the conflict point, where the statement contradicts our real world.
Learning What You Need from What You Did: Product Taxonomy Expansion with User Behaviors Supervision
Specifically, i) to fully exploit user behavioral information, we extract candidate hyponymy relations that match user interests from query-click concepts; ii) to enhance the semantic information of new concepts and better detect hyponymy relations, we model concepts and relations through both user-generated content and structural information in existing taxonomies and user click logs, by leveraging Pre-trained Language Models and Graph Neural Network combined with Contrastive Learning; iii) to reduce the cost of dataset construction and overcome data skews, we construct a high-quality and balanced training dataset from existing taxonomy with no supervision.
Can Pre-trained Language Models Interpret Similes as Smart as Human?
In this paper, we investigate the ability of PLMs in simile interpretation by designing a novel task named Simile Property Probing, i. e., to let the PLMs infer the shared properties of similes.
Unsupervised Editing for Counterfactual Stories
We also propose a new metric to alleviate the shortcomings of current automatic metrics and better evaluate the trade-off.
FedGEMS: Federated Learning of Larger Server Models via Selective Knowledge Fusion
Today data is often scattered among billions of resource-constrained edge devices with security and privacy constraints.
On Commonsense Cues in BERT for Solving Commonsense Tasks
We quantitatively investigate the presence of structural commonsense cues in BERT when solving commonsense tasks, and the importance of such cues for the model prediction.
