10 citations · 21 across the 8 of their papers we have counts for
9 papers
Beyond Single-shot Writing: Deep Research Agents are Unreliable at Multi-turn Report Revision
Bingsen Chen, Boyan Li, Ping Nie +3
Existing benchmarks for Deep Research Agents (DRAs) treat report generation as a single-shot writing task, which fundamentally diverges from how human researchers iteratively draft…
Likert or Not: LLM Absolute Relevance Judgments on Fine-Grained Ordinal Scales
Charles Godfrey, Ping Nie, Natalia Ostapuk +3
Large language models (LLMs) obtain state of the art zero shot relevance ranking performance on a variety of information retrieval tasks. The two most common prompts to elicit LLM…
Nearest Neighbor Multivariate Time Series Forecasting
Huiliang Zhang, Ping Nie, Lijun Sun +1
Multivariate time series (MTS) forecasting has a wide range of applications in both industry and academia. Recently, spatial-temporal graph neural networks (STGNNs) have gained pop…
Beyond Prompting: Making Pre-trained Language Models Better Zero-shot Learners by Clustering Representations
Yu Fei, Ping Nie, Zhao Meng +2
Recent work has demonstrated that pre-trained language models (PLMs) are zero-shot learners. However, most existing zero-shot methods involve heavy human engineering or complicated…
Anticipating the Unseen Discrepancy for Vision and Language Navigation
Yujie Lu, Huiliang Zhang, Ping Nie +4
Vision-Language Navigation requires the agent to follow natural language instructions to reach a specific target. The large discrepancy between seen and unseen environments makes i…
MIC: Model-agnostic Integrated Cross-channel Recommenders
Yujie Lu, Ping Nie, Shengyu Zhang +4
Semantically connecting users and items is a fundamental problem for the matching stage of an industrial recommender system. Recent advances in this topic are based on multi-channe…
Answering Any-hop Open-domain Questions with Iterative Document Reranking
Ping Nie, Yuyu Zhang, Arun Ramamurthy +1
Existing approaches for open-domain question answering (QA) are typically designed for questions that require either single-hop or multi-hop reasoning, which make strong assumption…
A Pairwise Probe for Understanding BERT Fine-Tuning on Machine Reading Comprehension
Jie Cai, Zhengzhou Zhu, Ping Nie +1
Pre-trained models have brought significant improvements to many NLP tasks and have been extensively analyzed. But little is known about the effect of fine-tuning on specific tasks…
DC-BERT: Decoupling Question and Document for Efficient Contextual Encoding
Yuyu Zhang, Ping Nie, Xiubo Geng +3
Recent studies on open-domain question answering have achieved prominent performance improvement using pre-trained language models such as BERT. State-of-the-art approaches typical…