5 citations · 20 across the 12 of their papers we have counts for
13 papers
Fundamental Problems With Model Editing: How Should Rational Belief Revision Work in LLMs?
Peter Hase, Thomas Hofweber, Xiang Zhou +2
The model editing problem concerns how language models should learn new facts about the world over time. While empirical research on model editing has drawn widespread attention, t…
Inducing Systematicity in Transformers by Attending to Structurally Quantized Embeddings
Yichen Jiang, Xiang Zhou, Mohit Bansal
Transformers generalize to novel compositions of structures and entities after being trained on a complex dataset, but easily overfit on datasets of insufficient complexity. We obs…
Data Factors for Better Compositional Generalization
Xiang Zhou, Yichen Jiang, Mohit Bansal
Recent diagnostic datasets on compositional generalization, such as SCAN (Lake and Baroni, 2018) and COGS (Kim and Linzen, 2020), expose severe problems in models trained from scra…
ReCEval: Evaluating Reasoning Chains via Correctness and Informativeness
Archiki Prasad, Swarnadeep Saha, Xiang Zhou +1
Multi-step reasoning ability is fundamental to many natural language tasks, yet it is unclear what constitutes a good reasoning chain and how to evaluate them. Most existing method…
Mutual Exclusivity Training and Primitive Augmentation to Induce Compositionality
Yichen Jiang, Xiang Zhou, Mohit Bansal
Recent datasets expose the lack of the systematic generalization ability in standard sequence-to-sequence models. In this work, we analyze this behavior of seq2seq models and ident…
Masked Part-Of-Speech Model: Does Modeling Long Context Help Unsupervised POS-tagging?
Xiang Zhou, Shiyue Zhang, Mohit Bansal
Previous Part-Of-Speech (POS) induction models usually assume certain independence assumptions (e.g., Markov, unidirectional, local dependency) that do not hold in real languages.…
GrIPS: Gradient-free, Edit-based Instruction Search for Prompting Large Language Models
Archiki Prasad, Peter Hase, Xiang Zhou +1
Providing natural language instructions in prompts is a useful new paradigm for improving task performance of large language models in a zero-shot setting. Recent work has aimed to…
Hidden Biases in Unreliable News Detection Datasets
Xiang Zhou, Heba Elfardy, Christos Christodoulopoulos +2
Automatic unreliable news detection is a research problem with great potential impact. Recently, several papers have shown promising results on large-scale news datasets with model…
Distributed NLI: Learning to Predict Human Opinion Distributions for Language Reasoning
Xiang Zhou, Yixin Nie, Mohit Bansal
We introduce distributed NLI, a new NLU task with a goal to predict the distribution of human judgements for natural language inference. We show that by applying additional distrib…
What Can We Learn from Collective Human Opinions on Natural Language Inference Data?
Yixin Nie, Xiang Zhou, Mohit Bansal
Despite the subjective nature of many NLP tasks, most NLU evaluations have focused on using the majority label with presumably high agreement as the ground truth. Less attention ha…
Deep Reinforcement Learning for On-line Dialogue State Tracking
Zhi Chen, Lu Chen, Xiang Zhou +1
Dialogue state tracking (DST) is a crucial module in dialogue management. It is usually cast as a supervised training problem, which is not convenient for on-line optimization. In…
Towards Robustifying NLI Models Against Lexical Dataset Biases
Xiang Zhou, Mohit Bansal
While deep learning models are making fast progress on the task of Natural Language Inference, recent studies have also shown that these models achieve high accuracy by exploiting…
The Curse of Performance Instability in Analysis Datasets: Consequences, Source, and Suggestions
Xiang Zhou, Yixin Nie, Hao Tan +1
We find that the performance of state-of-the-art models on Natural Language Inference (NLI) and Reading Comprehension (RC) analysis/stress sets can be highly unstable. This raises…