Alert button
Picture for Trevor Cohn

Trevor Cohn

Alert button

Pre-training Cross-lingual Open Domain Question Answering with Large-scale Synthetic Supervision

Feb 26, 2024
Fan Jiang, Tom Drummond, Trevor Cohn

Viaarxiv icon

Simpson's Paradox and the Accuracy-Fluency Tradeoff in Translation

Feb 20, 2024
Zheng Wei Lim, Ekaterina Vylomova, Trevor Cohn, Charles Kemp

Viaarxiv icon

Predicting Human Translation Difficulty with Neural Machine Translation

Dec 19, 2023
Zheng Wei Lim, Ekaterina Vylomova, Charles Kemp, Trevor Cohn

Viaarxiv icon

Boot and Switch: Alternating Distillation for Zero-Shot Dense Retrieval

Nov 27, 2023
Fan Jiang, Qiongkai Xu, Tom Drummond, Trevor Cohn

Viaarxiv icon

Noisy Self-Training with Synthetic Queries for Dense Retrieval

Nov 27, 2023
Fan Jiang, Tom Drummond, Trevor Cohn

Viaarxiv icon

Multi-EuP: The Multilingual European Parliament Dataset for Analysis of Bias in Information Retrieval

Nov 03, 2023
Jinrui Yang, Timothy Baldwin, Trevor Cohn

Viaarxiv icon

Language models are not naysayers: An analysis of language models on negation benchmarks

Jun 14, 2023
Thinh Hung Truong, Timothy Baldwin, Karin Verspoor, Trevor Cohn

Figure 1 for Language models are not naysayers: An analysis of language models on negation benchmarks
Figure 2 for Language models are not naysayers: An analysis of language models on negation benchmarks
Figure 3 for Language models are not naysayers: An analysis of language models on negation benchmarks
Figure 4 for Language models are not naysayers: An analysis of language models on negation benchmarks
Viaarxiv icon

A Reminder of its Brittleness: Language Reward Shaping May Hinder Learning for Instruction Following Agents

May 26, 2023
Sukai Huang, Nir Lipovetzky, Trevor Cohn

Figure 1 for A Reminder of its Brittleness: Language Reward Shaping May Hinder Learning for Instruction Following Agents
Figure 2 for A Reminder of its Brittleness: Language Reward Shaping May Hinder Learning for Instruction Following Agents
Figure 3 for A Reminder of its Brittleness: Language Reward Shaping May Hinder Learning for Instruction Following Agents
Figure 4 for A Reminder of its Brittleness: Language Reward Shaping May Hinder Learning for Instruction Following Agents
Viaarxiv icon

IMBERT: Making BERT Immune to Insertion-based Backdoor Attacks

May 25, 2023
Xuanli He, Jun Wang, Benjamin Rubinstein, Trevor Cohn

Figure 1 for IMBERT: Making BERT Immune to Insertion-based Backdoor Attacks
Figure 2 for IMBERT: Making BERT Immune to Insertion-based Backdoor Attacks
Figure 3 for IMBERT: Making BERT Immune to Insertion-based Backdoor Attacks
Figure 4 for IMBERT: Making BERT Immune to Insertion-based Backdoor Attacks
Viaarxiv icon

Mitigating Backdoor Poisoning Attacks through the Lens of Spurious Correlation

May 19, 2023
Xuanli He, Qiongkai Xu, Jun Wang, Benjamin Rubinstein, Trevor Cohn

Figure 1 for Mitigating Backdoor Poisoning Attacks through the Lens of Spurious Correlation
Figure 2 for Mitigating Backdoor Poisoning Attacks through the Lens of Spurious Correlation
Figure 3 for Mitigating Backdoor Poisoning Attacks through the Lens of Spurious Correlation
Figure 4 for Mitigating Backdoor Poisoning Attacks through the Lens of Spurious Correlation
Viaarxiv icon