Picture for Matthew Finlayson

Matthew Finlayson

Logits of API-Protected LLMs Leak Proprietary Information

Mar 15, 2024
Figure 1 for Logits of API-Protected LLMs Leak Proprietary Information
Figure 2 for Logits of API-Protected LLMs Leak Proprietary Information
Figure 3 for Logits of API-Protected LLMs Leak Proprietary Information
Figure 4 for Logits of API-Protected LLMs Leak Proprietary Information
Viaarxiv icon

Closing the Curious Case of Neural Text Degeneration

Add code
Oct 02, 2023
Figure 1 for Closing the Curious Case of Neural Text Degeneration
Figure 2 for Closing the Curious Case of Neural Text Degeneration
Figure 3 for Closing the Curious Case of Neural Text Degeneration
Figure 4 for Closing the Curious Case of Neural Text Degeneration
Viaarxiv icon

Attentiveness to Answer Choices Doesn't Always Entail High QA Accuracy

May 24, 2023
Figure 1 for Attentiveness to Answer Choices Doesn't Always Entail High QA Accuracy
Figure 2 for Attentiveness to Answer Choices Doesn't Always Entail High QA Accuracy
Figure 3 for Attentiveness to Answer Choices Doesn't Always Entail High QA Accuracy
Figure 4 for Attentiveness to Answer Choices Doesn't Always Entail High QA Accuracy
Viaarxiv icon

Lila: A Unified Benchmark for Mathematical Reasoning

Add code
Oct 31, 2022
Figure 1 for Lila: A Unified Benchmark for Mathematical Reasoning
Figure 2 for Lila: A Unified Benchmark for Mathematical Reasoning
Figure 3 for Lila: A Unified Benchmark for Mathematical Reasoning
Figure 4 for Lila: A Unified Benchmark for Mathematical Reasoning
Viaarxiv icon

Decomposed Prompting: A Modular Approach for Solving Complex Tasks

Add code
Oct 05, 2022
Figure 1 for Decomposed Prompting: A Modular Approach for Solving Complex Tasks
Figure 2 for Decomposed Prompting: A Modular Approach for Solving Complex Tasks
Figure 3 for Decomposed Prompting: A Modular Approach for Solving Complex Tasks
Figure 4 for Decomposed Prompting: A Modular Approach for Solving Complex Tasks
Viaarxiv icon

What Makes Instruction Learning Hard? An Investigation and a New Challenge in a Synthetic Environment

Add code
Apr 19, 2022
Figure 1 for What Makes Instruction Learning Hard? An Investigation and a New Challenge in a Synthetic Environment
Figure 2 for What Makes Instruction Learning Hard? An Investigation and a New Challenge in a Synthetic Environment
Figure 3 for What Makes Instruction Learning Hard? An Investigation and a New Challenge in a Synthetic Environment
Figure 4 for What Makes Instruction Learning Hard? An Investigation and a New Challenge in a Synthetic Environment
Viaarxiv icon

Causal Analysis of Syntactic Agreement Mechanisms in Neural Language Models

Add code
Jun 22, 2021
Figure 1 for Causal Analysis of Syntactic Agreement Mechanisms in Neural Language Models
Figure 2 for Causal Analysis of Syntactic Agreement Mechanisms in Neural Language Models
Figure 3 for Causal Analysis of Syntactic Agreement Mechanisms in Neural Language Models
Figure 4 for Causal Analysis of Syntactic Agreement Mechanisms in Neural Language Models
Viaarxiv icon