|
- Counterfactual Debiasing for Fact Verification - OpenReview
016 namely CLEVER, which is augmentation-free 017 and mitigates biases on the inference stage 018 Specifically, we train a claim-evidence fusion 019 model and a claim-only model independently 020 Then, we obtain the final prediction via sub-021 tracting output of the claim-only model from 022 output of the claim-evidence fusion model,
- Measuring Mathematical Problem Solving With the MATH Dataset
To find the limits of Transformers, we collected 12,500 math problems While a three-time IMO gold medalist got 90%, GPT-3 models got ~5%, with accuracy increasing slowly
- NetMoE: Accelerating MoE Training through Dynamic Sample Placement
Mixture of Experts (MoE) is a widely used technique to expand model sizes for better model quality while maintaining the computation cost constant
- Weakly-Supervised Affordance Grounding Guided by Part-Level. . .
In this work, we focus on the task of weakly supervised affordance grounding, where a model is trained to identify affordance regions on objects using human-object interaction images and egocentric object images without dense labels
- Large Language Models are Human-Level Prompt Engineers
We propose an algorithm for automatic instruction generation and selection for large language models with human level performance
- Training Large Language Model to Reason in a Continuous . . . - OpenReview
Large language models are restricted to reason in the “language space”, where they typically express the reasoning process with a chain-of-thoughts (CoT) to solve a complex reasoning problem
- Reasoning of Large Language Models over Knowledge Graphs with. . .
While large language models (LLMs) have made significant progress in processing and reasoning over knowledge graphs, current methods suffer from a high non-retrieval rate
- Ignore Previous Prompt: Attack Techniques For Language Models - OpenReview
Ignore Previous Prompt: Attack Techniques For Language Models Fábio Perez Ian Ribeiro AE Studio {fperez,ian ribeiro}@ae studio
|
|
|