Picture for Carsten Eickhoff

Carsten Eickhoff

Axiomatic Causal Interventions for Reverse Engineering Relevance Computation in Neural Retrieval Models

Add code
May 03, 2024
Viaarxiv icon

Controllable Topic-Focused Abstractive Summarization

Nov 12, 2023
Viaarxiv icon

Outlier Dimensions Encode Task-Specific Knowledge

Add code
Oct 26, 2023
Figure 1 for Outlier Dimensions Encode Task-Specific Knowledge
Figure 2 for Outlier Dimensions Encode Task-Specific Knowledge
Figure 3 for Outlier Dimensions Encode Task-Specific Knowledge
Figure 4 for Outlier Dimensions Encode Task-Specific Knowledge
Viaarxiv icon

Circuit Component Reuse Across Tasks in Transformer Language Models

Add code
Oct 12, 2023
Figure 1 for Circuit Component Reuse Across Tasks in Transformer Language Models
Figure 2 for Circuit Component Reuse Across Tasks in Transformer Language Models
Figure 3 for Circuit Component Reuse Across Tasks in Transformer Language Models
Figure 4 for Circuit Component Reuse Across Tasks in Transformer Language Models
Viaarxiv icon

Predictive Uncertainty-based Bias Mitigation in Ranking

Sep 18, 2023
Figure 1 for Predictive Uncertainty-based Bias Mitigation in Ranking
Figure 2 for Predictive Uncertainty-based Bias Mitigation in Ranking
Figure 3 for Predictive Uncertainty-based Bias Mitigation in Ranking
Figure 4 for Predictive Uncertainty-based Bias Mitigation in Ranking
Viaarxiv icon

One-Versus-Others Attention: Scalable Multimodal Integration

Add code
Jul 11, 2023
Figure 1 for One-Versus-Others Attention: Scalable Multimodal Integration
Figure 2 for One-Versus-Others Attention: Scalable Multimodal Integration
Figure 3 for One-Versus-Others Attention: Scalable Multimodal Integration
Figure 4 for One-Versus-Others Attention: Scalable Multimodal Integration
Viaarxiv icon

SSE: A Metric for Evaluating Search System Explainability

Add code
Jun 16, 2023
Viaarxiv icon

Stable Anisotropic Regularization

Add code
May 30, 2023
Figure 1 for Stable Anisotropic Regularization
Figure 2 for Stable Anisotropic Regularization
Figure 3 for Stable Anisotropic Regularization
Figure 4 for Stable Anisotropic Regularization
Viaarxiv icon

Enhancing the Ranking Context of Dense Retrieval Methods through Reciprocal Nearest Neighbors

May 25, 2023
Figure 1 for Enhancing the Ranking Context of Dense Retrieval Methods through Reciprocal Nearest Neighbors
Figure 2 for Enhancing the Ranking Context of Dense Retrieval Methods through Reciprocal Nearest Neighbors
Figure 3 for Enhancing the Ranking Context of Dense Retrieval Methods through Reciprocal Nearest Neighbors
Figure 4 for Enhancing the Ranking Context of Dense Retrieval Methods through Reciprocal Nearest Neighbors
Viaarxiv icon

Language Models Implement Simple Word2Vec-style Vector Arithmetic

Add code
May 25, 2023
Figure 1 for Language Models Implement Simple Word2Vec-style Vector Arithmetic
Figure 2 for Language Models Implement Simple Word2Vec-style Vector Arithmetic
Figure 3 for Language Models Implement Simple Word2Vec-style Vector Arithmetic
Figure 4 for Language Models Implement Simple Word2Vec-style Vector Arithmetic
Viaarxiv icon