DL and NLP

Interested in suggesting a paper and leading a session? Contact Evgenia Ilia on AmsterdamNLP’s slack.

Deep de Finetti: Recovering Topic Distributions from Large Language Models
01/02/23 11.30, L6.51 and probabll.zoom.

A Statistical Analysis of Summarization Evaluation Metrics Using Resampling Methods
04/10/23 14:00, L6.51 and probabll.zoom.

The impact of results blind science publishing on statistical consultation and evaluation
20/09/23 15:30, L6.51 and probabll.zoom.

Moving to a World Beyond 'p < 0.05'
13/09/23 14:00, L6.51 and probabll.zoom.

Neighboring Words Affect Human Interpretation of Saliency Explanations
04/07/23.

Navigating the Grey Area: Expressions of Overconfidence and Uncertainty in Language Models
25/05/23.

R-U-SURE? Uncertainty-Aware Code Suggestions By Maximizing Utility Across Random User Intents
11/05/23.

COLT5: Faster Long-Range Transformers with Conditional Computation
30/3.

Calibration Meets Explanation: A Simple and Effective Approach for Model Confidence Estimates
9/3 15:00.

“Will You Find These Shortcuts?” A Protocol for Evaluating the Faithfulness of Input Salience Methods for Text Classification
2/3 15:00, L2.07.

Epistemic Neural Networks
23/2 15:00, L2.07.

Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
17/02/23 15:00, L6.51.

Do Bayesian Neural Networks Need To Be Fully Stochastic?
02/12/22 15:00, L6.51.

Adaptive Label Smoothing with Self-Knowledge in Natural Language Generation
18/11/22 15:00, C3.2-11 (SP904).

Understanding Black-box Predictions via Influence Functions
11/11/22 15:00, L6.51.

Contrastive Decoding: Open-ended Text Generation as Optimization
04/11/22 15:00, L6.51.

Model Criticism for Long-Form Text Generation
02/11/22 16:30, L6.51.