Papers with Normalized Discounted Cumulative Gain

2 papers
Permutative Preference Alignment from Listwise Ranking of Human Judgments (2025.emnlp-main)

Copied to clipboard

Challenge: Existing methods to align Large Language Models with human preferences are based on the Bradley-Terry model, but when multiple responses are available, the B-T model fails to guarantee an accurate list ranking of the responses.
Approach: They propose an offline listwise approach that incorporates the Normalized Discounted Cumulative Gain (NDCG) as an alternative training objective for LLM alignment.
Outcome: The proposed approach outperforms existing pairwise and listwise methods on evaluation sets and general benchmarks such as AlpacaEval.
MOSAIC: Masked Objective with Selective Adaptation for In-domain Contrastive Learning (2026.findings-eacl)

Copied to clipboard

Challenge: a new framework for domain adaptation of text embedding models addresses the challenges of adapting general-domain text embeds to specialized domains.
Approach: They propose a framework for domain adaptation of text embedding models that integrates masked supervision and mangled objectives within a unified training pipeline.
Outcome: The proposed framework improves on high-resource and low-resourced domains while preserving the robustness of the original model.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations