Introduces the MEA benchmark for multi-target cross-lingual summarization across 24 languages and demonstrates that activation steering from English summarization representations improves performance.
Educational Psychology
7 Pith papers cite this work. Polarity classification is still indexing.
7
Pith papers citing it
representative citing papers
A proposed pipeline shows LLMs introduce detectable race and gender biases when summarizing life narratives, creating potential for representational harm in research.
LaMSUM is a novel multi-level LLM framework with voting methods for extractive summarization of large incident report collections that outperforms prior extractive methods.
Iterative peer-editing with audio enables human summaries of conversational speech to match the informativeness of transcript summaries and LLM outputs.
BM25-retrieved many-shot examples match much larger random sets for translating English into ten truly low-resource languages, and ICL still helps after fine-tuning.
Human reference summaries outperform LLM outputs in informativeness, faithfulness, and factuality, while LLMs lead only in fluency and coherence, indicating summarization remains an open problem.
Auto-Diagnose applies LLMs to summarize and diagnose root causes of integration test failures, reporting 90.14% accuracy on 71 manual cases and positive adoption after Google-wide rollout.
citing papers explorer
-
Understanding LLM Behavior in Multi-Target Cross-Lingual Summarization
Introduces the MEA benchmark for multi-target cross-lingual summarization across 24 languages and demonstrates that activation steering from English summarization representations improves performance.
-
Whose Story Gets Told? Positionality and Bias in LLM Summaries of Life Narratives
A proposed pipeline shows LLMs introduce detectable race and gender biases when summarizing life narratives, creating potential for representational harm in research.
-
LaMSUM: Amplifying Voices Against Harassment through LLM Guided Extractive Summarization of User Incident Reports
LaMSUM is a novel multi-level LLM framework with voting methods for extractive summarization of large incident report collections that outperforms prior extractive methods.
-
Beyond Transcripts: Iterative Peer-Editing with Audio Unlocks High-Quality Human Summaries of Conversational Speech
Iterative peer-editing with audio enables human summaries of conversational speech to match the informativeness of transcript summaries and LLM outputs.
-
An Empirical Study of Many-Shot In-Context Learning for Machine Translation of Low-Resource Languages
BM25-retrieved many-shot examples match much larger random sets for translating English into ten truly low-resource languages, and ICL still helps after fine-tuning.
-
Summarization is Not Dead Yet
Human reference summaries outperform LLM outputs in informativeness, faithfulness, and factuality, while LLMs lead only in fluency and coherence, indicating summarization remains an open problem.
-
LLM-Based Automated Diagnosis Of Integration Test Failures At Google
Auto-Diagnose applies LLMs to summarize and diagnose root causes of integration test failures, reporting 90.14% accuracy on 71 manual cases and positive adoption after Google-wide rollout.