Do LLMs Silence Human Voices? Unveiling the ‘Spiral of Silence’ in Retrieval‑Augmented Generation

This article reviews the ACL 2024 paper that investigates how large language model‑generated text influences retrieval‑augmented generation pipelines, revealing short‑term retrieval gains but a long‑term “spiral of silence” that marginalizes human‑generated content and homogenizes open‑domain QA results.

AI ImpactLLMOpen Domain QA

0 likes · 9 min read

Do LLMs Silence Human Voices? Unveiling the ‘Spiral of Silence’ in Retrieval‑Augmented Generation

DataFunSummit

Jul 29, 2022 · Artificial Intelligence

Integrating Knowledge Graphs with Neural Networks: Generative Pre‑Training, Differentiable Reasoning, and Fuzzy Logic Query Embedding

This article reviews recent work on combining knowledge graphs with neural networks, covering generative self‑supervised graph neural network pre‑training, differentiable logical reasoning over graphs, and a fuzzy‑logic based query‑embedding model that improves open‑domain question answering, especially for rare relations.

Graph Neural NetworkOpen Domain QAdifferentiable reasoning

0 likes · 22 min read

Integrating Knowledge Graphs with Neural Networks: Generative Pre‑Training, Differentiable Reasoning, and Fuzzy Logic Query Embedding

DataFunTalk

Jun 8, 2022 · Artificial Intelligence

Integrating Knowledge Graphs with Neural Networks: Generative Pre‑Training, Differentiable Reasoning, and Fuzzy Logic Query Embedding

This article reviews recent advances in combining knowledge graphs with neural networks, covering generative pre‑training of graph neural networks, wiki‑graph based open‑domain question answering, differentiable logical reasoning, and a fuzzy‑logic query‑embedding model that improves performance on sparse‑relation queries.

Artificial IntelligenceOpen Domain QAfuzzy logic

0 likes · 23 min read

Alibaba Cloud Developer

Jul 31, 2019 · Artificial Intelligence

How Alibaba’s Enriched BERT Set a New Record in Open‑Domain QA

Alibaba’s AI team introduced a multi‑stage, document‑ranking and paragraph‑ranking system built on an Enriched BERT model that topped the MS MARCO reading‑comprehension leaderboard, surpassing previous state‑of‑the‑art methods and even human performance on open‑domain QA tasks.

Alibaba AIEnriched BERTOpen Domain QA

0 likes · 5 min read

How Alibaba’s Enriched BERT Set a New Record in Open‑Domain QA