AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

The Heads Hypothesis: A Unifying Statistical Approach Towards Understanding Multi-Headed Attention in BERT

Madhura Pande, Aakriti Budhraja, Preksha Nema, Pratyush Kumar, Mitesh M. Khapra

AAAI 2021 · Architectures and efficiency › Attention and transformers · approach: Sparse and structured attention · 10 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Classify functional roles using normalized sieve bias and hypothesis testing rather than average scores aloneBERT attention heads · GLUE tasks (QNLI, QQP, MRPC, SST-2)statistical confidence of role assignmentImproved
Allow a head to receive multiple functional-role labels instead of assigning only oneBERT attention heads · GLUE tasks (QNLI, QQP, MRPC, SST-2)overlap of syntactic heads with local heads: 42.1% to 88.8% (with a mean of 70.9%)Mixed
Fine-tune BERT for a taskBERT_BASE · Final layers, across GLUE tasksdelimiter sieve biasHurt
Fine-tune BERT for a taskBERT_BASE · Final layers, across GLUE tasksblock sieve biasImproved
Fine-tune BERT for a taskBERT_BASE · Initial and middle layers, across GLUE tasksaverage sieve bias scoresNo clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy