Scaling Attention Head Analysis via Gradient-Based Attribution in Context-Aware Machine Translation
Paper recorded by Signals 4 on 2026-09-23 in cs.CL. Abstract reproduced from arXiv; link to the original below.
Published 2026-09-23 on arXiv · recorded by Signals 4 on 2026-09-24
Category: cs.CL · 自然语言处理 · first seen 2026-09-24
Abstract
In this paper, we introduce a gradient-based head attribution strategy where the Token-level Max-Margin loss is backpropagated to the attention maps. This framework enables a large-scale causal analysis of attention heads, making it suitable for LLMs. We evaluate our method on the task of disambiguation in Context-aware Machine Translation, where we analyze 50 phenomena across 4 models and 4 langu
Read on arXiv →
Cite this page: Scaling Attention Head Analysis via Gradient-Based Attribution in Context-Aware Machine Translation: the #10 most recent of 239 cs.CL papers we have recorded (as of 2026-09-23). Source: Signals 4 (Signals API) — https://data.jiangzhang.ca/signals4/t/papers/scaling-attention-head-analysis-via-gradient-based-attribution-in-context-aware-.html
Free to quote with attribution to “Signals 4 (Signals API)”. Machine-readable:
papers.json
Get 4 AI signals a day by email — free.
Get 4 AI signals a day by email — free