AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

Beyond Length: Quantifying Long-Range Information for Long-Context LLM Pretraining Data

Haoran Deng, Yingyu Lin, Zhenghao Lin, Xiao Liu, Yizhou Sun, Yian Ma, Yeyun Gong

ICLR 2026 · Large language models › Language, safety and interpretability · approach: Controlled text generation

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
select top 20% of long data by LongFilter score instead of unfiltered long dataLlama-3-8B continued pretraining · LongBench overall (Book)overall score: 37.47 -> 39.81Improved
select top 20% by LongFilter score instead of unfiltered dataLlama-3-8B continued pretraining · RULER overall (Book)overall score: 73.35 -> 78.95Improved
LongFilter selection vs unfiltered dataLlama-3-8B continued pretraining · LongBench overall (ArXiv)overall score: 38.58 -> 39.52Improved
LongFilter selection vs unfiltered dataLlama-3-8B continued pretraining · RULER overall (ArXiv)overall score: 69.28 -> 70.13Improved
LongFilter selection vs unfiltered dataLlama-3-8B continued pretraining · RULER overall (CommonCrawl)overall score: 72.59 -> 75.37Improved
LongFilter selection vs unfiltered dataLlama-3-8B continued pretraining · RULER Other subtask (CommonCrawl)score: 47.54 -> 31.71Hurt
LongFilter selection vs LongWanjuanLlama-3-8B continued pretraining · LongBench overall (CommonCrawl)overall score: 40.02 -> 40.66Improved
LongFilter selection vs unfiltered dataLlama-3-8B continued pretraining · HELMET overalltraining efficiencyImproved
LongFilter selection vs unfiltered dataLlama-3-8B continued pretraining · HELMET Recall (SubEM)recall: up to 10% gain at 1B tokensImproved
LongFilter selection vs LongWanjuanLlama-3-8B continued pretraining · LongBench Summ (ArXiv)score: 24.83 vs 24.48Mixed

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