Pith Number

pith:HAMJZSQV

pith:2024:HAMJZSQVKL4VHWOAINCHUE26II

not attested not anchored not stored refs resolved

LongRoPE: Extending LLM Context Window Beyond 2 Million Tokens

Chengruidong Zhang, Fan Yang, Jiahang Xu, Li Lyna Zhang, Mao Yang, Ning Shang, Yiran Ding, Yuanyuan Xu

LongRoPE extends pre-trained LLMs to 2048k token contexts via targeted non-uniform positional interpolation and a two-stage fine-tuning process.

arxiv:2402.13753 v1 · 2024-02-21 · cs.CL

Open paper page JSON Open Graph Bundle Merged state Verified badge What is a Pith Number?

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{HAMJZSQVKL4VHWOAINCHUE26II}

Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge

Record completeness

1 Bitcoin timestamp

2 Internet Archive

3 Author claim open · sign in to claim

4 Citations open

5 Replications open

✓ Portable graph bundle live · download bundle · merged state

The bundle contains the canonical record plus signed events. A mirror can host it anywhere and recompute the same current state with the deterministic merge algorithm.

Claims

C1strongest claim

LongRoPE extends the context window of pre-trained LLMs to an impressive 2048k tokens, with up to only 1k fine-tuning steps at within 256k training lengths, while maintaining performance at the original short context window.

C2weakest assumption

The two forms of non-uniformities in positional interpolation identified via efficient search are generalizable across models and tasks and provide a stable initialization that does not overfit to the search data.

C3one line summary

LongRoPE extends LLM context windows to 2048k tokens via search for non-uniform positional interpolation, progressive fine-tuning from 256k, and short-context readjustment.

References

15 extracted · 15 resolved · 6 Pith anchors

[1] Extending Context Window of Large Language Models via Positional Interpolation · arXiv:2306.15595

[2] The Pile: An 800GB Dataset of Diverse Text for Language Modeling · arXiv:2101.00027

[3] Single path one-shot neural architecture search with uniform sampling 2020

[4] Lm-infinite: Simple on-the-fly length generalization for large language models

[5] Measuring Massive Multitask Language Understanding 2009 · arXiv:2009.03300

Formal links

2 machine-checked theorem links

Cited by

29 papers in Pith

LLM-based User Profile Management for Recommender System

Jordan-RoPE: Non-Semisimple Relative Positional Encoding via Complex Jordan Blocks

SEGA: Spectral-Energy Guided Attention for Resolution Extrapolation in Diffusion Transformers

Benchmarking EngGPT2-16B-A3B against Comparable Italian and International Open-source LLMs

PaceLLM: Brain-Inspired Large Language Models for Long-Context Understanding

Receipt and verification

First computed	2026-05-17T23:38:53.019275Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

38189cca1552f953d9c043447a135e420636aa9e5d0a13a0ca99322e4750d280

Aliases

arxiv: 2402.13753 · arxiv_version: 2402.13753v1 · doi: 10.48550/arxiv.2402.13753 · pith_short_12: HAMJZSQVKL4V · pith_short_16: HAMJZSQVKL4VHWOA · pith_short_8: HAMJZSQV

Agent API

Resolver JSON Graph JSON Events JSON Schema Signing key

Verify this Pith Number yourself

curl -sH 'Accept: application/ld+json' https://pith.science/pith/HAMJZSQVKL4VHWOAINCHUE26II \
  | jq -c '.canonical_record' \
  | python3 -c "import sys,json,hashlib; b=json.dumps(json.loads(sys.stdin.read()), sort_keys=True, separators=(',',':'), ensure_ascii=False).encode(); print(hashlib.sha256(b).hexdigest())"
# expect: 38189cca1552f953d9c043447a135e420636aa9e5d0a13a0ca99322e4750d280

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "0379a2839dbbe711aa7363393e5effbbca314a73f949a18ec961afaed67dfb0f",
    "cross_cats_sorted": [],
    "license": "http://creativecommons.org/licenses/by-nc-nd/4.0/",
    "primary_cat": "cs.CL",
    "submitted_at": "2024-02-21T12:30:33Z",
    "title_canon_sha256": "46e4af97439ff2209ef39f593912fe97c0c8c33415282cb5b0df3aafaf04abf8"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2402.13753",
    "kind": "arxiv",
    "version": 1
  }
}