Pith Number

pith:IEDUSXUN

pith:2016:IEDUSXUNSRHHWCLZYQR7MYZ6LQ

not attested not anchored not stored refs resolved

Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation

Alex Rudnick, Apurva Shah, Cliff Young, George Kurian, Greg Corrado, Hideto Kazawa, Jason Riesa, Jason Smith, Jeff Klingner, Jeffrey Dean, Keith Stevens, Klaus Macherey, {\L}ukasz Kaiser, Macduff Hughes, Maxim Krikun, Melvin Johnson, Mike Schuster, Mohammad Norouzi, Nishant Patil, Oriol Vinyals, Qin Gao, Quoc V. Le, Stephan Gouws, Taku Kudo, Wei Wang, Wolfgang Macherey, Xiaobing Liu, Yonghui Wu, Yoshikiyo Kato, Yuan Cao, Zhifeng Chen

GNMT, a deep LSTM neural machine translation system with wordpieces and coverage penalties, reduces translation errors by an average of 60% compared to phrase-based systems.

arxiv:1609.08144 v2 · 2016-09-26 · cs.CL · cs.AI · cs.LG

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{IEDUSXUNSRHHWCLZYQR7MYZ6LQ}

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

Using a human side-by-side evaluation on a set of isolated simple sentences, it reduces translation errors by an average of 60% compared to Google's phrase-based production system.

C2weakest assumption

That the measured gains are attributable to the described architectural choices (attention placement, wordpieces, coverage penalty) rather than differences in training data scale or compute, and that results on simple sentences generalize to complex, domain-specific text.

C3one line summary

GNMT deploys 8-layer LSTMs with attention, wordpieces, low-precision inference, and coverage-penalized beam search to match state-of-the-art on WMT'14 En-Fr and En-De while cutting translation errors by 60% in human evaluations.

References

44 extracted · 44 resolved · 4 Pith anchors

[1] G., Steiner, B., Tucker, P., V asudevan, V., W arden, P., Wicke, M., Yu, Y., and Zheng, X 2016

[2] Neural machine translation by jointly learning to align and translate 2015

[3] Brown, P., Cocke, J., Pietra, S. D., Pietra, V. D., Jelinek, F., Mercer, R., and Roossin, P. A statistical approach to language translation. InProceedings of the 12th Conference on Computational Lingu 1988

[4] F., Cocke, J., Pietra, S 1990

[5] Brown, P. F., Pietra, V. J. D., Pietra, S. A. D., and Mercer, R. L. The mathematics of statistical machine translation: Parameter estimation.Comput. Linguist. 19, 2 (June 1993), 263–311 1993

Formal links

2 machine-checked theorem links

Cited by

87 papers in Pith

Ad Headline Generation using Self-Critical Masked Language Model

Brevity is the Soul of Inference Efficiency: Inducing Concision in VLMs via Data Curation

OTRO: Oblivious Tokenization Path with Square-Root ORAM

MATLAB-Based Layerwise Self-Adaptive Physics-Informed Neural Network in Applications to Multidimensional Coupled Burgers' Equations with High Reynolds Numbers

Gryphon: A Unified Architecture for Semantic-ID Generation and Item-Level Scoring in Industrial Recommendations

Receipt and verification

First computed	2026-07-04T21:27:34.571107Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

4107495e8d944e7b0979c423f6633e5c37fd1fe1cfb8fa27e6cc97c6099f2e03

Aliases

arxiv: 1609.08144 · arxiv_version: 1609.08144v2 · doi: 10.48550/arxiv.1609.08144 · pith_short_12: IEDUSXUNSRHH · pith_short_16: IEDUSXUNSRHHWCLZ · pith_short_8: IEDUSXUN

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/IEDUSXUNSRHHWCLZYQR7MYZ6LQ \
  | 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: 4107495e8d944e7b0979c423f6633e5c37fd1fe1cfb8fa27e6cc97c6099f2e03

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "f8728cf6bc353f04780df8a769aa5b6792264ff8dd0e04d7bfd682bf2a9cfe92",
    "cross_cats_sorted": [
      "cs.AI",
      "cs.LG"
    ],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.CL",
    "submitted_at": "2016-09-26T19:59:55Z",
    "title_canon_sha256": "7e1f30af6bb78f3b01e138dd4c95c0225405b345dc64dcb2de9ce5d5e167c230"
  },
  "schema_version": "1.0",
  "source": {
    "id": "1609.08144",
    "kind": "arxiv",
    "version": 2
  }
}