ARXIV:2601.19613 · DOCUMENT AI · SUBMITTED 02 APR · 02:30 UTC · FRESHNESS STALE

VerifiedSource: PDF linkedPartialPaperPack: 3 of 4 citation fields filledMissingMissing fields: authorsPartialProof: unverified proof status

Up to 36x Speedup: Mask-based Parallel Inference Paradigm for Key Information Extraction in MLLMs

arXiv

PIP offers a mask-based paradigm for a 5-36x speedup in key information extraction from documents, maintaining accuracy while enhancing efficiency.

Blocked on Code›Score7.0Evidence unverified

Opportunity summary

Pain PIP offers a mask-based paradigm for a 5-36x speedup in key information extraction from documents, maintaining accuracy while enhancing efficiency.

Evidence 0 refs | 0 sources | 17% coverage

Blocker Evidence unverified

Open Build Read PDF Signal Canvas Track

PROBLEM

PIP offers a mask-based paradigm for a 5-36x speedup in key information extraction from documents, maintaining accuracy while enhancing efficiency. However, their reliance on autoregressive inference, which generates outputs sequentially, creates a significant efficiency…

METHOD

Full abstract

Key Information Extraction (KIE) from visually-rich documents (VrDs) is a critical task, for which recent Large Language Models (LLMs) and Multi-Modal Large Language Models (MLLMs) have demonstrated strong potential. However, their reliance on autoregressive inference, which generates outputs sequentially, creates a significant efficiency bottleneck, especially as KIE tasks often involve extracting multiple, semantically independent fields. To overcome this limitation, we introduce PIP: a Parallel Inference Paradigm for KIE. Our approach reformulates the problem by using "[mask]" tokens as placeholders for all target values, enabling their simultaneous generation in a single forward pass. To facilitate this paradigm, we develop a tailored mask pre-training strategy and construct large-scale supervised datasets. Experimental results show that our PIP-models achieve a 5-36x inference speedup with negligible performance degradation compared to traditional autoregressive base models. By substantially improving efficiency while maintaining high accuracy, PIP paves the way for scalable and practical real-world KIE solutions.

RESULT

ScienceToStartup currently rates this 7.0/10 on the public viability pass. Experimental results show that our PIP-models achieve a 5-36x inference speedup with negligible performance degradation compared to traditional autoregressive base models.

WHY NOW

Document AI moved forward this cycle; last verified April 2026. Public score 7.0/10.

Continue into Read for claims, analysis, references, and neighboring papers.

Opportunity summary

Score7.0

PainPIP offers a mask-based paradigm for a 5-36x speedup in key information extraction from documents, maintaining accuracy while enhancing efficiency.

Evidence0 refs | 0 sources | 17% coverage

Blockermissing authors

Analysis summary

PIP offers a mask-based paradigm for a 5-36x speedup in key information extraction from documents, maintaining accuracy while enhancing efficiency.

VerifiedSource: PDF linkedPartialPaperPack: 3 of 4 citation fields filledMissingMissing fields: authorsPartialProof: unverified proof status

ARXIV:2601.19613 · DOCUMENT AI · SUBMITTED 02 APR · 02:30 UTC · FRESHNESS STALE

VerifiedSource: PDF linkedPartialPaperPack: 3 of 4 citation fields filledMissingMissing fields: authorsPartialProof: unverified proof status

Up to 36x Speedup: Mask-based Parallel Inference Paradigm for Key Information Extraction in MLLMs

arXiv

PIP offers a mask-based paradigm for a 5-36x speedup in key information extraction from documents, maintaining accuracy while enhancing efficiency.

Blocked on Code›Score7.0Evidence unverified

Opportunity summary

Pain PIP offers a mask-based paradigm for a 5-36x speedup in key information extraction from documents, maintaining accuracy while enhancing efficiency.

Evidence 0 refs | 0 sources | 17% coverage

Blocker Evidence unverified

Open Build Read PDF Signal Canvas Track

PROBLEM

METHOD

Full abstract

RESULT

WHY NOW

Document AI moved forward this cycle; last verified April 2026. Public score 7.0/10.

Continue into Read for claims, analysis, references, and neighboring papers.

Opportunity summary

Score7.0

PainPIP offers a mask-based paradigm for a 5-36x speedup in key information extraction from documents, maintaining accuracy while enhancing efficiency.

Evidence0 refs | 0 sources | 17% coverage

Blockermissing authors

Analysis summary

PIP offers a mask-based paradigm for a 5-36x speedup in key information extraction from documents, maintaining accuracy while enhancing efficiency.

VerifiedSource: PDF linkedPartialPaperPack: 3 of 4 citation fields filledMissingMissing fields: authorsPartialProof: unverified proof status

Paper Pack

10.48550/arXiv.2601.19613

Up to 36x Speedup: Mask-based Parallel Inference Paradigm for Key Information Extraction in MLLMs

PIP offers a mask-based paradigm for a 5-36x speedup in key information extraction from documents, maintaining accuracy while enhancing efficiency.

Abstract

Source availability

PDF linked

The paper record includes a public PDF URL.

Extraction status

Parse run linked

A document parse run is attached to this paper.

Proof status

unverified

0 refs; 0 sources; 17% coverage.

What was readable

linkedon file8 anchorsderived fallback23 indexednot indexed

Derived fallback: Estimated from adjacent evidence; not verified from source.

Viability

7.0

Time to MVP

MVP estimate missing

Commercial

No commercial flags on file

Export

Preparing verified analysis

lens / founder

PROBLEM

METHOD

RESULT

WHY NOW

Document AI moved forward this cycle; last verified April 2026. Public score 7.0/10.

Claim map

Abstract-backed public claims while anchored extraction refreshes.

Strong 0Mixed 0Weak 4

Evidencepartial
PIP offers a mask-based paradigm for a 5-36x speedup in key information extraction from documents, maintaining accuracy while enhancing efficiency. However, their reliance on autoregressive inference, which generates outputs sequentially, creates a significant efficiency bottleneck, especially as KIE tasks often involve extracting multiple, semantically independent fields.
Implicationpartial
Abstract-backed fallback claim; anchored extraction has not materialized a public claim row yet.
Verificationpartial
partial
Evidencepartial
Key Information Extraction (KIE) from visually-rich documents (VrDs) is a critical task, for which recent Large Language Models (LLMs) and Multi-Modal Large Language Models (MLLMs) have demonstrated strong potential. However, their reliance on autoregressive inference, which generates outputs sequentially, creates a significant efficiency bottleneck, especially as KIE tasks often involve extracting multiple, semantically independent fields.
Implicationpartial
Abstract-backed fallback claim; anchored extraction has not materialized a public claim row yet.
Verificationpartial
partial
Evidencepartial
ScienceToStartup currently rates this 7.0/10 on the public viability pass. Experimental results show that our PIP-models achieve a 5-36x inference speedup with negligible performance degradation compared to traditional autoregressive base models.
Implicationpartial
Abstract-backed fallback claim; anchored extraction has not materialized a public claim row yet.
Verificationpartial
partial
Evidencepartial
Document AI moved forward this cycle; last verified April 2026. Public score 7.0/10.
Implicationpartial
Abstract-backed fallback claim; anchored extraction has not materialized a public claim row yet.
Verificationpartial
partial

Constellation map

Paper-native neighborhood for concepts, methods, materials, markets, and competitors. Missing lanes stay labeled instead of disappearing behind commercialization gates.

Open full Signal Canvas

Concepts

not indexed

Methods

Materials

PDF linkedDocument parse run

Markets

Document AI

Competitors

not indexed

Competitive landscape

PIP offers a mask-based paradigm for a 5-36x speedup in key information extraction from documents, maintaining accuracy while enhancing efficiency.

Segment

Document AI

Adoption evidence

No public code link in the paper record yet

Commercial read

7.0/10 public viability

Direct

not classified

Adjacent

not classified

Substitute

not classified

Unknown

not classified

Buzz

No indexed public discussion is attached to 2601.19613 yet. That is a visibility signal, not a blank module: the monitor is watching the public channels below.

Hacker News

Not indexed yet

Bluesky

Not indexed yet

PDF

Preview the source document here, or use the hero PDF action for a new tab.

References(23)

DeepSeek-OCR: Contexts Optical Compression

2025Haoran Wei, Yaofeng Sun et al.

GLM-4.5: Agentic, Reasoning, and Coding (ARC) Foundation Models

2025GLM-4.5 Team Aohan Zeng, Xin Lv et al.

Large Language Diffusion Models

2025Shen Nie, Fengqi Zhu et al.

Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution

2024Peng Wang, Shuai Bai et al.

A Bounding Box is Worth One Token: Interleaving Layout and Text in a Large Language Model for Document Understanding

2024Jinghui Lu, Haiyang Yu et al.

How far are we to GPT-4V? Closing the gap to commercial multimodal models with open-source suites

2024Zhe Chen, Weiyun Wang et al.

LayoutLLM: Layout Instruction Tuning with Large Language Models for Document Understanding

2024Chuwei Luo, Yufan Shen et al.

DeepSeek-VL: Towards Real-World Vision-Language Understanding

2024Haoyu Lu, Wen Liu et al.

Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context

2024Machel Reid, Nikolay Savinov et al.

TextMonkey: An OCR-Free Large Multimodal Model for Understanding Document

2024Yuliang Liu, Biao Yang et al.

PaDeLLM-NER: Parallel Decoding in Large Language Models for Named Entity Recognition

2024Jinghui Lu, Ziwei Yang et al.

Llama 2: Open Foundation and Fine-Tuned Chat Models

2023Hugo Touvron, Louis Martin et al.

LLaVAR: Enhanced Visual Instruction Tuning for Text-Rich Image Understanding

2023Yanzhe Zhang, Ruiyi Zhang et al.

Visual Information Extraction in the Wild: Practical Dataset and End-to-end Solution

2023Jianfeng Kuang, Wei Hua et al.

MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Models

2023Deyao Zhu, Jun Chen et al.

Visual Instruction Tuning

2023Haotian Liu, Chunyuan Li et al.

GPT-4 Technical Report

2023OpenAI Josh Achiam, Steven Adler et al.

LLaMA: Open and Efficient Foundation Language Models

2023Hugo Touvron, Thibaut Lavril et al.

Spatial Dual-Modality Graph Reasoning for Key Information Extraction

2021Hongbin Sun, Zhanghui Kuang et al.

CORD: A Consolidated Receipt Dataset for Post-OCR Parsing

2019Seunghyun Park, Seung Shin et al.

Showing 20 of 23 references

CITED BY

No citing papers are indexed in the public S2S graph yet. This is an explicit zero-signal state, not a hidden lookup.

Foundation

Prior WorkParallel In-context Learning for Large Vision Language Models

7.0

Prior WorkPair-In, Pair-Out: Latent Multi-Token Prediction for Efficient LLMs

7.0

Prior WorkRethinking Model Efficiency: Multi-Agent Inference with Large Models

7.0

Prior WorkEnhancing Multimodal In-Context Learning via Inductive-Deductive Reasoning

7.0

Extension

Builds On ThisDivide-and-Conquer Inference for Large-Scale Visual Recognition with Multimodal Large Language Models

0.0

Builds On ThisVision-DeepResearch: Incentivizing DeepResearch Capability in Multimodal Large Language Models

5.0

Builds On ThisEfficient Inference of Large Vision Language Models

4.0

Builds On ThisOCR or Not? Rethinking Document Information Extraction in the MLLMs Era with Real-World Large-Scale Datasets

6.0

Commercially relevant

Higher ViabilityPOP: Prefill-Only Pruning for Efficient Large Model Inference

8.0

Higher ViabilityBeyond Scattered Acceptance: Fast and Coherent Inference for DLMs via Longest Stable Prefixes

8.0

Conflicting

none indexed

Owned Distribution

Subscribe to the weekly brief

Get the weekly shortlist of commercializable papers, benchmark movers, and proof receipts that matter for product execution.

Agent drawer

5 surfaces preserved for agents. Humans can ignore.

Developer contracts, payload previews, evidence maps, and run controls stay here instead of the Read, Build, and Track workspace.

Run context

Paper: 2601.19613
Route: /paper/up-to-36x-speedup-mask-based-parallel-inference-paradigm-for-key-information-extraction-in-mllms
Active tab: read
Artifact: up-to-36x-speedup-mask-based-parallel-inference-paradigm-for-key-information-extraction-in-mllms

Available agents

Read extractor
Build planner
Track monitor
Competitive mapper
Related-paper scout

API/MCP endpoints

REST paper pack API/api/v1/paper/up-to-36x-speedup-mask-based-parallel-inference-paradigm-for-key-information-extraction-in-mllms/paper-pack
REST build passport API/api/v1/paper/up-to-36x-speedup-mask-based-parallel-inference-paradigm-for-key-information-extraction-in-mllms/build-passport
REST OpenAPI/api/openapi.json
MCP descriptor/api/mcp
MCP resourcesciencetostartup://surfaces/paper-workspace

Tool contracts

paper_packbuild_passportopportunity_kernelforesightsource_proofevidence_state

Payload preview

Inspect payload

{
  "contract_version": "paper-r2",
  "paper_id": "3947c266-5e09-444f-8dc3-5b806f604ce8",
  "arxiv_id": "2601.19613",
  "canonical_route": "/paper/up-to-36x-speedup-mask-based-parallel-inference-paradigm-for-key-information-extraction-in-mllms",
  "active_tab": "synced from current hash by the drawer client",
  "selected_artifact": "up-to-36x-speedup-mask-based-parallel-inference-paradigm-for-key-information-extraction-in-mllms",
  "endpoints": {
    "paper_pack": "/api/v1/paper/up-to-36x-speedup-mask-based-parallel-inference-paradigm-for-key-information-extraction-in-mllms/paper-pack",
    "build_passport": "/api/v1/paper/up-to-36x-speedup-mask-based-parallel-inference-paradigm-for-key-information-extraction-in-mllms/build-passport",
    "mcp_resource": "sciencetostartup://surfaces/paper-workspace"
  }
}

Schema validation

paper-r2 contract: present
JSON-LD twin: SSR emitted
OpenAPI path parity: /api/openapi.json
MCP resource parity: paper-workspace

Job trace

queued: drawer opened by user action
running: inspect or copy payload
succeeded: payload available in SSR
failed: route errors appear in evidence cards

Evidence map

sources used: page freshness, source proof anchors, JSON-LD
missing sources: exposed by PaperPack and EvidenceState chips
derived fallbacks: marked unverified before handoff

Page Freshness

Canonical route, proof status, last verified, refs, sources, and coverage.

Page Freshness

Paper proof surface

Canonical route: /paper/up-to-36x-speedup-mask-based-parallel-inference-paradigm-for-key-information-extraction-in-mllms

stale

Proof freshness: stale
Proof status: unverified
Display score: 7/10
Last proof check: 2026-04-02
Score updated: 2026-04-02
Score fresh until: 2026-05-02
References: 0
Source count: 0
Coverage: 17%

This page is showing the last landed evidence receipt and score bundle because the latest proof data is outside the freshness window.

OpenAlex: pending — this preprint is not yet indexed by OpenAlex.

Agent Handoff

Endpoint list, payload shape, route context, and copyable handoff data.

Agent Handoff

Up to 36x Speedup: Mask-based Parallel Inference Paradigm for Key Information Extraction in MLLMs

Canonical ID up-to-36x-speedup-mask-based-parallel-inference-paradigm-for-key-information-extraction-in-mllms | Route /paper/up-to-36x-speedup-mask-based-parallel-inference-paradigm-for-key-information-extraction-in-mllms

REST example

curl https://sciencetostartup.com/api/v1/agent-handoff/paper/up-to-36x-speedup-mask-based-parallel-inference-paradigm-for-key-information-extraction-in-mllms

MCP example

{
  "tool": "get_paper",
  "arguments": {
    "arxiv_id": "2601.19613"
  }
}

source_context

{
  "surface": "paper",
  "mode": "paper",
  "query": "Up to 36x Speedup: Mask-based Parallel Inference Paradigm for Key Information Extraction in MLLMs",
  "normalized_query": "2601.19613",
  "route": "/paper/up-to-36x-speedup-mask-based-parallel-inference-paradigm-for-key-information-extraction-in-mllms",
  "paper_ref": "up-to-36x-speedup-mask-based-parallel-inference-paradigm-for-key-information-extraction-in-mllms",
  "topic_slug": null,
  "benchmark_ref": null,
  "dataset_ref": null
}

Buildability Receipt

Verdict, compute envelope, blockers, signature state, and receipt links.

Paper proof page receipt window

Watch and verify: Up to 36x Speedup: Mask-based Parallel Inference Paradigm for Key Information Extraction in MLLMs

/buildability/up-to-36x-speedup-mask-based-parallel-inference-paradigm-for-key-information-extraction-in-mllms

Watchwatch

Subject: Up to 36x Speedup: Mask-based Parallel Inference Paradigm for Key Information Extraction in MLLMs

Verdict

Watch

Verdict is Watch because viability or proof quality is intermediate and should be re-evaluated before execution.

Time to first demo

Insufficient data

No first-demo timestamp, owner estimate, or elapsed demo receipt is attached to this surface.

Compute envelope

Structured compute envelope

Insufficient data

No data, compute, hardware, memory, latency, dependency, or serving requirement receipt is attached.

Evidence ids

Receipt path

/buildability/up-to-36x-speedup-mask-based-parallel-inference-paradigm-for-key-information-extraction-in-mllms

Paper ref

up-to-36x-speedup-mask-based-parallel-inference-paradigm-for-key-information-extraction-in-mllms

arXiv id

2601.19613

Freshness

Generated at

2026-04-02T02:30:40.136Z

Evidence freshness

stale

Last verification

2026-04-02T02:30:40.136Z

Sources

References

Coverage

17%

Hash state

Lineage hash

d5147ca5c513cc000985f92c74e53f860c78044163fe649e6277aaf85b09a120

Canonical opportunity-kernel lineage hash.

Signature state

External signature

unsigned_external

No founder, registry, pilot, or production-adoption signature is attached to this receipt.

Verification

not_verified

Verification is blocked until an external signature is provided.

Blockers

Missing: repo_url
Missing: references
Missing: proof_status
Missing: distribution_readiness_scores
Missing: paper_extraction_scorecards
Unknown: distribution readiness has not been computed yet
Unknown: proof verification has not been recorded yet

Verification pending / evidence receipt incomplete

repo_url

references

Missing proof, requirement, signature, approval, adoption, or telemetry fields are blockers and must not be inferred.

Open receipt API receipt Build Loop Signal Canvas Proof divergence Divergence API Brier outcomes API

Source Proof anchors

Visual citations from the paper document graph.

Source proof

Visual citation anchors from the paper document graph.

8 anchors

proof blockPage 282%

This equation captures one of the core mathematical components of the system. xt = arg max x P(x | x1, . . . , xt−1, I). To address this bottleneck, we

Page and bbox are available; crop image is pending.

proof blockPage 282%

This equation captures one of the core mathematical components of the system. P(x1, . . . , xT | I) = t=1 P(xt | x1, . . . , xt−1, I).

Page and bbox are available; crop image is pending.

proof blockPage 282%

This equation captures one of the core mathematical components of the system. P(xM | x\M, I) = Y t∈M P(xt | x\M, I).

Page and bbox are available; crop image is pending.

JSON-LD twin

The application/ld+json payload rendered for agents.

{
  "@context": "https://schema.org",
  "@graph": [
    {
      "@type": "WebPage",
      "@id": "https://sciencetostartup.com/paper/up-to-36x-speedup-mask-based-parallel-inference-paradigm-for-key-information-extraction-in-mllms#webpage",
      "url": "https://sciencetostartup.com/paper/up-to-36x-speedup-mask-based-parallel-inference-paradigm-for-key-information-extraction-in-mllms",
      "name": "Up to 36x Speedup: Mask-based Parallel Inference Paradigm for Key Information Extraction in MLLMs",
      "description": "PIP offers a mask-based paradigm for a 5-36x speedup in key information extraction from documents, maintaining accuracy while enhancing efficiency.",
      "isPartOf": {
        "@id": "https://sciencetostartup.com/#website"
      }
    },
    {
      "@type": "ScholarlyArticle",
      "@id": "https://sciencetostartup.com/paper/up-to-36x-speedup-mask-based-parallel-inference-paradigm-for-key-information-extraction-in-mllms#scholarlyArticle",
      "headline": "Up to 36x Speedup: Mask-based Parallel Inference Paradigm for Key Information Extraction in MLLMs",
      "description": "PIP offers a mask-based paradigm for a 5-36x speedup in key information extraction from documents, maintaining accuracy while enhancing efficiency.",
      "url": "https://sciencetostartup.com/paper/up-to-36x-speedup-mask-based-parallel-inference-paradigm-for-key-information-extraction-in-mllms",
      "sameAs": "https://arxiv.org/abs/2601.19613",
      "identifier": {
        "@type": "PropertyValue",
        "propertyID": "arXiv",
        "value": "2601.19613"
      },
      "isAccessibleForFree": true,
      "isPartOf": {
        "@id": "https://sciencetostartup.com/#website"
      },
      "datePublished": "2026-01-27T13:45:30.000Z",
      "citation": [
        {
          "@type": "ScholarlyArticle",
          "identifier": {
            "@type": "PropertyValue",
            "propertyID": "SemanticScholar",
            "value": "5e8b50dd5ed9851a541d337a4d9cfa9b75c0a33c"
          },
          "url": "https://www.semanticscholar.org/paper/5e8b50dd5ed9851a541d337a4d9cfa9b75c0a33c"
        },
        {
          "@type": "ScholarlyArticle",
          "identifier": {
            "@type": "PropertyValue",
            "propertyID": "SemanticScholar",
            "value": "e05795cdb5b0c11e72016c79122777c5b3a7ed48"
          },
          "url": "https://www.semanticscholar.org/paper/e05795cdb5b0c11e72016c79122777c5b3a7ed48"
        },
        {
          "@type": "ScholarlyArticle",
          "identifier": {
            "@type": "PropertyValue",
            "propertyID": "SemanticScholar",
            "value": "0d11a9674b68216b92e08cf7617a93fbd3fb91f4"
          },
          "url": "https://www.semanticscholar.org/paper/0d11a9674b68216b92e08cf7617a93fbd3fb91f4"
        },
        {
          "@type": "ScholarlyArticle",
          "identifier": {
            "@type": "PropertyValue",
            "propertyID": "SemanticScholar",
            "value": "a22b23b65cdbc0335b872ed7c7a7206c710f8997"
          },
          "url": "https://www.semanticscholar.org/paper/a22b23b65cdbc0335b872ed7c7a7206c710f8997"
        },
        {
          "@type": "ScholarlyArticle",
          "identifier": {
            "@type": "PropertyValue",
            "propertyID": "SemanticScholar",
            "value": "043cdced22f17231ad62ef0b3ca2533651abe4ab"
          },
          "url": "https://www.semanticscholar.org/paper/043cdced22f17231ad62ef0b3ca2533651abe4ab"
        },
        {
          "@type": "ScholarlyArticle",
          "identifier": {
            "@type": "PropertyValue",
            "propertyID": "SemanticScholar",
            "value": "887306ee7ebd9eebd7faea24106ec8e8f1a50987"
          },
          "url": "https://www.semanticscholar.org/paper/887306ee7ebd9eebd7faea24106ec8e8f1a50987"
        },
        {
          "@type": "ScholarlyArticle",
          "identifier": {
            "@type": "PropertyValue",
            "propertyID": "SemanticScholar",
            "value": "01771d36e818c1355de7885a78e69eb274314535"
          },
          "url": "https://www.semanticscholar.org/paper/01771d36e818c1355de7885a78e69eb274314535"
        },
        {
          "@type": "ScholarlyArticle",
          "identifier": {
            "@type": "PropertyValue",
            "propertyID": "SemanticScholar",
            "value": "b14e5138d4d1f3577b2390541dc7b730a41bb651"
          },
          "url": "https://www.semanticscholar.org/paper/b14e5138d4d1f3577b2390541dc7b730a41bb651"
        },
        {
          "@type": "ScholarlyArticle",
          "identifier": {
            "@type": "PropertyValue",
            "propertyID": "SemanticScholar",
            "value": "c811bedbe8f4c21d0cba9f9175f7c2eb203284a7"
          },
          "url": "https://www.semanticscholar.org/paper/c811bedbe8f4c21d0cba9f9175f7c2eb203284a7"
        },
        {
          "@type": "ScholarlyArticle",
          "identifier": {
            "@type": "PropertyValue",
            "propertyID": "SemanticScholar",
            "value": "1e0049ecb981fa356f984641585b21fd19e4e254"
          },
          "url": "https://www.semanticscholar.org/paper/1e0049ecb981fa356f984641585b21fd19e4e254"
        },
        {
          "@type": "ScholarlyArticle",
          "identifier": {
            "@type": "PropertyValue",
            "propertyID": "SemanticScholar",
            "value": "a3a273bbd169ed6e67c2735449bf609662a42cb0"
          },
          "url": "https://www.semanticscholar.org/paper/a3a273bbd169ed6e67c2735449bf609662a42cb0"
        },
        {
          "@type": "ScholarlyArticle",
          "identifier": {
            "@type": "PropertyValue",
            "propertyID": "SemanticScholar",
            "value": "104b0bb1da562d53cbda87aec79ef6a2827d191a"
          },
          "url": "https://www.semanticscholar.org/paper/104b0bb1da562d53cbda87aec79ef6a2827d191a"
        },
        {
          "@type": "ScholarlyArticle",
          "identifier": {
            "@type": "PropertyValue",
            "propertyID": "SemanticScholar",
            "value": "a9d5d97733ccb15002ff3cfb95b0a7d8ba5236e3"
          },
          "url": "https://www.semanticscholar.org/paper/a9d5d97733ccb15002ff3cfb95b0a7d8ba5236e3"
        },
        {
          "@type": "ScholarlyArticle",
          "identifier": {
            "@type": "PropertyValue",
            "propertyID": "SemanticScholar",
            "value": "8d75062df3a121e7926f6c79b70b8ffc571d5cff"
          },
          "url": "https://www.semanticscholar.org/paper/8d75062df3a121e7926f6c79b70b8ffc571d5cff"
        },
        {
          "@type": "ScholarlyArticle",
          "identifier": {
            "@type": "PropertyValue",
            "propertyID": "SemanticScholar",
            "value": "ca6a2bc279be5a3349a22bfd6866ed633d18734b"
          },
          "url": "https://www.semanticscholar.org/paper/ca6a2bc279be5a3349a22bfd6866ed633d18734b"
        },
        {
          "@type": "ScholarlyArticle",
          "identifier": {
            "@type": "PropertyValue",
            "propertyID": "SemanticScholar",
            "value": "a5036f31f0e629dc661f120b8c3b1f374d479ab8"
          },
          "url": "https://www.semanticscholar.org/paper/a5036f31f0e629dc661f120b8c3b1f374d479ab8"
        },
        {
          "@type": "ScholarlyArticle",
          "identifier": {
            "@type": "PropertyValue",
            "propertyID": "SemanticScholar",
            "value": "163b4d6a79a5b19af88b8585456363340d9efd04"
          },
          "url": "https://www.semanticscholar.org/paper/163b4d6a79a5b19af88b8585456363340d9efd04"
        },
        {
          "@type": "ScholarlyArticle",
          "identifier": {
            "@type": "PropertyValue",
            "propertyID": "SemanticScholar",
            "value": "57e849d0de13ed5f91d086936296721d4ff75a75"
          },
          "url": "https://www.semanticscholar.org/paper/57e849d0de13ed5f91d086936296721d4ff75a75"
        },
        {
          "@type": "ScholarlyArticle",
          "identifier": {
            "@type": "PropertyValue",
            "propertyID": "SemanticScholar",
            "value": "79d8ff95c1e5370692129eba6c00af50db9e3da5"
          },
          "url": "https://www.semanticscholar.org/paper/79d8ff95c1e5370692129eba6c00af50db9e3da5"
        },
        {
          "@type": "ScholarlyArticle",
          "identifier": {
            "@type": "PropertyValue",
            "propertyID": "SemanticScholar",
            "value": "c69942bf1b4f75e53cb62d0c5126c1cb4a5aa7bc"
          },
          "url": "https://www.semanticscholar.org/paper/c69942bf1b4f75e53cb62d0c5126c1cb4a5aa7bc"
        }
      ],
      "additionalProperty": [
        {
          "@type": "PropertyValue",
          "propertyID": "viabilityScore",
          "value": 7
        },
        {
          "@type": "PropertyValue",
          "propertyID": "researchDomain",
          "value": "Document AI"
        }
      ]
    },
    {
      "@type": "BreadcrumbList",
      "itemListElement": [
        {
          "@type": "ListItem",
          "position": 1,
          "name": "Home",
          "item": "https://sciencetostartup.com"
        },
        {
          "@type": "ListItem",
          "position": 2,
          "name": "Document AI",
          "item": "https://sciencetostartup.com/topics"
        },
        {
          "@type": "ListItem",
          "position": 3,
          "name": "Up to 36x Speedup: Mask-based Parallel Inference Paradigm fo",
          "item": "https://sciencetostartup.com/paper/up-to-36x-speedup-mask-based-parallel-inference-paradigm-for-key-information-extraction-in-mllms"
        }
      ]
    }
  ]
}

Up to 36x Speedup: Mask-based Parallel Inference Paradigm for Key Information Extraction in MLLMs

Up to 36x Speedup: Mask-based Parallel Inference Paradigm for Key Information Extraction in MLLMs

Claim map

Constellation map

Competitive landscape

Buzz

PDF

References(23)

Related Papers

Subscribe to the weekly brief

Build artifacts

Brief

Experiment plan

Validation checklist

Scientific founder

Translational engineer

Domain operator

GTM lead

Regulatory/clinical advisor

Timeline

Claim map

Constellation map

Competitive landscape

Buzz

PDF

References(23)

Related Papers

Subscribe to the weekly brief

Build artifacts

Brief

Experiment plan

Validation checklist

Scientific founder

Translational engineer

Domain operator

GTM lead

Regulatory/clinical advisor

Timeline