Pith. sign in

REVIEW 15 cited by

InstructUIE: Multi-task Instruction Tuning for Unified Information Extraction

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2304.08085 v1 pith:QCUI23Z2 submitted 2023-04-17 cs.CL cs.AI

Xiao Wang , Weikang Zhou , Can Zu , Han Xia , Tianze Chen , Yuansen Zhang , Rui Zheng , Junjie Ye
show 6 more authors
Qi Zhang Tao Gui Jihua Kang Jingsheng Yang Siyuan Li Chunsai Du
This is my paper · ORCID
classification cs.CLcs.AI
keywords extractioninformationunifiedinstructioninstructionsinstructuielargemethod
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Large language models have unlocked strong multi-task capabilities from reading instructive prompts. However, recent studies have shown that existing large models still have difficulty with information extraction tasks. For example, gpt-3.5-turbo achieved an F1 score of 18.22 on the Ontonotes dataset, which is significantly lower than the state-of-the-art performance. In this paper, we propose InstructUIE, a unified information extraction framework based on instruction tuning, which can uniformly model various information extraction tasks and capture the inter-task dependency. To validate the proposed method, we introduce IE INSTRUCTIONS, a benchmark of 32 diverse information extraction datasets in a unified text-to-text format with expert-written instructions. Experimental results demonstrate that our method achieves comparable performance to Bert in supervised settings and significantly outperforms the state-of-the-art and gpt3.5 in zero-shot settings.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 15 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. LA-RL: Label-Aware Self-Reflection for Reinforcement Learning in Information Extraction

    cs.CL 2026-07 conditional novelty 6.0 of 10

    Label-aware diagnostic reflection plus two-stage outcome GRPO improves same-backbone IE F1 over SFT, with larger gains under relation-extraction domain shift.

  2. Enhancing Automatic Term Extraction with Large Language Models via Syntactic Retrieval

    cs.CL 2025-06 conditional novelty 6.0 of 10

    Syntactic similarity retrieval of demonstrations improves LLM-based automatic term extraction in cross-domain settings, but gains are modest and in-domain lexical retrieval is often competitive or better.

  3. GuideX: Guided Synthetic Data Generation for Zero-Shot Information Extraction

    cs.CL 2025-05 conditional novelty 6.0 of 10

    A fully automated pipeline for generating annotation schemas, guidelines, and synthetic labeled examples from documents improves zero-shot NER after fine-tuning.

  4. Small Language Model Makes an Effective Long Text Extractor

    cs.CL 2025-02 conditional novelty 6.0 of 10

    A smaller span-based NER model with a compressed plus-shaped attention mechanism extracts long entities from very long texts with less memory than prior span-based methods.

  5. DE-NER : Zero-shot Named Entity Recognition via Dialogue Elicitation of Large Language Models

    cs.CL 2026-08 conditional novelty 5.0 of 10

    A self-play dialogue framework with a self-trained questioner improves zero-shot named entity recognition over basic prompting, but not consistently over the strongest existing methods.

  6. RMPL: Relation-aware Multi-task Progressive Learning with Stage-wise Training for Multimedia Event Extraction

    cs.CL 2026-02 conditional novelty 5.0 of 10

    A two-stage training pipeline—unified schema warm-up on heterogeneous supervision followed by task-specific fine-tuning—improves low-resource multimedia event extraction on M2E2 across three VLMs.

  7. MR-UIE: Multi-Perspective Reasoning with Reinforcement Learning for Universal Information Extraction

    cs.CL 2025-09 conditional novelty 5.0 of 10

    A pipeline that combines multi-perspective chain-of-thought reasoning with reinforcement learning for universal information extraction, showing modest gains that are overstated in the text.

  8. Joint Information Extraction Across Classical and Modern Chinese with Tea-MOELoRA

    cs.CL 2025-09 conditional novelty 5.0 of 10

    Tea-MOELORA uses separate task and era gates over LoRA experts to jointly train relation and event extraction across classical and modern Chinese, improving F1 over joint LoRA and existing LoRA-MoE baselines on most datasets.

  9. Skill-based Explanations for Serendipitous Course Recommendation

    cs.AI 2025-08 reject novelty 5.0 of 10

    A user study of skill-based explanations in a course recommender found no significant overall effect on interest, unexpectedness, or serendipity, but a significant reduction in neutral responses among undeclared students.

  10. Selecting and Merging: Towards Adaptable and Scalable Named Entity Recognition with Large Language Models

    cs.CL 2025-06 conditional novelty 5.0 of 10

    SaM selects and merges a few domain-expert LoRA models at inference time, outperforming a unified NER model by about 10% F1 on CrossNER and MIT.

  11. KnowCoder-V2: Deep Knowledge Analysis

    cs.AI 2025-06 conditional novelty 5.0 of 10

    KnowCoder-V2 augments deep research with offline knowledge organization and code-based knowledge computation, reporting gains on information extraction, KBQA, and LLM-judged report generation.

  12. Named Entity Recognition in Historical Italian: The Case of Giacomo Leopardi's Zibaldone

    cs.CL 2025-05 conditional novelty 5.0 of 10

    A fine-tuned GliNER model reaches 68.98% exact F1 and 75.64% fuzzy F1 on a new Italian historical NER benchmark, outperforming zero-shot LLaMa3.1-8B and zero-shot GliNER.

  13. DIRECT: Direct Decoding for Efficient and Aligned Sequence Labeling with Large Language Models

    cs.CL 2026-07 conditional novelty 4.0 of 10

    DPO after SFT plus constrained template-filling decoding improves LLM sequence-labeling accuracy and cuts inference time by reusing KV cache for non-label tokens.

  14. PlanE: Meta Planning of Data, Tuning, and Inference for Extractive-based LLMs

    cs.AI 2026-05 conditional novelty 4.0 of 10

    A quadratic meta-planner trained on a few model-dataset runs selects the optimal data-tuning-inference configuration for extractive LLMs, matching grid search on three IE tasks.

  15. Low-Resource Fine-Tuning for Multi-Task Structured Information Extraction with a Billion-Parameter Instruction-Tuned Model

    cs.CL 2025-09 reject novelty 3.0 of 10

    A 1B LLaMA model fine-tuned with LoRA on 100-1000 synthetic samples shows high ROUGE-L and JSON parse rates on three extraction tasks, but the comparison against zero-shot 7B/8B models does not support the claim that ...

Pith tools