Pith. sign in

REVIEW 5 major objections 5 minor 48 references

KnowCoder-V2: Deep Knowledge Analysis

T0 review · 5 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read Treating knowledge as Python code, a 14B model claims to beat larger deep-research systems on report quality.

desk verdict A credible IE/KBQA system whose headline deep-research report claim is undermined by a non-independent LLM judge. read the letter →

arxiv 2506.06881 v1 pith:N5U5VHHR submitted 2025-06-07 cs.AI

Zixuan Li , Wenxuan Liu , Long Bai , Chunmao Zhang , Wei Li , Fenghui Zhang , Quanxin Jin , Ruoyun He
show 8 more authors
Zhuo Chen Zhilei Hu Fei Wang Bingbing Xu Xuhui Jiang Xiaolong Jin Jiafeng Guo Xueqi Cheng
This is my paper
classification cs.AI
keywords deepresearchknowledgeorganizationcomputationcodegenerationinformationextractionontologyexpansionbasequestionansweringreport
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper argues that today's deep research systems fall short when the task requires insight, because they treat knowledge as web pages and reason by writing text. The proposed fix is to make knowledge computable: define an ontology as Python classes, convert source documents into Python objects during an offline organization phase, and then let the model generate and run analysis code over those objects while writing the report. KnowCoder-V2, a 14-billion-parameter code model fine-tuned for this, uses import statements to recall class schemas and writes code both for extraction and for computing results such as citation counts and publication distributions. On more than thirty datasets the model is competitive or better on taxonomy building, entity and relation extraction, event extraction, and knowledge-base question answering; in a report-generation test on a self-constructed dataset about scientist achievements, the full framework scores 8.9 average versus 7.7 for the strongest commercial deep-research baseline, with the biggest gap on the 'insight' axis.

What carries the argument

The load-bearing mechanism is unified code generation built on a code-as-knowledge representation: ontology concepts are Python classes and knowledge instances are Python objects. Extraction becomes an import clause plus object instantiation; update is merging objects with the same name; reasoning is generating analysis code that runs over a retrieved search_results list. The framework surrounds this with a cycle of ontology search, code generation, instance query, and code execution, plus an evaluation agent that sends failed or unsatisfactory executions back through the loop. Ontology alignment uses the same code representation: candidate relations are retrieved by embedding concept definitions and then confirmed by an LLM reading the class code.

What would settle it

Re-run the report-generation comparison on a pre-registered third-party dataset with expert human raters who check every cited number against its source; if the KnowCoder-V2 reports no longer beat the strongest commercial deep-research system on average quality or on insight, the paper's headline claim fails. A cheaper check is to verify whether the code-generated reports contain citation counts or bibliographic entries that do not exist in the source documents.

Watch

Extended reading notes

Core claim

KnowCoder-V2's central claim is that deep research systems gain a new capability when the same code-generation mechanism is used for both organizing knowledge and computing over it. The ontology is written as Python classes, and extracted facts become Python objects; extraction is object instantiation, knowledge update is code merge, and analysis is executable code run against a list called search_results. Because the class definitions live in the model's parameters, the model can recall a schema with a short import clause instead of receiving all definitions in the prompt, which removes the prompt-length bottleneck of earlier code-based extraction. The paper reports that this design outperforms or matches strong baselines on taxonomy expansion, English and Chinese information extraction, robustness to perturbations, and WebQSP question answering, and that the KDR framework's reports are rated above commercial and open deep-research baselines on all five dimensions, with the largest advantage in insight (8.7 versus 4.1–5.3).

Load-bearing premise

The report-generation result assumes that scores given by two large-language-model judges, one of which powers parts of the same system, on a self-built and unpublished dataset about scientists are a valid measure of report quality and insight.

Editorial extensions

If this is right

  • Deep research no longer needs to treat each task as a fresh online search; shared domain knowledge can be organized once and reused across tasks.
  • Large-scale schemas stop being a prompt-length problem, because a model that has learned class definitions can activate them with an import statement.
  • Analyses that require counting, aggregating, comparing across timestamps, or charting become executable computations rather than prose guesswork.
  • A 14-billion-parameter fine-tuned model can match or exceed much larger commercial deep-research systems on report completeness, thoroughness, factuality, coherence, and insight, if the reported evaluations hold.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The 'insight' advantage probably comes from verifiable computation: citation counts, distributions, and charts can be checked, whereas free-text summaries cannot, so code execution may reduce fabricated numbers in generated reports.
  • The two-phase design suggests a natural test for other domains: pick any field with a stable ontology, build the offline knowledge base once, and compare report insight against retrieval-only systems.
  • If import-based schema recall is the active mechanism, improvements in code-pretrained base models should transfer directly to better knowledge organization with no new architectural change; scaling the base model would test this.
  • The human-in-the-loop ontology construction and the name-merge update rule are the least automated parts; replacing them with learned update and alignment could determine how far the approach scales beyond curated topics.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 5 minor

Summary. The paper proposes KnowCoder-V2, a 14B code-oriented LLM (fine-tuned from Qwen-2.5-Coder-14B-Instruct with LoRA) that represents ontologies as Python classes and knowledge instances as Python objects, and uses unified code generation both for knowledge organization (instantiation code) and knowledge reasoning (analysis code). The authors also introduce the Knowledgeable Deep Research (KDR) framework, which adds an offline knowledge organization phase and an online knowledge computation cycle to conventional deep research systems. Evaluation covers taxonomy expansion, English and Chinese information extraction (NER, RE, ED, EAE), a robustness benchmark (RUIE), KBQA on WebQSP, and a report-generation task on a self-constructed scientist-achievement dataset. The paper reports strong gains over prior IE models on most benchmarks, state-of-the-art Hits@1 on WebQSP, and higher LLM-judged report quality than OpenAI Deep Research, Grok3, Qwen3 Deep Research, and WebThinker.

Significance. If the headline claim is correct, the paper would show that a compact 14B model, combined with an offline knowledge base and code-based computation, can outperform large commercial deep research agents on report quality and insight, while also advancing the state of the art in universal information extraction. The paper's strengths include a broad evaluation across many standard IE benchmarks, a new robustness benchmark comparison, and a clean conceptual unification of organization and reasoning through code generation, with no fitted constants or circular derivations in the IE and KBQA experiments. The IE and KBQA results are independently meaningful and would likely stand on their own. However, the distinguishing claim made in the abstract and in Section 5.3.2 — that the KDR framework 'can generate high-quality reports with insightful analytical results' — is currently supported only by a self-constructed, unreleased dataset scored by two LLM judges, one of which is part of the evaluated system. The significance of the paper as a whole therefore depends on whether this evaluation weakness can be repaired; the underlying idea and the IE contributions are solid.

major comments (5)
  1. [§5.3.2, Table 6; Figure 3 caption] The report-generation claim, which is the paper's headline, rests on an evaluation that is not independent of the system. Scores are averaged from DeepSeek-V3 and DeepSeek-R1, and the Figure 3 caption explicitly states that DeepSeek-R1 'handle[s] the parts of the KDR framework that are not related to knowledge analysis tasks.' Thus one of the two judges is also part of the KDR pipeline being evaluated. The dataset is self-constructed and unreleased (Section 5.1), no human evaluation is reported, and no inter-judge agreement is given. This is a load-bearing weakness: the 8.9 versus 7.7 average advantage over OpenAI Deep Research (Table 6) may reflect a systematic judge preference for outputs that resemble KDR's own computational artifacts rather than a generalizable quality advantage. Please add a human evaluation on a sample of reports, use at least one judge that is not involved in the KDR pipeline, report agreement statistics, and release the dataset and generated reports. If these are not feasible, the deep-research claim should be substantially softened.
  2. [§4.3, §5.3.2] The paper states in Section 4.3 that for ontology alignment and knowledge reasoning tasks, 'we are currently generating the corresponding training data and plan to release the trained models in the future.' This means KnowCoder-V2 is not trained on report-generation or KBQA-style reasoning tasks in this paper. Yet the report-generation experiments in Table 6 attribute the improvement to KnowCoder-V2's knowledge computation capability. This is a mismatch between the method description and the evaluation: the reported gains could come from the KDR framework, the base model's general code ability, or the human-in-the-loop ontology design, rather than from the claimed fine-tuned capabilities of KnowCoder-V2. The authors should clarify exactly which components of KnowCoder-V2 are used in the report-generation and KBQA experiments, and ideally provide a version of the model that is actually trained on the reasoning tasks, or redesign the claims to match the evaluated system.
  3. [§5.3.2, Table 6] There is no ablation isolating the contribution of the knowledge computation cycle to report quality. The comparison in Table 6 is between the full KDR+KnowCoder-V2 system and closed- or open-source deep research systems, but it does not compare against KDR without the knowledge computation cycle, KDR with a base Qwen-2.5-Coder model, or KDR with a different LLM in place of KnowCoder-V2. As a result, the statement that 'Due to the knowledge computation steps, KnowCoder-V2 is able to generate more comprehensive and insightful experimental results' (Section 5.3.2) is not directly supported by the experiments. Such ablations are needed to attribute the reported 8.9 score to the proposed components.
  4. [§5.3.2, Table 5] The KBQA evaluation is limited to a single dataset, WebQSP, and the experimental setup uses the ground-truth topic entity as initial context, following ToG. The paper reports Hits@1 but no variance or significance testing, and no analysis of failure cases. Given that this is one of only two knowledge-reasoning evaluations and that the model is not trained on reasoning data (Section 4.3), a single dataset is too thin to support the claim that 'KnowCoder-V2 is capable of generating accurate analysis code for the KBQA task.' Additional KBQA benchmarks or a more careful statement of scope would strengthen this section.
  5. [§5.1, Table 1] Table 1 groups KnowCoder-V2 under the '1-shot Setting' alongside NL and CodeTaxo, but the paper never specifies what the one-shot exemplar is, how KnowCoder-V2 is prompted for taxonomy expansion, or how the two-stage alignment with SimCSE and LLM re-ranking (Section 4.1.2) is combined with the one-shot protocol. This makes the strong results on WordNet and Graphine difficult to reproduce and difficult to interpret. Please provide the exact prompt and the one-shot example, or clarify whether KnowCoder-V2 in Table 1 is used as a prompting method or as a fine-tuned model.
minor comments (5)
  1. [§5.1] The sentence 'Table ?? presents the statistics of these taxonomies' contains an unresolved table reference; the table and its statistics are missing from the manuscript.
  2. [Throughout] There are numerous typos and grammatical errors that should be corrected, including 'Insigt' for 'Insight' in Section 5.3.2, 'exising' in the Introduction, 'eveluate' in Section 5.1, 'trianed' in Section 4.1.2, and 'filed search_results' in Section 4.2.2.
  3. [Tables 2 and 3] The table headers are inconsistent: some columns are labeled 'Avg' while others are left blank or contain dashes, and KnowCoder-V1's Chinese NER entries are listed as 40.6 and 16.3 with no overall average. Please unify the notation and explain all dashes.
  4. [Section 5.3.1] The paper reports average improvements of 3.03% and 2.92% F1 over 'all baselines' but does not report standard deviations or significance tests. Given that many baseline differences are within 1-2 points, it would be helpful to state whether the overall averages are computed across the same set of datasets for all systems.
  5. [Section 4.1.3] The claim that internalizing class definitions via 'import' improves efficiency is plausible, but no prompt-length or wall-clock efficiency measurement is reported in the experiments, even though Section 4.1.3 motivates this design by efficiency. A small table or figure quantifying prompt-length reduction would make the efficiency claim concrete.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the report-generation judge overlap is an evaluation-independence caveat, not a circular derivation.

full rationale

The paper's claimed derivation is not circular by construction. KnowCoder-V2 is a fine-tuned code-generation model whose knowledge-organization and KBQA results are evaluated against external benchmarks (taxonomy expansion datasets, NER/RE/ED/EAE corpora, WebQSP), so those results do not depend on the paper's own conclusions. The class-based representation is adopted from KnowCoder-V1 by design choice, not used as evidence for correctness, and the RobustUIE/AlignXIE self-citations supply augmentation pipelines rather than load-bearing support for the headline claim. The report-generation evaluation is the weakest point: the dataset is self-constructed ('there are no suitable datasets'), the scores are averaged from DeepSeek-V3 and DeepSeek-R1, and DeepSeek-R1 also handles parts of KDR (Figure 3 caption). That overlap could inflate the 8.9 versus 7.7 margin, and the paper itself notes that training data for ontology-alignment and reasoning tasks is still to be released. But this is an empirical soundness and reproducibility risk, not a case where a prediction is equivalent to its input: no equation, fitted parameter, or definition makes the report score a function of the system's own design. No self-definitional reduction, fitted-input prediction, or load-bearing self-citation chain is present, so the circularity score is low.

Assumptions & free parameters 6 free parameters · 5 assumptions · 0 invented entities

The central claim does not rest on a mathematical derivation, so the ledger is mostly about representation and evaluation choices. The model adapts a known code-as-ontology idea from the authors' KnowCoder-V1, adds fine-tuning and data augmentation, and embeds it in a deep research loop. The main unevidenced choices are the human-in-the-loop ontology validation, the heuristic knowledge update, and the LLM-judge evaluation of report quality. No new physical or formal entities are introduced.

free parameters (6)
  • LoRA rank = 32
    Adapter rank chosen by hand; defined in Section 4.3. It controls the capacity of the fine-tuned model, though the central claim does not hinge on this exact value.
  • Learning rate = 3e-4
    Training hyperparameter used for LoRA fine-tuning in Section 4.3.
  • Sequence length = 2048 tokens
    Limits the prompt and extraction context; relevant to the import-based class strategy.
  • Batch size = 256
    Training batch size reported in Section 4.3.
  • Inference temperature = 0
    Fixed decoding temperature; removes sampling variance in reported scores.
  • Augmented sample count = 20,000
    Loss-guided selection of informative robustness samples in Section 4.3.
assumptions (5)
  • domain assumption Python classes as ontology and objects as knowledge instances faithfully represent the domain knowledge needed for deep analysis.
    Section 4.1 and Figure 3; if this representation omits relational or contextual information, knowledge computation results will be incomplete.
  • domain assumption Human-in-the-loop concept validation is available for every new research task.
    Sections 3.1 and 4.1.1 require LLM-proposed concepts to be reviewed by human experts; this is not automated and limits scalability.
  • domain assumption DeepSeek-V3 and DeepSeek-R1 average scores measure report quality and insight.
    Section 5.3.2 uses only these LLM judges, with no human evaluation or external benchmark.
  • domain assumption Elasticsearch can retrieve the class definitions required for code generation.
    Section 4.2.2 Ontology Search depends on this retrieval step.
  • domain assumption Merging same-name entities and keeping the latest conflicting value preserves knowledge correctness.
    Section 4.1.4 explicitly calls this a heuristic and leaves more precise update mechanisms to future work.

how reviews work

0 comments
Cite this review

Pith. "Pith review of KnowCoder-V2: Deep Knowledge Analysis." pith.science (2026). https://pith.science/paper/N5U5VHHR

@misc{pith2026250606881,
  author       = {Pith},
  title        = {Pith review of: KnowCoder-V2: Deep Knowledge Analysis},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/N5U5VHHR}},
  note         = {Machine review of arXiv:2506.06881}
}
read the original abstract

Deep knowledge analysis tasks always involve the systematic extraction and association of knowledge from large volumes of data, followed by logical reasoning to discover insights. However, to solve such complex tasks, existing deep research frameworks face three major challenges: 1) They lack systematic organization and management of knowledge; 2) They operate purely online, making it inefficient for tasks that rely on shared and large-scale knowledge; 3) They cannot perform complex knowledge computation, limiting their abilities to produce insightful analytical results. Motivated by these, in this paper, we propose a \textbf{K}nowledgeable \textbf{D}eep \textbf{R}esearch (\textbf{KDR}) framework that empowers deep research with deep knowledge analysis capability. Specifically, it introduces an independent knowledge organization phase to preprocess large-scale, domain-relevant data into systematic knowledge offline. Based on this knowledge, it extends deep research with an additional kind of reasoning steps that perform complex knowledge computation in an online manner. To enhance the abilities of LLMs to solve knowledge analysis tasks in the above framework, we further introduce \textbf{\KCII}, an LLM that bridges knowledge organization and reasoning via unified code generation. For knowledge organization, it generates instantiation code for predefined classes, transforming data into knowledge objects. For knowledge computation, it generates analysis code and executes on the above knowledge objects to obtain deep analysis results. Experimental results on more than thirty datasets across six knowledge analysis tasks demonstrate the effectiveness of \KCII. Moreover, when integrated into the KDR framework, \KCII can generate high-quality reports with insightful analytical results compared to the mainstream deep research framework.

Figures

Figures reproduced from arXiv: 2506.06881 by the authors.

Figure 1
Figure 1. Report generation performance of the Knowledgeable Deep Research framework em [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Demonstration of the proposed knowledgeable deep research framework. The mustard yel [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Illustration of the Knowledgeable Deep Research Framework implemented by [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Two cases of reports generated by the KDR framework. [PITH_FULL_IMAGE:figures/full_fig_p014_4.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

48 extracted references · 25 canonical work pages

  1. [1]

    Gpt-4 technical report

    Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, et al. Gpt-4 technical report. arXiv preprint arXiv:2303.08774,

    arXiv

  2. [7]

    IEPile: Unearthing Large-Scale Schema-Based Information Extraction Corpus

    Honghao Gui, Lin Yuan, Hongbin Ye, Ningyu Zhang, Mengshu Sun, Lei Liang, and Huajun Chen. Iepile: Unearthing large-scale schema-based information extraction corpus. arXiv preprint arXiv:2402.14710, 2024a. Honghao Gui, Lin Yuan, Hongbin Ye, Ningyu Zhang, Mengshu Sun, Lei Liang, and Huajun Chen. Iepile: Unearthing large-scale schema-based information extrac...

    work page Pith review arXiv

  3. [8]

    Harsha Gurulingappa, Abdul Mateen Rajput, Angus Roberts, Juliane Fluck, Martin Hofmann-Apitius, and Luca Toldo

    URL https://arxiv.org/abs/2311.02962. Harsha Gurulingappa, Abdul Mateen Rajput, Angus Roberts, Juliane Fluck, Martin Hofmann-Apitius, and Luca Toldo. Development of a benchmark corpus to support the automatic extraction of drug-related adverse effects from medical case reports. Journal of Biomedical Informatics, 45(5): 885–892,

    arXiv

  4. [11]

    SemEval-2010 task 8: Multi-way classification of semantic relations between pairs of nominals

    Iris Hendrickx, Su Nam Kim, Zornitsa Kozareva, Preslav Nakov, Diarmuid Ó Séaghdha, Sebastian Padó, Marco Pennacchiotti, Lorenza Romano, and Stan Szpakowicz. SemEval-2010 task 8: Multi-way classification of semantic relations between pairs of nominals. In Katrin Erk and Carlo Strapparava (eds.), Proceedings of the 5th International Workshop on Semantic Eva...

    work page 2010

  5. [12]

    URL https: //arxiv.org/abs/2106.09685. Binyuan Hui, Jian Yang, Zeyu Cui, Jiaxi Yang, Dayiheng Liu, Lei Zhang, Tianyu Liu, Jiajun Zhang, Bowen Yu, Keming Lu, Kai Dang, Yang Fan, Yichang Zhang, An Yang, Rui Men, Fei Huang, Bo Zheng, Yibo Miao, Shanghaoran Quan, Yunlong Feng, Xingzhang Ren, Xuancheng Ren, Jingren Zhou, and Junyang Lin. Qwen2.5-coder technica...

    arXiv

  6. [13]

    Jinhao Jiang, Kun Zhou, Zican Dong, Keming Ye, Wayne Xin Zhao, and Ji-Rong Wen

    URL https: //arxiv.org/abs/2409.12186. Jinhao Jiang, Kun Zhou, Zican Dong, Keming Ye, Wayne Xin Zhao, and Ji-Rong Wen. Structgpt: A general framework for large language model to reason over structured data, 2023a. Song Jiang, Qiyue Yao, Qifan Wang, and Yizhou Sun. A single vector is not enough: Taxonomy expansion via box embeddings. In Proceedings of the ...

    arXiv 2023

  7. [17]

    URL https://doi.org/10.1093/database/ baw068

    doi: 10.1093/database/baw068. URL https://doi.org/10.1093/database/ baw068. Peng Li, Tianxiang Sun, Qiong Tang, Hang Yan, Yuanbin Wu, Xuanjing Huang, and Xipeng Qiu. Codeie: Large code generation models are better few-shot information extractors, 2023a. URL https://arxiv.org/abs/2305.05711. 16 KnowCoder Tianle Li, Xueguang Ma, Alex Zhuang, Yu Gu, Yu Su, a...

    arXiv

  8. [18]

    URL https://doi.org/10.1007/ 978-3-030-60457-8_44

    doi: 10.1007/978-3-030-60457-8\_44. URL https://doi.org/10.1007/ 978-3-030-60457-8_44 . Zixuan Li, Yutao Zeng, Yuxin Zuo, Weicheng Ren, Wenxuan Liu, Miao Su, Yucan Guo, Yantao Liu, Lixiang Lixiang, Zhilei Hu, et al. Knowcoder: Coding structured knowledge into llms for universal information extraction. In Proceedings of the 62nd Annual Meeting of the Assoc...

    work page doi:10.1007/978-3-030-60457-8

Show all 48 references
  1. [19]

    Towards event extraction with massive types: Llm-based collaborative annotation and partitioning extraction

    Wenxuan Liu, Zixuan Li, Long Bai, Yuxin Zuo, Daozhu Xu, Xiaolong Jin, Jiafeng Guo, and Xueqi Cheng. Towards event extraction with massive types: Llm-based collaborative annotation and partitioning extraction. arXiv preprint arXiv:2503.02628,

    arXiv

  2. [22]

    Taxollama: Wordnet-based model for solving multiple lexical semantic tasks

    Viktor Moskvoretskii, Ekaterina Neminova, Alina Lobanova, Alexander Panchenko, and Irina Nikishina. Taxollama: Wordnet-based model for solving multiple lexical semantic tasks. arXiv preprint arXiv:2403.09207,

    arXiv

  3. [23]

    Cross-lingual name tagging and linking for 282 languages

    17 KnowCoder Xiaoman Pan, Boliang Zhang, Jonathan May, Joel Nothman, Kevin Knight, and Heng Ji. Cross-lingual name tagging and linking for 282 languages. In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pp. 194...

    1946

  4. [25]

    URL https: //doi.org/10.1093/bioinformatics/btt580

    doi: 10.1093/BIOINFORMATICS/BTT580. URL https: //doi.org/10.1093/bioinformatics/btt580. Sebastian Riedel, Limin Yao, and Andrew McCallum. Modeling relations and their mentions without labeled text. In ECML/PKDD,

    doi

  5. [27]

    URL https://arxiv.org/abs/2310.03668. Erik F. Tjong Kim Sang and Fien De Meulder. Introduction to the conll-2003 shared task: Language- independent named entity recognition. In Walter Daelemans and Miles Osborne (eds.),Proceedings of the Seventh Conference on Natural Language ...

    2003 arXiv

  6. [29]

    V34I05.6401

    doi: 10.1609/AAAI. V34I05.6401. URL https://doi.org/10.1609/aaai.v34i05.6401. Zhihong Shao, Yeyun Gong, Yelong Shen, Minlie Huang, Nan Duan, and Weizhu Chen. Enhancing retrieval-augmented large language models with iterative retrieval-generation synergy. arXiv preprint arXiv:2...

    arXiv

  7. [30]

    Taxoexpan: Self-supervised taxonomy expansion with position-enhanced graph neural network

    Jiaming Shen, Zhihong Shen, Chenyan Xiong, Chi Wang, Kuansan Wang, and Jiawei Han. Taxoexpan: Self-supervised taxonomy expansion with position-enhanced graph neural network. InProceedings of The Web Conference 2020, pp. 486–497,

    2020

  8. [31]

    A unified taxonomy-guided instruction tun- ing framework for entity set expansion and taxonomy expansion

    Yanzhen Shen, Yu Zhang, Yunyi Zhang, and Jiawei Han. A unified taxonomy-guided instruction tun- ing framework for entity set expansion and taxonomy expansion. arXiv preprint arXiv:2402.13405,

    arXiv

  9. [32]

    TIARA: Multi-grained retrieval for robust question answering over large knowledge base

    Yiheng Shu, Zhiwei Yu, Yuhan Li, Börje Karlsson, Tingting Ma, Yuzhong Qu, and Chin-Yew Lin. TIARA: Multi-grained retrieval for robust question answering over large knowledge base. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pp. 81...

    2022

  10. [34]

    R1-searcher: Incentivizing the search capability in llms via reinforcement learning

    Huatong Song, Jinhao Jiang, Yingqian Min, Jie Chen, Zhipeng Chen, Wayne Xin Zhao, Lei Fang, and Ji-Rong Wen. R1-searcher: Incentivizing the search capability in llms via reinforcement learning. arXiv preprint arXiv:2503.05592,

    arXiv

  11. [35]

    Think-on-graph: Deep and responsible reasoning of large language model on knowledge graph

    18 KnowCoder Jiashuo Sun, Chengjin Xu, Lumingyuan Tang, Saizhuo Wang, Chen Lin, Yeyun Gong, Lionel M Ni, Heung-Yeung Shum, and Jian Guo. Think-on-graph: Deep and responsible reasoning of large language model on knowledge graph. arXiv preprint arXiv:2307.07697,

    arXiv

  12. [36]

    Low-resource taxonomy enrichment with pretrained language models

    Kunihiro Takeoka, Kosuke Akimoto, and Masafumi Oyamada. Low-resource taxonomy enrichment with pretrained language models. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pp. 2747–2758,

    2021

  13. [37]

    MultiNERD: A multilingual, multi-genre and fine-grained dataset for named entity recognition (and disambiguation)

    Simone Tedeschi and Roberto Navigli. MultiNERD: A multilingual, multi-genre and fine-grained dataset for named entity recognition (and disambiguation). In Findings of the Association for Computational Linguistics: NAACL 2022 , pp. 801–812, Seattle, United States, July

    2022

  14. [38]

    doi: 10.18653/v1/2022.findings-naacl.60

    Association for Computational Linguistics. doi: 10.18653/v1/2022.findings-naacl.60. URL https://aclanthology.org/2022.findings-naacl.60. Christopher Walker, Stephanie Strassel, Julie Medero, and Kazuaki Maeda. Ace 2005 multilingual training corpus,

    2022 doi

  15. [39]

    Keheng Wang, Feiyu Duan, Sirui Wang, Peiguang Li, Yunsen Xian, Chuantao Yin, Wenge Rong, and Zhang Xiong

    URL https://catalog.ldc.upenn.edu/LDC2006T06. Keheng Wang, Feiyu Duan, Sirui Wang, Peiguang Li, Yunsen Xian, Chuantao Yin, Wenge Rong, and Zhang Xiong. Knowledge-driven cot: Exploring faithful reasoning in llms for knowledge-intensive question answering, 2023a. Suyuchen Wang, ...

    2022

  16. [40]

    Instructuie: Multi-task instruction tuning for unified information extraction, 2023b

    Xiao Wang, Weikang Zhou, Can Zu, Han Xia, Tianze Chen, Yuansen Zhang, Rui Zheng, Junjie Ye, Qi Zhang, Tao Gui, Jihua Kang, Jingsheng Yang, Siyuan Li, and Chunsai Du. Instructuie: Multi-task instruction tuning for unified information extraction, 2023b. URL https://arxiv. org/ab...

    arXiv

  17. [41]

    Agentic reasoning: Reasoning llms with tools for the deep research

    Junde Wu, Jiayuan Zhu, and Yuyuan Liu. Agentic reasoning: Reasoning llms with tools for the deep research. arXiv preprint arXiv:2502.04644,

    arXiv

  18. [42]

    Hongyuan Xu, Yunong Chen, Zichen Liu, Yanlong Wen, and Xiaojie Yuan

    URL https://arxiv.org/abs/2312.15548. Hongyuan Xu, Yunong Chen, Zichen Liu, Yanlong Wen, and Xiaojie Yuan. Taxoprompt: A prompt- based generation method with taxonomic context for self-supervised taxonomy expansion. In IJCAI, pp. 4432–4438,

    arXiv

  19. [43]

    Activerag: Autonomously knowledge assimilation and accommodation through retrieval-augmented agents

    Zhipeng Xu, Zhenghao Liu, Yukun Yan, Shuo Wang, Shi Yu, Zheni Zeng, Chaojun Xiao, Zhiyuan Liu, Ge Yu, and Chenyan Xiong. Activerag: Autonomously knowledge assimilation and accommodation through retrieval-augmented agents. arXiv preprint arXiv:2402.13547,

  20. [44]

    Donghan Yu, Sheng Zhang, Patrick Ng, Henghui Zhu, Alexander Hanbo Li, Jun Wang, Yiqun Hu, William Wang, Zhiguo Wang, and Bing Xiang

    URL https://arxiv.org/abs/2406.11192. Donghan Yu, Sheng Zhang, Patrick Ng, Henghui Zhu, Alexander Hanbo Li, Jun Wang, Yiqun Hu, William Wang, Zhiguo Wang, and Bing Xiang. Decaf: Joint decoding of answers and logical forms for question answering over knowledge bases,

    arXiv

  21. [45]

    Codetaxo: Enhancing taxonomy expansion with limited examples via code language prompts

    19 KnowCoder Qingkai Zeng, Yuyang Bai, Zhaoxuan Tan, Zhenyu Wu, Shangbin Feng, and Meng Jiang. Codetaxo: Enhancing taxonomy expansion with limited examples via code language prompts. arXiv preprint arXiv:2408.09070,

    arXiv

  22. [47]

    URL http://arxiv.org/abs/2403.13372

    Association for Computational Linguistics. URL http://arxiv.org/abs/2403.13372. Yuxiang Zheng, Dayuan Fu, Xiangkun Hu, Xiaojie Cai, Lyumanshan Ye, Pengrui Lu, and Pengfei Liu. Deepresearcher: Scaling deep research via reinforcement learning in real-world environments. arXiv pr...

    arXiv

  23. [48]

    Jizhao Zhu, Akang Shi, Zixuan Li, Long Bai, Xiaolong Jin, Jiafeng Guo, and Xueqi Cheng

    URL https: //arxiv.org/abs/2308.03279. Jizhao Zhu, Akang Shi, Zixuan Li, Long Bai, Xiaolong Jin, Jiafeng Guo, and Xueqi Cheng. Towards robust universal information extraction: Benchmark, evaluation, and solution. arXiv preprint arXiv:2503.03201,

    arXiv

  24. [49]

    Alignxie: Improving multilingual information extraction by cross-lingual alignment

    Yuxin Zuo, Wenxuan Jiang, Wenxuan Liu, Zixuan Li, Long Bai, Hanbin Wang, Yutao Zeng, Xiaolong Jin, Jiafeng Guo, and Xueqi Cheng. Alignxie: Improving multilingual information extraction by cross-lingual alignment. arXiv preprint arXiv:2411.04794,

    arXiv

  25. [1996]

    From local to global: A graph rag approach to query-focused summarization

    Darren Edge, Ha Trinh, Newman Cheng, Joshua Bradley, Alex Chao, Apurva Mody, Steven Truitt, Dasha Metropolitansky, Robert Osazuwa Ness, and Jonathan Larson. From local to global: A graph rag approach to query-focused summarization. arXiv preprint arXiv:2404.16130,

    arXiv

  26. [2003]

    Taneeya Satyapanich, Francis Ferraro, and Tim Finin

    URL https://aclanthology.org/W03-0419/. Taneeya Satyapanich, Francis Ferraro, and Tim Finin. CASIE: extracting cybersecurity event information from text. In The Thirty-Fourth AAAI Conference on Artificial Intelligence, AAAI 2020, The Thirty-Second Innovative Applications of Ar...

    2020

  27. [2006]

    Jiao Li, Yueping Sun, Robin J

    URL https://aclanthology.org/W06-0115/. Jiao Li, Yueping Sun, Robin J. Johnson, Daniela Sciaky, Chih-Hsuan Wei, Robert Leaman, Allan Peter Davis, Carolyn J. Mattingly, Thomas C. Wiegers, and Zhiyong Lu. Biocreative V CDR task corpus: a resource for chemical disease relation ex...

    2016

  28. [2010]

    A linear programming formulation for global inference in natural language tasks

    Dan Roth and Wen-tau Yih. A linear programming formulation for global inference in natural language tasks. In Proceedings of the Eighth Conference on Computational Natural Language Learning (CoNLL-2004) at HLT-NAACL 2004 , pp. 1–8, Boston, Massachusetts, USA, May 6 - May 7

    2004

  29. [2012]

    doi: https://doi.org/10.1016/j.jbi.2012.04.008

    ISSN 1532-0464. doi: https://doi.org/10.1016/j.jbi.2012.04.008. URL https: //www.sciencedirect.com/science/article/pii/S1532046412000615. Text Mining and Natural Language Processing in Pharmacogenomics. 15 KnowCoder Gaole He, Yunshi Lan, Jing Jiang, Wayne Xin Zhao, and Ji-Rong...

    2012 doi

  30. [2014]

    Semeval-2016 task 13: Taxonomy extraction evaluation (texeval-2)

    Georgeta Bordea, Els Lefever, and Paul Buitelaar. Semeval-2016 task 13: Taxonomy extraction evaluation (texeval-2). In Proceedings of the 10th international workshop on semantic evaluation (semeval-2016), pp. 1081–1091,

    2016

  31. [2016]

    Learning to reason with search for llms via reinforcement learning

    Mingyang Chen, Tianpeng Li, Haoze Sun, Yijie Zhou, Chenzheng Zhu, Fan Yang, Zenan Zhou, Weipeng Chen, Haofen Wang, Jeff Z Pan, et al. Learning to reason with search for llms via reinforcement learning. arXiv preprint arXiv:2503.19470,

    arXiv

  32. [2017]

    doi: 10.18653/v1/P17-1178

    Association for Computational Linguistics. doi: 10.18653/v1/P17-1178. URL https://aclanthology.org/P17-1178. Sampo Pyysalo and Sophia Ananiadou. Anatomical entity mention recognition at literature scale. Bioinform., 30(6):868–875,

    doi

  33. [2018]

    doi: 10.18653/v1/P18-1144

    Association for Computational Linguistics. doi: 10.18653/v1/P18-1144. URL [https://aclanthology.org/P18-1144](https: //aclanthology.org/P18-1144). Wayne Xin Zhao, Kun Zhou, Junyi Li, Tianyi Tang, Xiaolei Wang, Yupeng Hou, Yingqian Min, Beichen Zhang, Junjie Zhang, Zican Dong, ...

    doi

  34. [2019]

    Zequn Liu, Shukai Wang, Yiyang Gu, Ruiyi Zhang, Ming Zhang, and Sheng Wang

    URL http://arxiv.org/abs/1906.02437. Zequn Liu, Shukai Wang, Yiyang Gu, Ruiyi Zhang, Ming Zhang, and Sheng Wang. Graphine: A dataset for graph-aware terminology definition generation. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pp...

    1906 arXiv

  35. [2020]

    URL https://doi.org/10.1007/978-3-030-68763-2_48

    doi: 10.1007/978-3-030-68763-2\_48. URL https://doi.org/10.1007/978-3-030-68763-2_48 . Gina-Anne Levow. The third international chinese language processing bakeoff: Word segmentation and named entity recognition. In Hwee Tou Ng and Olivia O. Y . Kwong (eds.),Proceedings of the...

    doi

  36. [2021]

    URL https://doi.org/10.1145/3437963.3441753

    doi: 10.1145/3437963.3441753. URL https://doi.org/10.1145/3437963.3441753. Xiaoxin He, Yijun Tian, Yifei Sun, Nitesh Chawla, Thomas Laurent, Yann LeCun, Xavier Bresson, and Bryan Hooi. G-retriever: Retrieval-augmented generation for textual graph understanding and question ans...

  37. [2022]

    doi: 10.18653/v1/2022.emnlp-main.555

    Association for Computational Linguistics. doi: 10.18653/v1/2022.emnlp-main.555. URL https://aclanthology.org/ 2022.emnlp-main.555. Charlie Snell, Jaehoon Lee, Kelvin Xu, and Aviral Kumar. Scaling llm test-time compute optimally can be more effective than scaling model paramet...

    2022 arXiv

  38. [2023]

    Open deep search: Democratizing search with open-source reasoning agents

    Salaheddin Alzubi, Creston Brooks, Purva Chiniya, Edoardo Contente, Chiara von Gerlach, Lu- cas Irwin, Yihan Jiang, Arda Kaz, Windsor Nguyen, Sewoong Oh, et al. Open deep search: Democratizing search with open-source reasoning agents. arXiv preprint arXiv:2503.20201,

    arXiv

  39. [2024]

    Simcse: Simple contrastive learning of sentence embeddings

    Tianyu Gao, Xingcheng Yao, and Danqi Chen. Simcse: Simple contrastive learning of sentence embeddings. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pp

    2021

  40. [2025]

    Biomedical named entity recognition at scale

    Veysel Kocaman and David Talby. Biomedical named entity recognition at scale. In Alberto Del Bimbo, Rita Cucchiara, Stan Sclaroff, Giovanni Maria Farinella, Tao Mei, Marco Bertini, Hugo Jair Escalante, and Roberto Vezzani (eds.),Pattern Recognition. ICPR International Workshop...

    2021

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.