Pith. sign in

REVIEW 2 major objections 6 minor 45 references

SUCEA: Reasoning-Intensive Retrieval for Adversarial Fact-checking through Claim Decomposition and Editing

T0 review · 2 major / 6 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 Adversarial fact-checking improves when first-round retrieval guides rewriting of decomposed sub-claims before a second retrieval pass, lifting accuracy by 7.5 points on FOOL METWICE.

desk verdict SUCEA's decompose-retrieve-edit-retrieve recipe is genuinely new and worth peer review, but its headline 'significant' gains are not statistically anchored on samples of 200. read the letter →

arxiv 2506.04583 v1 pith:IO6VZLNN submitted 2025-06-05 cs.CL cs.AI

Hongjun Liu , Yilun Zhao , Arman Cohan , Chen Zhao This is my paper
classification cs.CLcs.AI
keywords adversarialfact-checkingclaimdecompositiondecontextualizationevidence-guidededitingiterativeretrievalreasoning-intensiveFOOLMETWICEWICE
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

Adversarial fact-checking claims are deliberately written to avoid lexical overlap with the evidence that would verify them, so retrieval-augmented systems fail before the reasoning stage. The paper claims this can be fixed without fine-tuning by making the query itself more evidence-like: split the claim into atomic, context-free sub-claims; retrieve once purely for hints; rewrite each sub-claim with names, numbers, and corrections taken from that evidence; then retrieve again and predict the label from the combined evidence. On FOOL METWICE the framework lifts fact-checking accuracy by 7.5 points over a retrieval-augmented baseline with Llama-3.1-70B and improves top-10 retrieval accuracy by 11.0 points under TFIDF, with smaller consistent gains on WICE. A sympathetic reading of the results is that evidence-grounded claim editing, not decomposition alone, is what makes reasoning-intensive retrieval tractable for adversarial claims.

What carries the argument

The load-bearing mechanism is the evidence-augmented claim editing loop. After claim segmentation and decontextualization turn the input into atomic standalone sub-claims, a first retrieval round returns top-k passages that are used only as hints; an LLM then rewrites each sub-claim under explicit constraints to add missing named entities, numerical values, and locations and to replace vague or adversarial wording with evidence-backed details, while being told not to add parametric knowledge. The edited sub-claim is submitted for a second retrieval round, which is what recovers the ground-truth evidence that the original adversarial phrasing hid. Segmentation reduces distraction for the retriever, and a final reranking step selects the most relevant evidence before the entailment label is predicted.

What would settle it

Take any FOOL METWICE claim, run SUCEA as described, then rerun it with the first-round evidence replaced by random unrelated Wikipedia passages before claim editing; if accuracy stays near 73.5%, the evidence-guidance is not load-bearing, and if it collapses, the stated dependence on first-round hints is confirmed.

Watch

Extended reading notes

Core claim

The central claim is that the bottleneck in adversarial fact-checking is retrievability, not reasoning: the evidence exists in the corpus, but the claim has been paraphrased, stripped of names, or reworded so the retriever cannot find it. SUCEA turns the retrieval query into the object of repair by first decomposing the claim into decontextualized sub-claims, then running a provisional retrieval whose top passages act as hints, then instructing an LLM to rewrite each sub-claim to fill in missing entities, numbers, and locations and to correct counterfactual phrasing using only information present in those passages. A second retrieval on the edited sub-claims, followed by LLM reranking and label prediction, produces the reported gains: 65.5% to 73.5% accuracy on FOOL METWICE with Llama-3.1-70B, a 7.5-point improvement, and an 11.0-point gain in top-10 retrieval accuracy under TFIDF, with consistent improvements on WICE as well.

Load-bearing premise

The framework's second-round retrieval and final verdict depend on the first-round passages being topically close enough to the truth that the LLM's rewrite moves toward the evidence rather than away from it.

Editorial extensions

If this is right

  • On FOOL METWICE, SUCEA improves fact-checking accuracy from 65.5% to 73.5% with Llama-3.1-70B, and the gain is largest for supported claims, which rise from 52.4% to 64.1%.
  • Retrieval accuracy at Top-10 under TFIDF rises from 23.5% to 34.5%, showing that edited sub-claims are substantially easier for a purely lexical retriever to match.
  • Ablations show that editing without evidence guidance loses up to 8 points of retrieval accuracy under TFIDF, and removing segmentation costs up to 7 points, so both modules carry the result.
  • The framework works without fine-tuning across backbone LLMs of different sizes and also improves accuracy on WICE, indicating the mechanism transfers to long-form, non-adversarial claims.
  • Retrieval quality keeps improving with additional retrieve-and-edit rounds, with the largest gains in the first two iterations.

Reading between the lines

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

  • Beyond the paper: if evidence-grounded query editing is the active ingredient, the same pattern should transfer to other reasoning-intensive retrieval tasks, such as retrieving code from a natural-language description or finding documents for multi-hop questions, where the query is also deliberately abstract.
  • Beyond the paper: the method has no guardrail for a failed first round, so a cheap extension is to detect low-confidence first-round retrieval and widen k, switch retrievers, or abstain from editing; the paper's error analysis, which finds 15 of 50 failures tied to parametric knowledge, suggests a second check comparing the edited sub-claim against the first-round passages could reduce contaminati
  • Beyond the paper: because gains are larger for TFIDF than for Contriever, the framework's benefit may shrink as dense retrievers become more paraphrase-robust; a testable prediction is that SUCEA's advantage over a stronger dense retriever will be smaller.
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

2 major / 6 minor

Summary. The paper proposes SUCEA, a training-free modular framework for adversarial fact-checking. The pipeline decomposes each claim into decontextualized sub-claims, performs a first round of retrieval, uses the retrieved passages as hints to rewrite each sub-claim, retrieves again with the edited queries, reranks the aggregated evidence with an LLM, and finally predicts an entailment label. Experiments on FOOL METWICE (200 sampled claims) and WICE (358 claims) with two backbone LLMs (GPT-4o-mini and Llama-3.1-70B) and two retrievers (Contriever and TFIDF) report consistent improvements in retrieval accuracy and end-to-end fact-checking accuracy over RALM and four decomposition-based baselines. The paper also includes ablations, qualitative case studies, an error analysis, and an appendix with full results and prompts.

Significance. If the improvements are reliable, the paper makes a useful contribution: it shows that evidence-grounded, iterative claim editing, without any fine-tuning, can make lexical and dense retrievers substantially more effective on claims deliberately written to evade lexical overlap, and it provides a modular decomposition that is easy to reuse. The paper is honest about its limitations, includes a thoughtful error analysis, and releases code. The main weakness is that the central "significant improvement" claims are not backed by any statistical inference: all numbers are single point estimates on 200 or 358 claims, with no confidence intervals, significance tests, or repeated runs. This is fixable and does not invalidate the approach, but it must be addressed before the headline claims are acceptable.

major comments (2)
  1. [Abstract, §4.2, Tables 2 and 3] The abstract and Section 4.2 use the word "significantly" to describe SUCEA's improvements, but no significance test, confidence interval, or variance estimate is reported anywhere in the paper. On the 200-claim FOOL METWICE subset, the headline fact-checking gap of 75.0 vs. 67.5 corresponds to only about 15 claims; a two-proportion z-test on these counts gives z ≈ 1.74 (p ≈ 0.08), which is not significant at the conventional 95% level. Several WICE differences are even smaller; for example, Table 2 shows SUCEA at 39.0% versus PROGRAMFC at 38.4% under GPT-4o-mini with Contriever. The paper also does not report the random seed or stratification used to draw the 200-claim subset, so the error cannot be reconstructed. I recommend adding paired significance tests (e.g., McNemar for label accuracy, bootstrap intervals for retrieval metrics) and reporting them for all tables, and adjusting the "significant" wording to what the data actually support.
  2. [§5.1, Table 4] The ablation section claims that "all components significantly contribute" and that removing any component leads to "notable reduction," but several ablation drops under Contriever are within one or two points on the same 200-claim set; for example, Table 4 shows wo. claim editing at 49.5 vs. 51.0 for Llama-3.1-70B and Paraphrase wo. Evidence at 50.5 vs. 51.0, and wo. claim editing at 53.0 vs. 54.0 for GPT-4o-mini. These differences (1-2 claims) are within sampling noise, so the universal claim of significant contribution is not supported by the reported numbers. The TFIDF columns show larger drops and are more convincing. The authors should either provide significance evidence for all ablation cells or restrict the conclusion to the configurations where the effect is clearly larger than noise.
minor comments (6)
  1. [Table 1] The header "F OOLMETWICE" contains an extra space; change to "FOOL METWICE."
  2. [§5.1] The phrase "the need of guidence with evidence" contains a typo; it should read "guidance."
  3. [§4.2] The word "contriver" is used instead of "Contriever" in the sentence "when using contriver"; please fix for consistency.
  4. [References] Kamoi et al. is listed twice (2023a and 2023b) with the same WiCE paper; please consolidate into a single entry.
  5. [§4.1 / Experiments] The paper should report the random seed and sampling procedure used to select the 200-claim FOOL METWICE subset, to make the evaluation reproducible.
  6. [§4.2, Table 5] The statement that SUCEA "outperforms all baseline approaches" on WICE is contradicted in the full results in Table 5: at Top@3 with Llama-3.1-70B and Contriever, MINICHECK reaches 39.1% while SUCEA reaches 38.2%. Please qualify the claim to "in most configurations" or refer to the primary setting.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: SUCEA's evidence-guided editing loop is an explicit, ablated design choice rather than a prediction derived from its own inputs.

full rationale

SUCEA contains no fitted parameters, no definitional identities, and no load-bearing self-citations. The first-round evidence is used to rewrite sub-claims before second-round retrieval, which creates a feedback loop; however, the paper does not present this loop as a first-principles derivation, and the final entailment label is predicted from the original claim C against the aggregated evidence. The Paraphrase-wo-Evidence ablation (Table 4) directly tests the causal role of the evidence hint and shows that removing it degrades retrieval accuracy, so the contribution of evidence-guided editing is empirical and falsifiable rather than assumed. The error analysis in Section 5.3 documents cases where editing injects parametric knowledge, overgenerates, or produces overly fine-grained sub-claims, causing retrieval failure; this confirms that success is not guaranteed by construction. Self-citations (Zhao et al. 2024a-c) appear only as background and related-work references and do not carry the argument. The abstract's use of 'significantly' without reported significance tests is a statistical-reporting concern, not a circularity concern. No circular step can be exhibited, so the circularity score is at floor.

Assumptions & free parameters 1 free parameters · 4 assumptions · 0 invented entities

The method is training-free, so no fitted constants are used; one hand-chosen hyperparameter, the iteration count, is listed for completeness. The main assumptions are dataset-corpus coverage, correctness of gold evidence, and faithfulness of LLM segmentation and editing. No new entities are postulated.

free parameters (1)
  • Number of retrieval-editing rounds = 2 (default)
    SUCEA runs one round of editing and a second retrieval by default. Figure 10 shows that additional rounds continue to improve retrieval accuracy, so the default is a hand-chosen early stopping point rather than an optimized maximum.
assumptions (4)
  • domain assumption Gold evidence annotations in FOOL METWICE and WICE are complete and correct.
    Retrieval accuracy and final label accuracy are measured against these annotations; incomplete or noisy gold evidence would distort both. The datasets are described in Section 2.2 and used as ground truth in Section 4.1.
  • domain assumption The Dec. 20, 2018 Wikipedia corpus contains passages sufficient to verify every test claim.
    Retrieval cannot succeed if the corpus predates or lacks the relevant information. The fixed corpus is specified in Section 4.1 without a coverage analysis for either dataset.
  • ad hoc to paper LLM segmentation, decontextualization, and editing preserve the original claim's truth conditions.
    The framework assumes rewritten sub-claims are faithful and self-contained; Section 5.3 documents 15/50 parametric-knowledge errors and 22/50 over-segmentation errors in a 50-case sample, so the assumption is load-bearing and imperfect.
  • ad hoc to paper First-round retrieval returns top-k passages with enough correct anchors to guide editing.
    Editing is grounded only in the first-round passages (Section 3.2); if all top-k passages are irrelevant, the edited query inherits the error. The paper has no fallback for this case.

how reviews work

0 comments
Cite this review

Pith. "Pith review of SUCEA: Reasoning-Intensive Retrieval for Adversarial Fact-checking through Claim Decomposition and Editing." pith.science (2026). https://pith.science/paper/IO6VZLNN

@misc{pith2026250604583,
  author       = {Pith},
  title        = {Pith review of: SUCEA: Reasoning-Intensive Retrieval for Adversarial Fact-checking through Claim Decomposition and Editing},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IO6VZLNN}},
  note         = {Machine review of arXiv:2506.04583}
}
read the original abstract

Automatic fact-checking has recently received more attention as a means of combating misinformation. Despite significant advancements, fact-checking systems based on retrieval-augmented language models still struggle to tackle adversarial claims, which are intentionally designed by humans to challenge fact-checking systems. To address these challenges, we propose a training-free method designed to rephrase the original claim, making it easier to locate supporting evidence. Our modular framework, SUCEA, decomposes the task into three steps: 1) Claim Segmentation and Decontextualization that segments adversarial claims into independent sub-claims; 2) Iterative Evidence Retrieval and Claim Editing that iteratively retrieves evidence and edits the subclaim based on the retrieved evidence; 3) Evidence Aggregation and Label Prediction that aggregates all retrieved evidence and predicts the entailment label. Experiments on two challenging fact-checking datasets demonstrate that our framework significantly improves on both retrieval and entailment label accuracy, outperforming four strong claim-decomposition-based baselines.

Figures

Figures reproduced from arXiv: 2506.04583 by the authors.

Figure 1
Figure 1. Examples of challenges in fact-checking ad [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. An illustration of SUCEA pipeline (left) and its workflow for adversarial fact-checking using an ex￾ample from the FOOLMETWICE dataset (right). The input claim is first segmented and decontextualized by LLMs. For each sub-claim, a first round of evidence retrieval is conducted, followed by sub-claim revision and a second round of evidence retrieval. The final step is label prediction for the claim. dataset is collec… view at source ↗
Figure 3
Figure 3. Case studies demonstrating how SUCEA handles adversarial claims through Claim Segmentation and Claim Editing modules. Left:The original claim is split into sub-claims, enabling successful evidence retrieval. Right: Three cases showing different types of claim editing - missing key information, synonym substitution, and context omission - where the system edits claims to improve retrieval performance. The numbers (e.… view at source ↗
Figures from the paper (7 more)
Figure 4
Figure 4. Figure 4: Error analysis showing three main failure cases identified from a random sample of 50 cases: (1) [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]
Figure 5
Figure 5. Figure 5: Prompt template used for the Claim Segmenta [PITH_FULL_IMAGE:figures/full_fig_p011_5.png]
Figure 6
Figure 6. Figure 6: Prompt template for the Claim Decontextu [PITH_FULL_IMAGE:figures/full_fig_p014_6.png]
Figure 8
Figure 8. Figure 8: Prompt template used for LLM-based evi￾dence reranking. The prompt instructs the LLM to sort retrieved passages based on their relevance to the query. Adapted from (Sun et al., 2023) [PITH_FULL_IMAGE:figures/full_fig_p014_8.png]
Figure 7
Figure 7. Figure 7: Prompt template for the Evidence-Augmented [PITH_FULL_IMAGE:figures/full_fig_p014_7.png]
Figure 9
Figure 9. Figure 9: Prompt template for the entailment label pre [PITH_FULL_IMAGE:figures/full_fig_p015_9.png]
Figure 10
Figure 10. Figure 10: Ablation study results on FOOLMETWICE under multiple-round iteration setting. The performance of both GPT-4o-mini (left) and Llama-3.1-70B (right) improves as the number of iterations increases, with evaluation metrics shown at R@3, R@5, and R@10. Starting from the or…

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

45 extracted references · 12 canonical work pages

  1. [1]

    URL: " 'urlintro :=

    ENTRY address author booktitle chapter edition editor howpublished institution journal key month note number organization pages publisher school series title type volume year eprint doi pubmed url lastchecked label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block STRINGS urlintro eprinturl eprintpr...

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize " " * FUNCT...

  3. [3]

    AI@Meta. 2024. http://arxiv.org/abs/2407.21783 The llama 3 herd of models

    arXiv 2024

  4. [4]

    Mubashara Akhtar, Rami Aly, Christos Christodoulopoulos, Oana Cocarascu, Zhijiang Guo, Arpit Mittal, Michael Schlichtkrull, James Thorne, and Andreas Vlachos, editors. 2023 a . https://aclanthology.org/2023.fever-1.0/ Proceedings of the Sixth Fact Extraction and VERification Workshop (FEVER) . Association for Computational Linguistics, Dubrovnik, Croatia

    work page 2023

  5. [5]

    Mubashara Akhtar, Michael Schlichtkrull, Zhijiang Guo, Oana Cocarascu, Elena Simperl, and Andreas Vlachos. 2023 b . https://doi.org/10.18653/v1/2023.findings-emnlp.361 Multimodal automated fact-checking: A survey . In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 5430--5448, Singapore. Association for Computational Linguistics

    doi:10.18653/v1/2023.findings-emnlp.361 2023

  6. [6]

    Akari Asai, Zexuan Zhong, Danqi Chen, Pang Wei Koh, Luke Zettlemoyer, Hannaneh Hajishirzi, and Wen-tau Yih. 2024. Reliable, adaptable, and attributable language models with retrieval. arXiv preprint arXiv:2403.03187

    arXiv 2024

  7. [7]

    Orlando Ayala and Patrice Bechard. 2024. Reducing hallucination in structured outputs via retrieval-augmented generation. In Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 6: Industry Track), pages 228--238

    work page 2024

  8. [8]

    Max Bartolo, Alastair Roberts, Johannes Welbl, Sebastian Riedel, and Pontus Stenetorp. 2020. https://doi.org/10.1162/tacl_a_00338 Beat the ai: Investigating adversarial human annotation for reading comprehension . Transactions of the Association for Computational Linguistics, 8:662–678

    doi:10.1162/tacl_a_00338 2020

Show all 45 references
  1. [9]

    Canyu Chen and Kai Shu. 2023. Combating misinformation in the age of llms: Opportunities and challenges. AI Magazine

    2023

  2. [10]

    Jifan Chen, Aniruddh Sriram, Eunsol Choi, and Greg Durrett. 2022. http://arxiv.org/abs/2205.06938 Generating literal and implied subquestions to fact-check complex claims

    2022 arXiv

  3. [11]

    Julian Eisenschlos, Bhuwan Dhingra, Jannis Bulian, Benjamin B \"o rschinger, and Jordan Boyd-Graber. 2021 a . https://doi.org/10.18653/v1/2021.naacl-main.32 Fool me twice: Entailment from W ikipedia gamification . In Proceedings of the 2021 Conference of the North American Cha...

    2021 doi

  4. [12]

    Julian Martin Eisenschlos, Bhuwan Dhingra, Jannis Bulian, Benjamin Börschinger, and Jordan Boyd-Graber. 2021 b . http://arxiv.org/abs/2104.04725 Fool me twice: Entailment from wikipedia gamification

    2021 arXiv

  5. [13]

    Angela Fan, Aleksandra Piktus, Fabio Petroni, Guillaume Wenzek, Marzieh Saeidi, Andreas Vlachos, Antoine Bordes, and Sebastian Riedel. 2020. http://arxiv.org/abs/2011.05448 Generating fact checking briefs

    2020 arXiv

  6. [14]

    Luyu Gao, Zhuyun Dai, Panupong Pasupat, Anthony Chen, Arun Tejasvi Chaganty, Yicheng Fan, Vincent Zhao, Ni Lao, Hongrae Lee, Da-Cheng Juan, and Kelvin Guu. 2023. https://doi.org/10.18653/v1/2023.acl-long.910 RARR : Researching and revising what language models say, using langu...

    2023 doi

  7. [15]

    Yunfan Gao, Yun Xiong, Xinyu Gao, Kangxiang Jia, Jinliu Pan, Yuxi Bi, Yi Dai, Jiawei Sun, Meng Wang, and Haofen Wang. 2024. http://arxiv.org/abs/2312.10997 Retrieval-augmented generation for large language models: A survey

    2024 arXiv

  8. [16]

    Siddhant Garg and Goutham Ramakrishnan. 2020. https://doi.org/10.18653/v1/2020.emnlp-main.498 BAE : BERT -based adversarial examples for text classification . In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 6174--6181, O...

    2020 doi

  9. [17]

    Google. 2024. https://ai.google.dev/gemma/docs Gemma 2 model card

    2024

  10. [18]

    Shreya Goyal, Sumanth Doddapaneni, Mitesh M Khapra, and Balaraman Ravindran. 2023. A survey of adversarial defenses and robustness in nlp. ACM Computing Surveys, 55(14s):1--39

    2023

  11. [19]

    Dirk Groeneveld, Iz Beltagy, Evan Walsh, Akshita Bhagia, Rodney Kinney, Oyvind Tafjord, Ananya Jha, Hamish Ivison, Ian Magnusson, Yizhong Wang, Shane Arora, David Atkinson, Russell Authur, Khyathi Chandu, Arman Cohan, Jennifer Dumas, Yanai Elazar, Yuling Gu, Jack Hessel, Tusha...

    2024

  12. [20]

    Anisha Gunjal and Greg Durrett. 2024. Molecular facts: Desiderata for decontextualization in llm fact verification. arXiv preprint arXiv:2406.20079

    2024

  13. [21]

    Gautier Izacard, Mathilde Caron, Lucas Hosseini, Sebastian Riedel, Piotr Bojanowski, Armand Joulin, and Edouard Grave. 2022. http://arxiv.org/abs/2112.09118 Unsupervised dense information retrieval with contrastive learning

    2022 arXiv

  14. [22]

    Ziwei Ji, Nayeon Lee, Rita Frieske, Tiezheng Yu, Dan Su, Yan Xu, Etsuko Ishii, Ye Jin Bang, Andrea Madotto, and Pascale Fung. 2023. https://dl.acm.org/doi/10.1145/3571730 Survey of hallucination in natural language generation . ACM Computing Surveys, 55(12):1--38

    2023 doi

  15. [23]

    Robin Jia and Percy Liang. 2017. https://doi.org/10.18653/v1/D17-1215 Adversarial examples for evaluating reading comprehension systems . In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, pages 2021--2031, Copenhagen, Denmark. Associati...

    2017 doi

  16. [24]

    Ryo Kamoi, Tanya Goyal, Juan Diego Rodriguez, and Greg Durrett. 2023 a . https://doi.org/10.18653/v1/2023.emnlp-main.470 W i CE : Real-world entailment for claims in W ikipedia . In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 7...

    2023 doi

  17. [25]

    Ryo Kamoi, Tanya Goyal, Juan Diego Rodriguez, and Greg Durrett. 2023 b . http://arxiv.org/abs/2303.01432 Wice: Real-world entailment for claims in wikipedia

    2023 arXiv

  18. [26]

    Vladimir Karpukhin, Barlas Oğuz, Sewon Min, Patrick Lewis, Ledell Wu, Sergey Edunov, Danqi Chen, and Wen tau Yih. 2020. http://arxiv.org/abs/2004.04906 Dense passage retrieval for open-domain question answering

    2020 arXiv

  19. [27]

    Chaofan Li, MingHao Qin, Shitao Xiao, Jianlyu Chen, Kun Luo, Yingxia Shao, Defu Lian, and Zheng Liu. 2024. http://arxiv.org/abs/2409.15700 Making text embedders few-shot learners

    2024 arXiv

  20. [28]

    Sewon Min, Kalpesh Krishna, Xinxi Lyu, Mike Lewis, Wen tau Yih, Pang Wei Koh, Mohit Iyyer, Luke Zettlemoyer, and Hannaneh Hajishirzi. 2023 a . http://arxiv.org/abs/2305.14251 Factscore: Fine-grained atomic evaluation of factual precision in long form text generation

    2023 arXiv

  21. [29]

    Sewon Min, Kalpesh Krishna, Xinxi Lyu, Mike Lewis, Wen-tau Yih, Pang Koh, Mohit Iyyer, Luke Zettlemoyer, and Hannaneh Hajishirzi. 2023 b . https://doi.org/10.18653/v1/2023.emnlp-main.741 FA ct S core: Fine-grained atomic evaluation of factual precision in long form text genera...

    2023 doi

  22. [30]

    Mistral.AI. 2023. https://mistral.ai/news/mixtral-of-experts/ Mixtral of experts: A high quality sparse mixture-of-experts

    2023

  23. [31]

    Yixin Nie, Adina Williams, Emily Dinan, Mohit Bansal, Jason Weston, and Douwe Kiela. 2020. https://doi.org/10.18653/v1/2020.acl-main.441 Adversarial NLI : A new benchmark for natural language understanding . In Proceedings of the 58th Annual Meeting of the Association for Comp...

    2020 doi

  24. [32]

    OpenAI. 2023. https://api.semanticscholar.org/CorpusID:257532815 Gpt-4 technical report . ArXiv, abs/2303.08774

    2023 arXiv

  25. [33]

    Liangming Pan, Xiaobao Wu, Xinyuan Lu, Anh Tuan Luu, William Yang Wang, Min-Yen Kan, and Preslav Nakov. 2023. http://arxiv.org/abs/2305.12744 Fact-checking complex claims with program-guided reasoning

    2023 arXiv

  26. [34]

    M Towhidul Islam Tonmoy, Krishnav Rajbangshi, Shravani Nag, Aman Chadha, Amit P

    Vipula Rawte, S. M Towhidul Islam Tonmoy, Krishnav Rajbangshi, Shravani Nag, Aman Chadha, Amit P. Sheth, and Amitava Das. 2024. http://arxiv.org/abs/2403.19113 Factoid: Factual entailment for hallucination detection

    2024 arXiv

  27. [35]

    Hinrich Sch \"u tze, Christopher D Manning, and Prabhakar Raghavan. 2008. Introduction to information retrieval, volume 39. Cambridge University Press Cambridge

    2008

  28. [36]

    Hongjin Su, Howard Yen, Mengzhou Xia, Weijia Shi, Niklas Muennighoff, Han-yu Wang, Haisu Liu, Quan Shi, Zachary S Siegel, Michael Tang, et al. 2024. Bright: A realistic and challenging benchmark for reasoning-intensive retrieval. arXiv preprint arXiv:2407.12883

    2024 arXiv

  29. [37]

    Weiwei Sun, Lingyong Yan, Xinyu Ma, Shuaiqiang Wang, Pengjie Ren, Zhumin Chen, Dawei Yin, and Zhaochun Ren. 2023. http://arxiv.org/abs/2304.09542 Is chatgpt good at search? investigating large language models as re-ranking agents

    2023 arXiv

  30. [38]

    Liyan Tang, Philippe Laban, and Greg Durrett. 2024. https://doi.org/10.18653/v1/2024.emnlp-main.499 M ini C heck: Efficient fact-checking of LLM s on grounding documents . In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 8818--88...

    2024 doi

  31. [39]

    James Thorne, Andreas Vlachos, Christos Christodoulopoulos, and Arpit Mittal. 2018. https://doi.org/10.18653/v1/N18-1074 FEVER : a large-scale dataset for fact extraction and VER ification . In Proceedings of the 2018 Conference of the North A merican Chapter of the Associatio...

    2018 doi

  32. [40]

    Hugo Touvron, Louis Martin, Kevin Stone, Peter Albert, Amjad Almahairi, Yasmine Babaei, Nikolay Bashlykov, Soumya Batra, Prajjwal Bhargava, Shruti Bhosale, Dan Bikel, Lukas Blecher, Cristian Canton Ferrer, Moya Chen, Guillem Cucurull, David Esiobu, Jude Fernandes, Jeremy Fu, W...

    2023 arXiv

  33. [41]

    Eric Wallace, Shi Feng, Nikhil Kandpal, Matt Gardner, and Sameer Singh. 2019 a . Universal adversarial triggers for attacking and analyzing nlp. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference...

    2019

  34. [42]

    Eric Wallace, Pedro Rodriguez, Shi Feng, Ikuya Yamada, and Jordan Boyd-Graber. 2019 b . http://arxiv.org/abs/1809.02701 Trick me if you can: Human-in-the-loop generation of adversarial examples for question answering

    2019 arXiv

  35. [43]

    Yilun Zhao, Hongjun Liu, Yitao Long, Rui Zhang, Chen Zhao, and Arman Cohan. 2024 a . http://arxiv.org/abs/2311.09797 Financemath: Knowledge-intensive math reasoning in finance domains

    2024 arXiv

  36. [44]

    Yilun Zhao, Yitao Long, Yuru Jiang, Chengye Wang, Weiyuan Chen, Hongjun Liu, Yiming Zhang, Xiangru Tang, Chen Zhao, and Arman Cohan. 2024 b . http://arxiv.org/abs/2411.05764 Findver: Explainable claim verification over long and hybrid-content financial documents

    2024 arXiv

  37. [45]

    Yilun Zhao, Yitao Long, Hongjun Liu, Ryo Kamoi, Linyong Nan, Lyuhao Chen, Yixin Liu, Xiangru Tang, Rui Zhang, and Arman Cohan. 2024 c . http://arxiv.org/abs/2311.09805 Docmath-eval: Evaluating math reasoning capabilities of llms in understanding long and specialized documents

    2024 arXiv

Pith tools

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