Pith. sign in

REVIEW 3 major objections 5 minor 2 cited by

RAGentA: Multi-Agent Retrieval-Augmented Generation for Attributed Question Answering

T0 review · 3 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read A four-agent RAG pipeline reports +10.7% faithfulness over standard RAG, with correctness nearly unchanged.

desk verdict Four-agent RAG pipeline with fine-grained citations is clearly described, but the headline faithfulness gain may be an evaluation artifact because the judge likely saw different passage sets for baseline vs. system. read the letter →

arxiv 2506.16988 v2 pith:WLB5KUVG submitted 2025-06-20 cs.IR

Ines Besrour , Jingbo He , Tobias Schreieder , Michael Färber This is my paper · ORCID
classification cs.IR
keywords retrieval-augmentedgenerationmulti-agentsystemsattributedquestionansweringin-linecitationsfaithfulnesshybridretrievalanswercorrectnesslargelanguagemodels
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 tries to show that a retrieval-augmented generation system can produce answers that are both correct and faithfully grounded in sources by splitting the work among four specialized LLM agents. The main claim is that adding an agent that writes in-line citations and a reviser agent that checks whether every part of the question has been answered materially improves faithfulness, the degree to which answers are supported by retrieved documents. On the authors' 500-question synthetic benchmark, the framework scores 0.7044 for faithfulness against 0.6362 for a standard single-prompt RAG pipeline (+10.7%), while correctness rises only slightly from 0.8256 to 0.8348 (+1.1%). The framework also combines sparse and dense retrieval, improving Recall@20 by 12.5% over the best single retriever. If the gains hold under independent evaluation, the framework offers a training-free recipe for making LLM answers verifiable in settings where users need to check sources.

What carries the argument

The load-bearing mechanism is the interplay between Agent-3's claim-level citation generation and Agent-4's completeness check. Agent-3 emits each fact as a claim with markers in the standardized [X] format, converting grounding from an implicit property into an output that can be inspected and scored. Agent-4 parses the query into sub-components, tags each one as fully, partially, or not answered, and for unanswered gaps issues follow-up queries to the retrieval system, excluding documents already seen, then merges the supplemental answers. The other load-bearing piece is the hybrid retrieval score $S_{\text{hybrid}}(d) = \alpha S_{\text{sparse}}(d) + (1-\alpha)S_{\text{dense}}(d)$ with $\alpha=0.35$, which fuses exact-term matching with dense embeddings so the agents start from a broader relevant pool. Agent-2's relevance filter, scored as $\log p(\text{Yes}) - \log p(\text{No})$ with a dynamic threshold, then trims that pool before generation.

What would settle it

Mechanically append [X] citation markers to Standard RAG answers without changing their text and re-run the same LLM judge; if faithfulness jumps toward RAGentA's 0.70, the judge is rewarding formatting, not grounding. Alternatively, have human raters score a random sample of the 500 answers and compare the faithfulness gap.

Watch

Extended reading notes

Core claim

RAGentA's central claim is that reliability in RAG is improved less by better retrieval alone than by making attribution an explicit, inspectable step of generation. Four agents cooperate: an initial predictor generates a candidate answer for each retrieved document; a judge scores each document's usefulness and filters out noisy ones using a dynamic threshold; a final predictor writes the answer as claims, each with inline citation markers; and a reviser agent decomposes the original question, marks each portion as fully, partially, or not answered, and when gaps remain reformulates queries, retrieves new documents, and merges the new material into a unified answer. The authors report that this pipeline yields a +10.7% faithfulness gain over standard RAG with only +1.1% correctness gain, and that the hybrid sparse-dense retriever recovers 12.5% more relevant documents at rank 20 than the best single retriever. The faithfulness figure is the paper's central assertion: claim-level attribution plus second-pass completion creates answers that are better grounded in the retrieved evidence.

Load-bearing premise

The entire faithfulness advantage rests on the automatic judge used for scoring agreeing with the official challenge evaluator; no agreement study is reported, so a judge that rewards the presence of citation markers could explain the +10.7% gap.

Editorial extensions

If this is right

  • If the reported scores hold, any RAG deployment that already retrieves top-k documents can adopt the four-agent loop without retraining and obtain more source-grounded answers, at the cost of extra LLM calls.
  • In-line citations in the standardized [X] format make individual claims auditable, so users can verify each statement against its cited passage rather than trusting the whole answer.
  • The hybrid retrieval result implies that fusing sparse and dense signals is a cheap way to improve recall before any agent reasoning begins, since it outperforms both single retrievers on the benchmark.
  • Agent-4's dynamic refinement should matter most for multi-aspect questions, where a single retrieval pass often misses one component; the paper's fully/partially/not-answered decomposition directly targets that failure mode.
  • Because correctness gains are small and the authors note the second-stage retrieval currently adds limited value, the framework's practical advantage is in trustworthiness rather than raw accuracy.

Reading between the lines

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

  • Implicit in the results but not tested: the faithfulness gap may be inflated by the judge rewarding the mere presence of citation markers; a citation-precision audit, checking whether each [X] marker actually supports its claim, would separate formatting from true grounding.
  • The framework's advantage likely concentrates on the benchmark's multi-aspect and comparison questions, where Agent-4 can actually find missing pieces; a per-category breakdown would show where the gain lives and where correctness stays flat.
  • A practical extension suggested by the architecture is to make the number of revision rounds budget-aware, since Agent-4's second-stage retrieval is the main source of the noted computational overhead, and to early-stop when all components are tagged fully answered.
  • A testable variant would remove Agent-2's relevance filtering and keep only citation generation; if faithfulness stays near 0.70, filtering is not the driver, and simpler pipelines could be used.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 5 minor

Summary. The manuscript proposes RAGentA, a training-free multi-agent RAG framework for attributed question answering. Retrieval combines BM25 and E5 via a weighted interpolation (Eq. 1, alpha=0.35) to select top-20 documents. Four agents then work in sequence: Agent-1 generates per-document answers, Agent-2 filters documents using a log-probability relevance score with a dynamic threshold (n=0.5), Agent-3 produces a final answer with in-line citations, and Agent-4 checks completeness, optionally reformulates the query, retrieves additional documents, and merges revised answers. The system is evaluated on 500 synthetic QA pairs generated with DataMorgana from the FineWeb index. Reported results are Recall@20 of 0.5650 for the hybrid retriever versus 0.5020 for BM25 (+12.5%), correctness of 0.8346 versus 0.8256 for Standard RAG (+1.1%), and faithfulness of 0.7044 versus 0.6362 (+10.7%). The paper concludes that the multi-agent architecture and hybrid retrieval improve grounding and answer quality.

Significance. If the evaluation is sound, the paper offers a useful, reproducible, training-free recipe for attributed QA. The hybrid retrieval result is clearly positive and the architectural extension of MAIN-RAG with citation generation and a reviser agent is a plausible mechanism for improving faithfulness. Strengths include public code, adoption of previously published hyperparameters rather than test-set tuning, and an honest discussion of the small correctness gain and computational overhead. The central caveat is that the headline faithfulness and correctness comparison rests entirely on a local LLM judge whose input context appears asymmetric between the two systems and which has not been validated against the official LiveRAG autoevaluator; until that is addressed, the +10.7% faithfulness advantage is not yet demonstrated.

major comments (3)
  1. [§5.2, Table 3] The paper does not specify which document context the LLM judge receives for the Standard RAG baseline. For RAGentA the 'cited passages' are the documents selected by its in-line citations, while Standard RAG produces no citations; if the baseline was scored against the full top-20 retrieved set, the faithfulness metric measures different objects for the two systems. Selective citation can raise 'grounded in the provided passages' even if the answer is no better grounded in the shared corpus. Please state the exact judge input for both systems and re-run the comparison with identical passage sets (e.g., full top-20 for both, or the baseline's own extracted citations), and report both settings.
  2. [§5.2] The Llama-3.3-70B-Instruct judge is substituted for the closed-source Claude-3.5 Sonnet autoevaluator without any agreement study, calibration check, or human audit. The entire reported correctness and faithfulness comparison depends on this judge agreeing with the official evaluator, and citation-dense outputs could be rated more leniently by the local judge. Please report an agreement metric (e.g., Cohen's kappa or Spearman correlation) on a sample, or a human-audited subset of 50-100 items, and show that the relative ranking of RAGentA versus Standard RAG is stable under the official or human scoring.
  3. [§5.1, §5.2] All conclusions are based on a single synthetic dataset of 500 questions with no variance estimates or significance tests. The MRR@20 gain (+2.0%) and correctness gain (+1.1%) are small and could easily arise from sampling noise. Please report bootstrap confidence intervals or paired significance tests (e.g., Wilcoxon signed-rank) for the retrieval, correctness, and faithfulness comparisons, and ideally validate on an established benchmark such as ALCE or the official LiveRAG evaluation set.
minor comments (5)
  1. [Table 3, §5.2] Table 3 reports RAGentA correctness as 0.8346, while the text in §5.2 states 0.8348; please reconcile the discrepancy.
  2. [Abstract, §5.2] The abstract reports +1.09% and +10.72%, while §5.2 reports +1.1% and +10.7%; unify the numerical precision.
  3. [Figure 1] Figure 1 is dense and the paths from Agent-4 to reformulation, retrieval, and answer merging are difficult to follow; a step-numbered legend or a simplified sequence diagram would improve readability.
  4. [§4.1, Eq. (1)] The alpha=0.35 value is taken from a regulatory-text study [13]; briefly justify why this value transfers to open-domain QA over FineWeb, or report sensitivity to alpha in the retrieval results.
  5. [§4.2.4] The 'completeness check' in Agent-4 and the 'coverage' component of the correctness metric are conceptually related but not formally connected; a sentence defining how completeness is operationalized would help the reader.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: RAGentA's reported gains are empirical and its hyperparameters come from prior external work, not from fitting the test set.

full rationale

RAGentA does not present a derivation whose conclusion is built into its premises. The retrieval fusion in Eq. (1) uses α=0.35 adopted from Mosquera et al. [13], and the Agent-2 filtering threshold uses n=0.5 adopted from MAIN-RAG [2]; neither constant is fitted to the 500-question evaluation set, so the retrieval and answer-quality results are not forced by construction. The architecture is described as an extension of MAIN-RAG rather than justified by a self-citation chain, and none of the cited prior works overlaps with the current authors. The faithfulness metric is defined as grounding in retrieved documents (after Es et al. [3]), and RAGentA is designed to produce in-line citations; this aligns the system with the metric's definition but does not make the metric's score equal to the system's output by construction, since the judge still assesses whether the final answer is supported. The only substantive concern is an evaluation-protocol asymmetry: the judge receives "the predicted answer, the ground-truth answer, and the cited passages" (Sec. 5.2), and Standard RAG emits no citations, so the passage context given for the baseline is unspecified. That is a validity threat to the +10.7% faithfulness comparison, but it is not a circular reduction and would not change the circularity score under the specified patterns.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The claims rest on two external hyperparameters (alpha=0.35, n=0.5), a judge proxy whose fidelity to the official evaluator is unmeasured, and a synthetic dataset whose representativeness is asserted. No new physical entities are introduced. The agents are software components, not postulated entities requiring independent evidence.

free parameters (2)
  • alpha = 0.35
    Hybrid fusion weight in Eq. 1: S_hybrid = alpha*S_sparse + (1-alpha)*S_dense. Set from prior work (Mosquera et al.) on regulatory texts; not tuned on this benchmark, but it is still a chosen scalar that directly controls retrieval scores.
  • n = 0.5
    Filtering stringency in adjusted_tau_q = tau_q - n*sigma (Section 4.2.2). Adopted from MAIN-RAG's empirical analysis rather than re-tuned on RAGentA's dataset.
assumptions (4)
  • domain assumption The Llama-3.3-70B judge yields the same correctness and faithfulness ratings as the official Claude-3.5 Sonnet autoevaluator.
    Invoked in Section 5.2 as the reason the challenge evaluation can be reproduced locally; no agreement or calibration analysis is supplied.
  • domain assumption The 500 synthetic QA pairs from DataMorgana over the FineWeb index constitute a valid representative testbed for attributed QA.
    Section 3 describes generation and category distributions, but no human validation or comparison to a real QA benchmark (e.g., ALCE, LiveRAG official set) is reported.
  • domain assumption Hybrid fusion with alpha=0.35 from a regulatory-text domain transfers to the FineWeb domain.
    Section 4.1 adopts the value from Mosquera et al. without a sensitivity analysis on this corpus.
  • domain assumption The MAIN-RAG dynamic threshold with n=0.5 remains appropriate when Agent-3 is changed to emit citations and Agent-4 is added.
    Section 4.2.2 adopts n=0.5 from MAIN-RAG; no experiment measures how filtering stringency interacts with the revised pipeline.

how reviews work

0 comments
Cite this review

Pith. "Pith review of RAGentA: Multi-Agent Retrieval-Augmented Generation for Attributed Question Answering." pith.science (2026). https://pith.science/paper/WLB5KUVG

@misc{pith2026250616988,
  author       = {Pith},
  title        = {Pith review of: RAGentA: Multi-Agent Retrieval-Augmented Generation for Attributed Question Answering},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/WLB5KUVG}},
  note         = {Machine review of arXiv:2506.16988}
}
read the original abstract

We present RAGentA, a multi-agent retrieval-augmented generation (RAG) framework for attributed question answering (QA) with large language models (LLMs). With the goal of trustworthy answer generation, RAGentA focuses on optimizing answer correctness, defined by coverage and relevance to the question and faithfulness, which measures the extent to which answers are grounded in retrieved documents. RAGentA uses a multi-agent architecture that iteratively filters retrieved documents, generates attributed answers with in-line citations, and verifies completeness through dynamic refinement. Central to the framework is a hybrid retrieval strategy that combines sparse and dense methods, improving Recall@20 by 12.5% compared to the best single retrieval model, resulting in more correct and well-supported answers. Evaluated on a synthetic QA dataset derived from the FineWeb index, RAGentA outperforms standard RAG baselines, achieving gains of 1.09% in correctness and 10.72% in faithfulness. These results demonstrate the effectiveness of our multi-agent RAG architecture and hybrid retrieval strategy in advancing trustworthy QA with LLMs.

Figures

Figures reproduced from arXiv: 2506.16988 by the authors.

Figure 1
Figure 1. Architecture of the RAGentA framework: (1) A hybrid retriever selects top-20 documents. (2) Agent-1 generates an [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

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

  1. Claim2Source at CheckThat! 2026: Improving Multilingual Scientific Claim-Source Retrieval with Verification-based Re-Ranking

    cs.IR 2026-07 accept novelty 4.0 of 10

    A three-stage pipeline with bilingual claims, metadata-enhanced sources, and verification-based LLM re-ranking reaches 0.7628 average MRR@5 and ranks first on CheckThat! 2026 Task 1.

  2. SIGIR 2025 -- LiveRAG Challenge Report

    cs.CL 2025-07 conditional novelty 3.0 of 10

    In the SIGIR 2025 LiveRAG Challenge, all 25 active RAG teams beat the no-RAG baseline on LLM-judged correctness, and LLM scores correlated with human scores at r=0.88.

Reference graph

Works this paper leans on

30 extracted references · 9 canonical work pages · cited by 2 Pith papers

  1. [1]

    Moshe Berchansky, Daniel Fleischer, Moshe Wasserblat, and Peter Izsak. 2024. CoTAR: Chain-of-Thought Attribution Reasoning with Multi-level Granularity. In Findings of the Association for Computational Linguistics: EMNLP 2024 . ACL, Miami, Florida, USA, 236–246. https://doi.org/10.18653/v1/2024.findings-emnlp. 13

    doi:10.18653/v1/2024.findings-emnlp 2024

  2. [2]

    Chia-Yuan Chang, Zhimeng Jiang, Vineeth Rakesh, Menghai Pan, Chin- Chia Michael Yeh, Guanchu Wang, Mingzhi Hu, Zhichao Xu, Yan Zheng, Ma- hashweta Das, and Na Zou. 2024. MAIN-RAG: Multi-Agent Filtering Retrieval- Augmented Generation. arXiv:2501.00332 [cs.CL]

    arXiv 2024

  3. [3]

    Shahul Es, Jithin James, Luis Espinosa Anke, and Steven Schockaert. 2024. RAGAs: Automated Evaluation of Retrieval Augmented Generation. In Proceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics: System Demonstrations . ACL, St. Julians, Malta, 150–158. https: //aclanthology.org/2024.eacl-demo.16/

    work page 2024

  4. [4]

    Constanza Fierro, Reinald Kim Amplayo, Fantine Huot, Nicola De Cao, Joshua Maynez, Shashi Narayan, and Mirella Lapata. 2024. Learning to Plan and Generate Text with Citations. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) . ACL, Bangkok, Thailand, 11397–11417. https://doi.org/10.18653/v1...

    work page doi:10.18653/v1/2024.acl-long.615 2024

  5. [5]

    Simone Filice, Guy Horowitz, David Carmel, Zohar Karnin, Liane Lewin-Eytan, and Yoelle Maarek. 2025. Generating Diverse Q&A Benchmarks for RAG Evalua- tion with DataMorgana. arXiv:2501.12789 [cs.CL]

    arXiv 2025

  6. [6]

    Luyu Gao, Zhuyun Dai, Panupong Pasupat, Anthony Chen, Arun Tejasvi Cha- ganty, Yicheng Fan, Vincent Zhao, Ni Lao, Hongrae Lee, Da-Cheng Juan, and Kelvin Guu. 2023. RARR: Researching and Revising What Language Models Say, Using Language Models. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers...

    doi:10.18653/v1/2023.acl-long.910 2023

  7. [7]

    Tianyu Gao, Howard Yen, Jiatong Yu, and Danqi Chen. 2023. Enabling Large Language Models to Generate Text with Citations. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing . ACL, Singapore, 6465–6488. https://doi.org/10.18653/v1/2023.emnlp-main.398

    doi:10.18653/v1/2023.emnlp-main.398 2023

  8. [8]

    Chengyu Huang, Zeqiu Wu, Yushi Hu, and Wenya Wang. 2024. Training Language Models to Generate Text with Citations via Fine-grained Rewards. In Proceedings of the 62nd Annual Meeting of the Association for Computa- tional Linguistics (Volume 1: Long Papers) . ACL, Bangkok, Thailand, 2926–2949. https://doi.org/10.18653/v1/2024.acl-long.161

    doi:10.18653/v1/2024.acl-long.161 2024

Show all 30 references
  1. [9]

    Jie Huang and Kevin Chang. 2024. Citation: A Key to Building Responsible and Accountable Large Language Models. In Findings of the Association for Computational Linguistics: NAACL 2024 . ACL, Mexico City, Mexico, 464–473. https://doi.org/10.18653/v1/2024.findings-naacl.31

    2024 doi

  2. [10]

    Lei Huang, Weijiang Yu, Weitao Ma, Weihong Zhong, Zhangyin Feng, Haotian Wang, Qianglong Chen, Weihua Peng, Xiaocheng Feng, Bing Qin, and Ting Liu. 2025. A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questions. ACM Trans. Inf. S...

    2025 doi

  3. [11]

    Jisoo Jang and Wen-Syan Li. 2024. AU-RAG: Agent-based Universal Retrieval Augmented Generation. In Proceedings of the 2024 Annual International ACM SIGIR Conference on Research and Development in Information Retrieval in the Asia Pacific Region (Tokyo, Japan) (SIGIR-AP 2024). ...

    2024

  4. [12]

    Patrick Lewis, Ethan Perez, Aleksandra Piktus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Heinrich Küttler, Mike Lewis, Wen-tau Yih, Tim Rocktäschel, Sebastian Riedel, and Douwe Kiela. 2020. Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks. In Advances in ...

    2020

  5. [13]

    Jhon Stewar Rayo Mosquera, Carlos Raúl De La Rosa Peredo, and Mario Garrido Córdoba. 2025. A Hybrid Approach to Information Retrieval and Answer Genera- tion for Regulatory Texts. InProceedings of the 1st Regulatory NLP Workshop (Reg- NLP 2025). ACL, Abu Dhabi, UAE, 31–35. htt...

    2025

  6. [14]

    Nilay Patel, Shivashankar Subramanian, Siddhant Garg, Pratyay Banerjee, and Amita Misra. 2024. Towards Improved Multi-Source Attribution for Long-Form Answer Generation. In Proceedings of the 2024 Conference of the North Ameri- can Chapter of the Association for Computational ...

    2024 doi

  7. [15]

    Ronak Pradeep, Nandan Thakur, Shivani Upadhyay, Daniel Campos, Nick Craswell, and Jimmy Lin. 2025. The Great Nugget Recall: Automating Fact Extrac- tion and RAG Evaluation with Large Language Models. arXiv:2504.15068 [cs.IR]

    2025 arXiv

  8. [16]

    Jirui Qi, Gabriele Sarti, Raquel Fernández, and Arianna Bisazza. 2024. Model Internals-based Answer Attribution for Trustworthy Retrieval-Augmented Gen- eration. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing. ACL, Miami, Florida, USA...

    2024

  9. [17]

    Pritika Ramu, Koustava Goswami, Apoorv Saxena, and Balaji Vasan Srinivasan

  10. [18]

    S. E. Robertson and S. Walker. 1994. Some simple effective approximations to the 2-Poisson model for probabilistic weighted retrieval. In ACM SIGIR’94 (Dublin, Ireland). Springer-Verlag, Berlin, Heidelberg, 232–241. https://doi.org/10.1007/ 978-1-4471-2099-5_24

    1994

  11. [19]

    Robertson, Steve Walker, Susan Jones, Micheline Hancock-Beaulieu, and Mike Gatford

    Stephen E. Robertson, Steve Walker, Susan Jones, Micheline Hancock-Beaulieu, and Mike Gatford. 1994. Okapi at TREC-3. In TREC’94, Vol. 500-225. NIST, Gaithersburg, USA, 109–126. http://trec.nist.gov/pubs/trec3/papers/city.ps.gz

    1994

  12. [20]

    Yasser Saeid and Thomas Kopinski. 2024. AgentFusion: A Multi-Agent Ap- proach to Accurate Text Generation. In2024 International Conference on Electrical and Computer Engineering Researches (ICECER) . IEEE, Gaborone, Botswana, 1–8. https://doi.org/10.1109/ICECER62944.2024.10920460

    2024

  13. [21]

    Pfohl, Heather Cole-Lewis, Darlene Neal, Qazi Mamunur Rashid, Mike Schaekermann, Amy Wang, Dev Dash, Jonathan H

    Karan Singhal, Tao Tu, Juraj Gottweis, Rory Sayres, Ellery Wulczyn, Mohamed Amin, Le Hou, Kevin Clark, Stephen R. Pfohl, Heather Cole-Lewis, Darlene Neal, Qazi Mamunur Rashid, Mike Schaekermann, Amy Wang, Dev Dash, Jonathan H. Chen, Nigam H. Shah, Sami Lachgar, Philip Andrew M...

    2025

  14. [22]

    Ross Taylor, Marcin Kardas, Guillem Cucurull, Thomas Scialom, Anthony Hartshorn, Elvis Saravia, Andrew Poulton, Viktor Kerkez, and Robert Stojnic

  15. [23]

    Liang Wang, Nan Yang, Xiaolong Huang, Binxing Jiao, Linjun Yang, Daxin Jiang, Rangan Majumder, and Furu Wei. 2024. Text Embeddings by Weakly-Supervised Contrastive Pre-training. arXiv:2212.03533 [cs.CL]

    2024 arXiv

  16. [24]

    Chi, Quoc V

    Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Brian Ichter, Fei Xia, Ed H. Chi, Quoc V. Le, and Denny Zhou. 2022. Chain-of-thought prompting elicits reasoning in large language models. InProceedings of the 36th International Conference on Neural Information Processin...

    2022

  17. [25]

    Diji Yang, Jinmeng Rao, Kezhen Chen, Xiaoyuan Guo, Yawen Zhang, Jie Yang, and Yi Zhang. 2024. IM-RAG: Multi-Round Retrieval-Augmented Generation Through Learning Inner Monologues. In Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Info...

    2024

  18. [26]

    Hashimoto

    Tianyi Zhang, Faisal Ladhak, Esin Durmus, Percy Liang, Kathleen McKeown, and Tatsunori B. Hashimoto. 2024. Benchmarking Large Language Models for News Summarization. Transactions of the Association for Computational Linguistics 12 (2024), 39–57. https://doi.org/10.1162/tacl_a_00632

    2024 doi

  19. [27]

    Jun Zhao, Can Zu, Xu Hao, Yi Lu, Wei He, Yiwen Ding, Tao Gui, Qi Zhang, and Xuanjing Huang. 2024. LONGAGENT: Achieving Question Answering for 128k- Token-Long Documents through Multi-Agent Collaboration. InProceedings of the 2024 Conference on Empirical Methods in Natural Lang...

    2024 doi

  20. [28]

    Junda Zhu, Lingyong Yan, Haibo Shi, Dawei Yin, and Lei Sha. 2024. ATM: Adversarial Tuning Multi-agent System Makes a Robust Retrieval-Augmented Generator. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing. ACL, Miami, Florida, USA, 10902...

    2024

  21. [2022]

    arXiv:2211.09085 [cs.CL]

    Galactica: A Large Language Model for Science. arXiv:2211.09085 [cs.CL]

    arXiv

  22. [2024]

    In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing

    Enhancing Post-Hoc Attributions in Long Document Comprehension via Coarse Grained Answer Decomposition. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing . ACL, Miami, Florida, USA, 17790–17806. https://doi.org/10.18653/v1/2024.emnlp-main.985

    2024 doi

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.