GE-Chat: A Graph Enhanced RAG Framework for Evidential Response Generation of LLMs

Hua Wei; Jiaxing Zhang; Kuan-Ru Liou; Longchao Da; Parth Mitesh Shah

REVIEW 5 major objections 6 minor 26 references

GE-Chat: A Graph Enhanced RAG Framework for Evidential Response Generation of LLMs

T0 review · 5 major / 6 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 GE-Chat shows that a knowledge graph built from a user's uploaded document can lift an LLM's evidence from whole chunks to exact supporting sentences, and applying it consistently improves evidential responses across models.

desk verdict Plausible engineering contribution with a useful released dataset, but the central empirical claim is undermined by a length confound in the evaluation metric and the absence of numeric results. read the letter →

arxiv 2505.10143 v1 pith:C6Y3UPFD submitted 2025-05-15 cs.CL

Longchao Da , Parth Mitesh Shah , Kuan-Ru Liou , Jiaxing Zhang , Hua Wei This is my paper · ORCID

classification cs.CL

keywords LLMsevidentialansweringretrieval-augmentedgenerationknowledgegraphchain-of-thoughtnaturallanguageinferencesentence-levelevidencehallucination

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

GE-Chat is a retrieval-augmented generation framework that tries to show that an LLM's answers to questions about an uploaded document can be paired with exact evidence sentences from that document, rather than whole paragraphs or the model's own possibly unreliable source reflection. The paper argues that building a knowledge graph from the document, using chain-of-thought reasoning to identify relevant entities, searching an n-hop sub-graph to retrieve source chunks, and then selecting sentences by an entailment-based score consistently improves evidence quality across a range of open and closed LLMs. If this is right, users could quickly see which parts of an answer are grounded in the uploaded material and which are not, without having to trust the model's own claims about its sources. The paper also releases a 1000-case dataset with human-annotated evidence sentences so the claim can be checked.

What carries the argument

The load-bearing object is the derived knowledge graph $\mathcal{G}$, built by an LLM extracting entities from document chunks and probing their relations. Around it, three mechanisms carry the argument: a chain-of-thought template that produces answer sentences plus reasoning steps; a sub-graph search that matches each reasoning step to entities in $\mathcal{G}$ and expands to second-hop neighbours to retrieve source chunks; and an optimization step that scores each candidate sentence $s_n$ by $F(s_n) = \alpha\,\mathrm{prob}(s_n \vdash S') - \beta\, \mathrm{len}(s_n)$, with $\alpha=\beta=0.5$, where the entailment probability comes from an off-the-shelf NLI model applied to the sentence pair. The graph anchors generated content to the document's original entities, the n-hop search converts that anchor into a small set of source chunks, and the entailment objective selects short sentences that logically support the answer.

What would settle it

Take a document with a known answer and ground-truth evidence sentence, and run the pipeline twice: once with true entity names and once with entity names replaced by random tokens before relation probing and sub-graph search. If the final selected evidence sentence is identical and still correct, the knowledge graph is not doing the claimed work; if it changes or becomes wrong, the graph is load-bearing. A second check is to inspect a case where the selected sentence does not entail the answer by human judgment and see whether the NLI ranking placed that sentence first.

Watch

Extended reading notes

Core claim

The paper's central claim is that evidence for an LLM's conclusion can be identified at the sentence level by constraining the search to the original document: first parse the document into a knowledge graph, elicit the model's chain-of-thought, map those reasoning steps to graph entities, expand through two-hop relations to retrieve source chunks, and then choose the sentence in each chunk that best entails the answer while staying concise. On a new 1000-case dataset spanning ten domains, the authors report that wrapping existing LLMs in GE-Chat consistently improves their evidential-response scores compared with the models' direct source reflection. They treat this as evidence that the framework generalizes across models, including smaller ones with weak instruction-following, and that sentence-level evidence can be produced in real time from the document graph.

Load-bearing premise

The load-bearing premise is that the knowledge graph built by the LLM accurately reflects the document's entities and relations and that the entailment model reliably ranks candidate sentences; if either is noisy, the retrieved evidence can point to the wrong sentences.

Editorial extensions

If this is right

Users will see exact supporting sentences from the uploaded document beside each answer, instead of a whole highlighted chunk or the model's own guess at its sources.
Models that answer well but struggle to reflect on their own sources will still produce reliable evidence, because the pipeline does not depend on the model's source-reflection ability.
Every selected evidence sentence is anchored to entities and relations from the original document, so responses remain traceable to the uploaded material by construction.
The pipeline is fast enough for deployment: entity extraction is $O(n\cdot l)$, relation probing is $O(m^2)$, and the two-hop search is $O(2e)$.

Reading between the lines

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

This suggests the framework's evidence quality is partly bottlenecked by the LLM's graph-building ability; swapping in a deterministic entity and relation extractor in the same pipeline would test whether graph fidelity, not generation quality, drives the reported gains.
The k-hop search could be extended to dynamic expansion for multi-hop questions, so the framework may generalize from evidence for single answers to evidence supporting a full reasoning chain.
The entailment-based sentence optimizer is independent of the graph search, so it could be bolted onto other retrieval pipelines as a general post-processor that turns chunk-level hits into sentence-level citations.
Because the released dataset includes human-annotated evidence, it could support a follow-up calibration study that compares the cosine-and-conciseness score against human judgments of which evidence actually persuades a reader.

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.

Desk Editor's Note

read the letter

Short version: GE-Chat is a sensible pipeline—GraphRAG construction, CoT elicitation, sub-graph search, then NLI-based sentence selection—for giving sentence-level evidence for LLM answers. The released 1,000-case dataset with human-annotated evidence is a real asset for anyone working on evidential QA. The complexity analysis is straightforward and the writing is clear about what the components do. That part is solid.

The soft spot is the evaluation, and it is load-bearing. Equation (4) defines Evidence score as cos(E_i, E_gt_i) * min(1, L_gt_i / L_i). That penalty rewards shorter outputs and never punishes overly short ones. Meanwhile the method's own optimization objective in Eq. (2) explicitly minimizes sentence length with a fixed 0.5 weight, and the paper admits the direct-reflection baselines “generate too many words.” So the comparison is structurally tilted: the proposed method is tuned to be concise, the baselines are not, and conciseness is rewarded in the metric. The “consistent improvement” in Fig. 4 could easily be mostly a length artifact rather than better evidence identification. On top of that, the text of the paper gives no numeric scores, no per-model breakdown, no error bars, and no significance tests. I can’t tell from the submission how large the gains are or whether they are reliable. The decision to exclude GPT-4o from the comparison, while explained, also means the strongest baseline is missing.

The stress-test note you forwarded hits the right spot. I don't think the KG noise concern is the main issue—that is a standard engineering risk. The metric confound is the thing that needs to be addressed before the central claim can be accepted.

The citation pattern is mostly fine. The self-citations to EvidenceChat are appropriate because that is the direct predecessor. The reliance on LightRAG and the RAGAS conciseness metric is also legitimate.

Who is this for? People building RAG systems that need to show users where answers came from, and anyone who wants a sentence-level evidence dataset. As a peer reviewer, I would engage with it—the dataset and framework are worth a serious look—but I would insist on seeing numeric results, a length-controlled comparison (e.g., truncating baseline outputs to the same length as groundtruth), and ideally an ablation that removes the length penalty from the metric to separate relevance from conciseness. Right now the empirical claim is not verified, but the artifact and the problem formulation justify referee time.

Referee Report

5 major / 6 minor

Summary. The paper proposes GE-Chat, a retrieval-augmented generation framework that constructs a knowledge graph from an uploaded document, elicits chain-of-thought reasoning steps from an LLM, searches an n-hop sub-graph by entity matching to retrieve source chunks, and then applies an NLI-based, length-penalized objective to select concise evidence sentences. The authors release a 1000-case dataset covering ten domains and report, in a single figure, that applying GE-Chat to several LLMs improves an evidence-quality score relative to the models' direct source reflection.

Significance. If the claimed improvement is real, GE-Chat offers a practical and model-agnostic way to produce sentence-level evidence citations, which would be useful for trustworthiness assessment. The paper also contributes a public dataset and a demo, and its pipeline combines established components (GraphRAG, CoT, sub-graph search, NLI) in a coherent way. However, the current manuscript does not provide enough quantitative evidence to support the central claim: the experimental section reports only a figure with no numeric values, the evaluation metric conflates relevance with conciseness, and no comparison with prior evidence-highlighting baselines is made. The significance is therefore conditional on a substantially strengthened evaluation.

major comments (5)

[Sec. 3.2, Eq. (4) and Sec. 2.4, Eq. (2)] The evaluation metric in Eq. (4), Evidencescore = (1/N) sum [cos(E_i, E_gt_i) * min(1, L_gt_i / L_i)], rewards shorter outputs by the min(1, L_gt_i / L_i) factor and never penalizes outputs shorter than the ground truth, while the proposed method's selection objective in Eq. (2), F(s_n) = 0.5 * prob(s_n entails S') - 0.5 * len(s_n), explicitly optimizes for conciseness. Since the manuscript itself reports that the direct LLM baselines 'generate too many words' (Sec. 3.2), the observed 'consistent improvement' in Fig. 4 may be driven primarily by a length effect rather than by more accurate identification of supporting evidence. To address this confound, the authors should report cosine similarity and length (or conciseness) as separate metrics, add a length-controlled baseline, and include an ablation that removes the length term from F.
[Sec. 3.2, Eq. (4)] The metric computes cosine similarity over the entire generated evidence block E_i against the ground-truth block E_gt_i, which does not directly measure the sentence-level fine-grained evidence identification claimed in Sec. 1 and Sec. 2.4. A block-level cosine score can be high even if the selected sentences are not the exact ground-truth sentences. Please supplement the evaluation with sentence-level or token-level alignment metrics (for example, token-level F1, ROUGE, or overlap of selected sentence indices with the human-annotated evidence sentences).
[Sec. 3.2] All quantitative results are reported only in Fig. 4, with no per-model numeric values, no error bars, no standard deviations, and no significance tests. The central claim that GE-Chat 'consistently improves the performance of each model's evidential-based responses' cannot be verified from a single figure. Please provide a table with per-model Evidence scores for both the baseline and GE-Chat, along with the number of runs and a paired statistical test (e.g., Wilcoxon signed-rank test over the 1000 cases).
[Sec. 3.1] The ground-truth answers and evidence sentences are constructed with GPT-4o plus 'human-correction,' but no details are given about the annotation protocol, the number of annotators, their expertise, or inter-annotator agreement. Since all reported evidence scores measure agreement with this ground truth, its reliability is load-bearing. Please describe the annotation process and report agreement statistics (e.g., Cohen's kappa or a second-annotation overlap measure).
[Sec. 3.2] The only baseline is the direct LLM source reflection of the evidence. The manuscript does not compare GE-Chat with prior chunk-level evidence-retrieval or evidence-highlighting methods, such as the systems cited in refs. [7] and [13], or with simpler post-hoc sentence-selection baselines. Without such comparisons, it is unclear whether the reported improvement comes from the specific graph-based and CoT components or from the final NLI sentence selector. At minimum, please add a strong non-graph baseline (e.g., chunk retrieval followed by the same NLI-based sentence selection) to isolate the contribution of the knowledge-graph and CoT stages.

minor comments (6)

[Abstract and footnote 1] The sentence 'The datasets are released at 1' refers to a bare placeholder number; the actual footnote contains a Google Drive URL. Please provide a stable, citable dataset identifier (e.g., Zenodo or a DOI) and replace the placeholder.
[Sec. 2.4, Eq. (2)] The notation 'prob(s_n|− S′)' contains an unexplained minus sign before S'; the prose defines the intended quantity as the entailment probability prob(s_n entails S'). Please fix the notation for clarity.
[Sec. 3.2] The claim that 'the direct evidence retrieval ability of GPT4o is the best' is made without quantitative support in the text. If GPT-4o is used as a reference upper bound, please include its score in the figure or a table so the reader can calibrate the other results.
[Sec. 4 vs. Sec. 3.2] Sec. 4 states that the evaluation covers 'ten diverse LLMs,' but Sec. 3.2 says GE-Chat was applied to models except GPT-4o; please clarify the exact set of models tested and report their names.
[References] The ACM reference-format line lists the year as 2018 for a WWW '25 paper; please correct the placeholder year. Also, reference [3] (EvidenceChat) appears to be an unpublished manuscript; if it is the authors' own prior work, please provide a stable link or a published version.
[Fig. 4] Figure 4 is the only quantitative result but the axes, units, and the set of models are not described in the caption or text. Please make the figure self-contained and accessible.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: GE-Chat's evidence-selection objective and the evaluation metric are not definitionally locked, and the cited prior work is not load-bearing.

full rationale

The paper's derivation chain is self-contained with respect to the loaded evaluation. The GE-Chat evidence-selection objective in Eq. (2) is F(s_n) = alpha * prob(s_n entails S') - beta * len(s_n), with fixed alpha = beta = 0.5 and an external DeBERTa-large NLI model. The evaluation metric in Eq. (4) is Evidencescore = (1/N) * sum [ cos(E_i, E_gt_i) * min(1, L_gt_i / L_i) ]. These are related only in that both reward conciseness, but they are not the same function: the evaluation metric uses cosine similarity over the full generated evidence block against groundtruth evidence, while the method's objective uses sentence-level entailment probability. The length penalty in the metric could advantage the method because GE-Chat explicitly minimizes length, but this is a metric-validity concern, not a circular reduction; the claimed improvement still depends on the cosine-similarity term and on the retrieval pipeline. Self-citations, including [3] EvidenceChat and [1] uncertainty quantification, are used descriptively to characterize prior work and do not supply the central argument or forbid alternatives. The CoT prompting, sub-graph search, and NLI-based sentence selection all rely on external models or standard techniques, not on results derived from the evaluation metric. Therefore no step in the derivation reduces by construction to its own inputs, and no fitted parameter is renamed as a prediction. Score 0 is appropriate.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The central claim rests on two empirical assumptions: the LLM-built knowledge graph and CoT entity matches are accurate enough to locate the right chunks, and the NLI entailment probability correctly ranks supporting sentences. These are not derived in the paper and are tested only indirectly through the final Evidence score. The evaluation metric and human-annotated groundtruth are also assumed valid.

free parameters (3)

alpha = 0.5
Weight of entailment probability in objective F; chosen by hand in Sec 2.4, not tuned or justified.
beta = 0.5
Penalty for sentence length in objective F; chosen by hand in Sec 2.4.
k = 2
Number of hops in sub-graph searching; fixed implementation choice in Sec 2.3.

assumptions (5)

domain assumption The LLM can extract accurate entities and relations from document chunks to construct the knowledge graph.
Invoked in Sec 2.1; if entity extraction or relation probing is noisy, the graph does not support reliable retrieval.
domain assumption CoT reasoning steps contain entities that can be matched to knowledge graph entities for localization.
Sec 2.3 anchors generated content to source chunks via entity matching; this matching is assumed reliable.
domain assumption NLI entailment probability from DeBERTa-large correctly measures whether a sentence supports the answer sentence.
Sec 2.4 uses prob(s_n entails S') as the meaningfulness term in the objective; its validity is assumed without validation.
domain assumption The evaluation metric (cosine similarity and length ratio) reflects true evidence quality.
Sec 3.1 defines Evidencescore without validating it against human judgment or ablating its components.
domain assumption Groundtruth evidence annotations are reliable.
Sec 3.1 describes human-corrected GPT-4o generated references but reports no inter-annotator agreement.

how reviews work

0 comments

Cite this review

Pith. "Pith review of GE-Chat: A Graph Enhanced RAG Framework for Evidential Response Generation of LLMs." pith.science (2026). https://pith.science/paper/C6Y3UPFD

@misc{pith2026250510143,
  author       = {Pith},
  title        = {Pith review of: GE-Chat: A Graph Enhanced RAG Framework for Evidential Response Generation of LLMs},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/C6Y3UPFD}},
  note         = {Machine review of arXiv:2505.10143}
}

read the original abstract

Large Language Models are now key assistants in human decision-making processes. However, a common note always seems to follow: "LLMs can make mistakes. Be careful with important info." This points to the reality that not all outputs from LLMs are dependable, and users must evaluate them manually. The challenge deepens as hallucinated responses, often presented with seemingly plausible explanations, create complications and raise trust issues among users. To tackle such issue, this paper proposes GE-Chat, a knowledge Graph enhanced retrieval-augmented generation framework to provide Evidence-based response generation. Specifically, when the user uploads a material document, a knowledge graph will be created, which helps construct a retrieval-augmented agent, enhancing the agent's responses with additional knowledge beyond its training corpus. Then we leverage Chain-of-Thought (CoT) logic generation, n-hop sub-graph searching, and entailment-based sentence generation to realize accurate evidence retrieval. We demonstrate that our method improves the existing models' performance in terms of identifying the exact evidence in a free-form context, providing a reliable way to examine the resources of LLM's conclusion and help with the judgment of the trustworthiness.

Figures

Figures reproduced from arXiv: 2505.10143 by the authors.

**Figure 2.** A Simple Abstraction for Sub-graph Searching. [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗

**Figure 3.** The demonstration of the deployed GE-Chat framework.The user can upload PDF or relevant files, and the highlighted evidence comes along with the answers the LLMs made. For more examples please check the live demo video. 3 Experiment 3.1 Experiment Setup Dataset construction To address the scarcity of evidence sources in prior research, we created a dataset with 1000 cases to evaluate evidence generation quality acro… view at source ↗

Figures from the paper (1 more)

**Figure 4.** Figure 4: The comparison between the original LLMs evi [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

26 extracted references · 10 canonical work pages

[7]

Rujun Han, Yuhao Zhang, Peng Qi, Yumo Xu, Jenyuan Wang, Lan Liu, William Yang Wang, Bonan Min, and Vittorio Castelli. 2024. RAG-QA Arena: Evaluating Domain Robustness for Long-form Retrieval Augmented Question Answering. arXiv:2407.13998 [cs.CL] https://arxiv.org/abs/2407.13998

arXiv 2024
[13]

Demiao Lin. 2024. Revolutionizing retrieval-augmented generation with en- hanced PDF structure recognition. arXiv preprint arXiv:2401.12599 (2024)

arXiv 2024
[1]

Longchao Da, Tiejin Chen, Lu Cheng, and Hua Wei. 2024. Llm uncertainty quantification through directional entailment graph and claim level response augmentation. arXiv preprint arXiv:2407.00994 (2024)

arXiv 2024
[2]

Longchao Da, Kuanru Liou, Tiejin Chen, Xuesong Zhou, Xiangyong Luo, Yezhou Yang, and Hua Wei. 2024. Open-ti: Open traffic intelligence with augmented language model. International Journal of Machine Learning and Cybernetics (2024), 1–26

work page 2024
[3]

Longchao Da, Parth Mitesh Shah, Ananya Singh, and Hua Wei. 2024. Evi- denceChat: A RAG Enhanced LLM Framework for Trustworthy and Evidential Response Generation. (2024)

work page 2024
[4]

Hanxing Ding, Liang Pang, Zihao Wei, Huawei Shen, and Xueqi Cheng. 2024. Retrieve only when it needs: Adaptive retrieval augmentation for hallucination mitigation in large language models. arXiv preprint arXiv:2402.10612 (2024)

arXiv 2024
[5]

Asbjørn Følstad and Marita Skjuve. 2019. Chatbots for customer service: user experience and motivation. In Proceedings of the 1st international conference on conversational user interfaces. 1–9

work page 2019
[6]

Zirui Guo, Lianghao Xia, Yanhua Yu, Tu Ao, and Chao Huang. 2024. LightRAG: Simple and Fast Retrieval-Augmented Generation.arXiv preprint arXiv:2410.05779 (2024)

arXiv 2024

Show all 26 references

[8]

Wenbo Hu, Yifan Xu, Yi Li, Weiyue Li, Zeyuan Chen, and Zhuowen Tu. 2024. Bliva: A simple multimodal llm for better handling of text-rich visual questions. In Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 38. 2256–2264

2024
[9]

Ziwei Ji, Nayeon Lee, Rita Frieske, Tiezheng Yu, Dan Su, Yan Xu, Etsuko Ishii, Ye Jin Bang, Andrea Madotto, and Pascale Fung. 2023. Survey of hallucination in natural language generation. Comput. Surveys 55, 12 (2023), 1–38

2023
[10]

Ziwei Ji, Tiezheng Yu, Yan Xu, Nayeon Lee, Etsuko Ishii, and Pascale Fung
[11]

Jonathan Larson and Steven Truitt. 2024. GraphRAG: Unlocking LLM discovery on narrative private data

2024
[12]

Leonie aka helloiamleonie. 2024. evaluatingrag1. https://blog.langchain.dev/ evaluating-rag-pipelines-with-ragas-langsmith/

2024
[14]

Zhen Lin, Shubhendu Trivedi, and Jimeng Sun. 2023. Generating with confidence: Uncertainty quantification for black-box large language models. arXiv preprint arXiv:2305.19187 (2023)

2023 arXiv
[15]

Chandran Nandkumar and Luka Peternel. 2024. Enhancing Supermarket Robot Interaction: A Multi-Level LLM Conversational Interface for Handling Diverse Customer Intents. arXiv preprint arXiv:2406.11047 (2024)

2024
[16]

Ragas authors. 2024. conciseness. https://docs.ragas.io/en/stable/concepts/ metrics/summarization_score.html

2024
[17]

Jon Saad-Falcon, Joe Barrow, Alexa Siu, Ani Nenkova, Ryan A Rossi, and Franck Dernoncourt. 2023. PDFTriage: question answering over long, structured docu- ments. arXiv preprint arXiv:2309.08872 (2023)

2023 arXiv
[18]

Kurt Shuster, Spencer Poff, Moya Chen, Douwe Kiela, and Jason Weston. 2021. Retrieval augmentation reduces hallucination in conversation. arXiv preprint arXiv:2104.07567 (2021)

2021 arXiv
[19]

Lei Wang, Jiabang He, Shenshen Li, Ning Liu, and Ee-Peng Lim. 2024. Mitigating fine-grained hallucination by fine-tuning large vision-language models with caption rewrites. In International Conference on Multimedia Modeling . Springer, 32–45

2024
[20]

Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Fei Xia, Ed Chi, Quoc V Le, Denny Zhou, et al. 2022. Chain-of-thought prompting elicits reasoning in large language models. Advances in neural information processing systems 35 (2022), 24824–24837

2022
[21]

Rongxuan Wei, Kangkang Li, and Jiaming Lan. 2024. Improving Collaborative Learning Performance Based on LLM Virtual Assistant. In 2024 13th International Conference on Educational and Information Technology (ICEIT) . IEEE, 1–6

2024
[22]

Lifan Yuan, Ganqu Cui, Hanbin Wang, Ning Ding, Xingyao Wang, Jia Deng, Boji Shan, Huimin Chen, Ruobing Xie, Yankai Lin, et al. 2024. Advancing llm reasoning generalists with preference trees. arXiv preprint arXiv:2404.02078 (2024)

2024 arXiv
[23]

Jiaxing Zhang, Zhuomin Chen, Longchao Da, Dongsheng Luo, Hua Wei, et al
[24]

Zhuosheng Zhang, Aston Zhang, Mu Li, and Alex Smola. 2022. Automatic chain of thought prompting in large language models. arXiv preprint arXiv:2210.03493 (2022)

2022 arXiv
[2023]

In Findings of the Association for Computational Linguistics: EMNLP 2023

Towards mitigating LLM hallucination via self reflection. In Findings of the Association for Computational Linguistics: EMNLP 2023 . 1827–1843

2023
[2024]

Advances in Neural Information Processing Systems 37 (2024), 79282–79306

Regexplainer: Generating explanations for graph neural networks in regression tasks. Advances in Neural Information Processing Systems 37 (2024), 79282–79306

2024

Pith tools

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

[1] [7]

Rujun Han, Yuhao Zhang, Peng Qi, Yumo Xu, Jenyuan Wang, Lan Liu, William Yang Wang, Bonan Min, and Vittorio Castelli. 2024. RAG-QA Arena: Evaluating Domain Robustness for Long-form Retrieval Augmented Question Answering. arXiv:2407.13998 [cs.CL] https://arxiv.org/abs/2407.13998

arXiv 2024

[2] [13]

Demiao Lin. 2024. Revolutionizing retrieval-augmented generation with en- hanced PDF structure recognition. arXiv preprint arXiv:2401.12599 (2024)

arXiv 2024

[3] [1]

Longchao Da, Tiejin Chen, Lu Cheng, and Hua Wei. 2024. Llm uncertainty quantification through directional entailment graph and claim level response augmentation. arXiv preprint arXiv:2407.00994 (2024)

arXiv 2024

[4] [2]

Longchao Da, Kuanru Liou, Tiejin Chen, Xuesong Zhou, Xiangyong Luo, Yezhou Yang, and Hua Wei. 2024. Open-ti: Open traffic intelligence with augmented language model. International Journal of Machine Learning and Cybernetics (2024), 1–26

work page 2024

[5] [3]

Longchao Da, Parth Mitesh Shah, Ananya Singh, and Hua Wei. 2024. Evi- denceChat: A RAG Enhanced LLM Framework for Trustworthy and Evidential Response Generation. (2024)

work page 2024

[6] [4]

Hanxing Ding, Liang Pang, Zihao Wei, Huawei Shen, and Xueqi Cheng. 2024. Retrieve only when it needs: Adaptive retrieval augmentation for hallucination mitigation in large language models. arXiv preprint arXiv:2402.10612 (2024)

arXiv 2024

[7] [5]

Asbjørn Følstad and Marita Skjuve. 2019. Chatbots for customer service: user experience and motivation. In Proceedings of the 1st international conference on conversational user interfaces. 1–9

work page 2019

[8] [6]

Zirui Guo, Lianghao Xia, Yanhua Yu, Tu Ao, and Chao Huang. 2024. LightRAG: Simple and Fast Retrieval-Augmented Generation.arXiv preprint arXiv:2410.05779 (2024)

arXiv 2024