Context-Guided Dynamic Retrieval for Improving Generation Quality in RAG Models

Binrong Zhu; Guiran Liu; Hanlu Zhang; Hongye Zheng; Jacky He; Xiaokai Wang

REVIEW 4 major objections 6 minor 4 cited by

Context-Guided Dynamic Retrieval for Improving Generation Quality in RAG Models

T0 review · 4 major / 6 minor · reviewed 2026-08-16 · deepseek-v4-flash

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

Pith's one-line read A state-aware retrieval vector, recomputed at each generation step, improves retrieval-augmented generation quality.

desk verdict The architecture is a plausible incremental dynamic RAG variant, but the NQ retrieval setup appears to leak gold answers, and the experiments never actually test dynamic retrieval against a real open-domain index. read the letter →

arxiv 2504.19436 v1 pith:3SB3JYDZ submitted 2025-04-28 cs.CL cs.LG

Jacky He , Guiran Liu , Binrong Zhu , Hanlu Zhang , Hongye Zheng , Xiaokai Wang This is my paper · ORCID

classification cs.CLcs.LG

keywords retrieval-augmentedgenerationdynamicretrievalstate-awarevectordifferentiabledocumentmatchingjointretrieval-generationtrainingopen-domainquestionansweringNaturalQuestionsknowledgescheduling

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

The paper argues that a retrieval-augmented generator should not decide once, from the initial question, which documents to use. Instead, it recomputes a state-aware retrieval vector at every generation step from the original query and the text produced so far, so the knowledge brought in can track the model's evolving information needs and avoid the context drift of a static document block. If that mechanism works as claimed, retrieval and generation become one jointly trained system: a differentiable matching path lets the generation objective shape what the retriever selects, and the result is higher BLEU and ROUGE-L scores on open-domain question answering. The paper reports those gains on Natural Questions across several large language models, with GPT-4o scoring highest, and it reports that retrieval vectors built with more context are more robust when queries are ambiguous.

What carries the argument

The load-bearing object is the state-aware retrieval vector $\mathbf{q}'_t$ and the differentiable softmax path that turns it into the dynamic context embedding $\mathbf{c}_t$. The vector is produced by an MLP over the concatenated original query and current decoder hidden state, which lets each generated token alter what will be retrieved next; the softmax scores $\alpha_i$ then fuse the top documents into $\mathbf{c}_t$, which is passed to the generator. Because the matching is differentiable, training can minimize $L_{\mathrm{total}} = L_{\mathrm{gen}} + \lambda L_{\mathrm{ret}}$ end-to-end, so the retrieval vectors evolve in the direction that lowers text-generation loss. This is the mechanism that is supposed to deliver the paper's measured BLEU and ROUGE-L improvements.

What would settle it

Run the same architecture on an open-domain corpus where the target passage must be found among thousands of unrelated documents, and compare retrieval recall and generation BLEU/ROUGE-L against a static retriever; if the dynamic mechanism does not improve either metric when the gold passage is not guaranteed to be present, the central claim would be falsified.

Watch

Extended reading notes

Core claim

The central claim is that dynamic retrieval representation, a context embedding $\mathbf{c}_t$ recomputed at every decoding step, improves RAG generation quality by preventing context drift. Concretely, the retrieval query is rebuilt as $\mathbf{q}'_t = \mathrm{MLP}([\mathbf{q}; \mathbf{h}_t])$, documents are scored by a softmax over scaled dot products between $\mathbf{q}'_t$ and pre-encoded document vectors $\mathbf{d}_i$, and the weighted context $\mathbf{c}_t = \sum_i \alpha_i \mathbf{d}_i$ is fed into the Transformer generator together with the hidden state. Since the entire matching path is differentiable, the retriever learns from the generation loss as well as from a contrastive retrieval loss, so the knowledge selected at each step is shaped by what the generator still needs. The paper's experiments on Natural Questions are offered as evidence that this dynamic loop outperforms static query retrieval and query-plus-context variants, and that the advantage grows as the retrieval vectors incorporate more contextual signal.

Load-bearing premise

The load-bearing assumption is that the Natural Questions passages used as the knowledge base form a realistic retrieval setting, with the gold answer present but not trivially given; if the index contains only question-associated passages, the retrieval module is not actually being challenged, and the reported gains may reflect the generator receiving the answer context rather than the dynamic retrieval mechanism.

Editorial extensions

If this is right

Static RAG pipelines can be changed so that retrieval is conditioned on the partial output, letting knowledge selection follow the information need as it emerges rather than fixing it from the first query.
End-to-end training of retriever and generator should let weaker generators compensate for missing parametric knowledge by learning when to pull external context.
On ambiguous or multi-document inputs, generation consistency should degrade less, because later tokens can steer the retriever toward more relevant passages.
Under the paper's across-model comparison, the dynamic loop's quality gain sits on top of base-model strength: stronger generators realize larger BLEU and ROUGE-L improvements.

Reading between the lines

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

The paper does not test multi-turn dialogue, but the same state-conditional mechanism suggests its largest gains would appear there, since the accumulated dialogue history is a much richer steering signal than a single question.
If the Natural Questions knowledge base is just the passages paired with each question, the retrieval module is choosing among passages that already contain the answer; replacing that index with a large open-domain corpus and measuring passage recall would show whether the gain is genuine retrieval improvement or the generator being handed the gold context.
The differentiable matching path could be extended to adaptive retrieval frequency, so the model decides at each step whether to retrieve zero, one, or many documents; that would make the reported latency and knowledge-scheduling claims directly testable.

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

Colleague,

The punchline: this paper's central claim isn't supported by its own experiments. The retrieval setup looks like it leaks the gold answer, and there's no static-retrieval baseline in the model comparison. The method itself—making the retrieval vector a function of the generation state and training retrieval and generation jointly via a differentiable attention path—is a reasonable, if incremental, way to frame dynamic RAG. The components (query-conditioned hidden-state MLP, soft attention over document vectors, contrastive retrieval loss) are all familiar from DR-RAG, DRAGIN, and REALM-style work, and the authors cite some of that work. The combination is coherent, but I can't call it new in any meaningful sense.

What the paper does well: it's clearly written, the architecture diagram matches the equations, and the authors do run experiments across several LLMs. The ablation in Figure 2 at least compares their proposed retrieval-vector construction against a static query encoding and a context-augmented variant, which is the right kind of control.

Where it falls apart: Section III.A says the questions in NQ are used as queries, the associated Wikipedia passages as the external knowledge base D, and the provided gold passage as the supervision signal. That implies each question's candidate set contains the very passage from which the answer was drawn. With the answer inside D, a retrieval module that picks the gold passage hands the generator the answer context, and the "dynamic" advantage is confounded with leakage. The stress-test note makes exactly this point, and after reading the paper I think it's decisive. There are also no error bars or significance testing anywhere, and BLEU/ROUGE-L are weak metrics for open-domain QA. The robustness study uses an unspecified ambiguity labeling, so those numbers are uninterpretable. The method section leaves out training details (batch size, learning rate, lambda schedule, number of retrieval documents) that would be needed to reproduce the work.

The reference list is padded with unrelated topics (dermatology, traffic scheduling), which doesn't help confidence.

Bottom line: this is a desk-reject for a serious venue. The evaluation design breaks the central claim. If the authors redo the experiments with a genuine open-domain index and include a static top-k baseline and proper metrics (e.g., exact match, answer F1), the architecture might earn a workshop-level contribution. As it stands, I wouldn't cite it or recommend anyone spend referee time on it.

Referee Report

4 major / 6 minor

Summary. The paper proposes a dynamic retrieval mechanism for retrieval-augmented generation (RAG). The method introduces a state-aware retrieval vector q'_t computed from the query and the generator's hidden state via an MLP, a differentiable document selection module based on scaled dot-product attention, and a joint loss combining a generation cross-entropy term with a contrastive retrieval loss. The authors evaluate on the Natural Questions dataset with several LLMs (GPT-3.5, GPT-4, GPT-4o, Qwenmax, DeepSeek) and report BLEU and ROUGE-L scores, an ablation over retrieval vector construction methods, and a robustness study under query ambiguity. The central claim is that dynamically adjusting the knowledge input during generation avoids context drift and significantly improves generation quality.

Significance. The motivation is plausible and the proposed mechanism is simple enough to be implemented in existing RAG pipelines. Conditioning retrieval on the generation state and making the matching path differentiable are both active and reasonable research directions. However, the manuscript provides no code, no data, no hyperparameter specifications, and its experiments are not designed to support the headline claim. The retrieval pool appears to contain the gold answer passage for each question, which confounds the results with answer leakage, and no static RAG baseline is included. If the evaluation were properly redone on a genuine open-domain index with standard baselines, the method might show value, but as it stands the empirical evidence is not convincing. The paper also has significant presentation issues, including garbled equations and missing figures.

major comments (4)

[III.A (Datasets)] The candidate set D is built from 'the associated Wikipedia passages' for each NQ question, and 'the provided gold passage is used as the supervision signal.' Because the gold answer passage is contained in D for every query, the retrieval module is never tested against a genuine open-domain index in which the answer may be absent. The reported generation improvements are therefore confounded by answer leakage: the generator is effectively handed the answer context, so the experiments cannot support the claim that dynamic retrieval improves generation quality. The authors should re-evaluate on a full-corpus retrieval setting, report retrieval metrics such as Recall@k, and include a condition in which the gold passage is excluded from D.
[III.B (Table 1 and Figure 2)] The experiments do not compare dynamic retrieval against a static RAG baseline. Table 1 varies only the generator LLM under the same dynamic mechanism. Figure 2's 'basic static query method' is not a conventional static top-k RAG baseline; it is a different retrieval-vector construction within the same differentiable pipeline. Without an identical-condition static RAG comparator, the claimed 'significant improvements' of the dynamic mechanism cannot be attributed to the proposed method. In addition, Figure 2 is only described textually, and no error bars, confidence intervals, or significance tests are provided for any reported BLEU/ROUGE-L values.
[III.B (Figure 3)] The robustness study is not reproducible. The paper does not specify how queries were classified into low, moderate, and high ambiguity, who performed the labeling, or how many queries were in each subset. The 'robustness' percentages (92.3%, 87.5%, 81.2%) are never defined, and the figure itself is absent from the manuscript. Without this information, the reported decline under ambiguity cannot be interpreted.
[II (Joint loss) and III.A (splits)] Several parameters and implementation details needed to reproduce the central result are missing. The hyperparameter λ in the joint loss L_total = L_gen + λ L_ret is never specified or analyzed, the size of the NQ split used in each experiment is not stated, and the retrieval loss construction (positive/negative document pairs) is not described. The conclusion asserts 'significant' improvements, but no statistical tests, variance measures, or exact test-set sizes are reported anywhere.

minor comments (6)

[II (Equations)] The mathematical notation in Section II is not typeset correctly (e.g., the equation for q'_t appears as ']);([' tt hqMLPq =' and the retrieval probability formula is garbled). This must be fixed before the method can be understood.
[I (Figures)] Figures 1 and 3 are referenced in the text but do not appear in the manuscript; only Figure 2 is described numerically. The missing figures make the architecture and the robustness results impossible to verify.
[I (Typos)] The manuscript contains several formatting errors: the header 'Abstract-This' runs together, 'Keywords-dynamic retrieval' lacks a space, and the affiliation block repeats 'San Francisco State University' twice. A full proofread is needed.
[II (References)] References [17]-[19] are cited as the basis for the hierarchical semantic encoding, differentiable matching, and dynamic scheduling, but the titles of those papers do not transparently support these specific techniques; the authors should cite the original methodological sources (e.g., RAG, DPR, REALM, Atlas) instead.
[III.B (Metrics)] The evaluation relies on BLEU and ROUGE-L for open-domain question answering without justifying these lexical-overlap metrics. Reporting exact-match or answer F1 alongside would make the results more meaningful.
[II (Naming)] The term 'multi-level perceptive retrieval vector construction' is an overstatement for a single-layer MLP over the concatenation of q and h_t; a more measured description would match the actual architecture.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation: the dynamic-retrieval mechanism is not defined in terms of the reported BLEU/ROUGE-L outcomes, and no load-bearing result is imported from the authors' prior work.

full rationale

The paper's derivation chain is a standard RAG architecture: the state-aware retrieval vector q'_t = MLP([q; h_t]) (Eq. 1), softmax retrieval weights (Eq. 2), context embedding c_t = sum alpha_i d_i, and Transformer generation (Eq. 3) are all constructive definitions, not fitted targets. The joint loss L_total = L_gen + lambda L_ret is an optimization objective, and the empirical claims are supported by benchmark comparisons (Table 1, Figures 2 and 3) on the external Natural Questions dataset. There is no equation that reduces the predicted quantities to the method's inputs by construction, and no parameter is fitted to BLEU/ROUGE-L and then renamed as a prediction. The citations to Cai et al., Yu et al., and Kai et al. are architectural inspiration, not load-bearing uniqueness theorems or self-citations by the present authors. The evaluation concern that NQ's associated Wikipedia passages contain the gold answer is a validity/leakage threat, not a circularity: it does not make the BLEU/ROUGE-L scores definitionally equal to the retrieval-vector construction. Under the stated criteria, the correct finding is no significant circularity.

Assumptions & free parameters 1 free parameters · 5 assumptions · 0 invented entities

The paper introduces no new physical or mathematical entities. Its central claim rests on standard mathematical operations and several unvalidated domain assumptions about the retrieval setup, the evaluation metrics, and the training objective.

free parameters (1)

lambda (retrieval loss weight)
Hyperparameter balancing L_gen and L_ret in the joint loss; no value or selection procedure is given in the paper.

assumptions (5)

standard math Softmax over scaled dot-product attention yields a valid retrieval distribution (Eq. 2).
The retrieval score alpha_i is a softmax of query-document dot products, a standard differentiable approximation.
ad hoc to paper The hidden state h_t of the generator captures the current knowledge gap and is a useful conditioning signal for retrieval.
The method updates the retrieval vector as MLP([q; h_t]) with no evidence or ablation isolating this assumption.
domain assumption Natural Questions 'associated Wikipedia passages' form a realistic retrieval corpus.
Section III-A states these passages are used as the knowledge base; if they are the gold-containing passages, retrieval is trivial.
domain assumption BLEU and ROUGE-L are valid measures of generation quality for this task.
All quality claims rest on these token-overlap metrics, which are not standard for open-domain QA and may not reflect answer correctness.
ad hoc to paper A contrastive retrieval loss with positive and negative document pairs is well-defined.
The paper mentions L_ret but never specifies how positives and negatives are sampled, making the training objective underspecified.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Context-Guided Dynamic Retrieval for Improving Generation Quality in RAG Models." pith.science (2026). https://pith.science/paper/3SB3JYDZ

@misc{pith2026250419436,
  author       = {Pith},
  title        = {Pith review of: Context-Guided Dynamic Retrieval for Improving Generation Quality in RAG Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/3SB3JYDZ}},
  note         = {Machine review of arXiv:2504.19436}
}

read the original abstract

This paper focuses on the dynamic optimization of the Retrieval-Augmented Generation (RAG) architecture. It proposes a state-aware dynamic knowledge retrieval mechanism to enhance semantic understanding and knowledge scheduling efficiency in large language models for open-domain question answering and complex generation tasks. The method introduces a multi-level perceptive retrieval vector construction strategy and a differentiable document matching path. These components enable end-to-end joint training and collaborative optimization of the retrieval and generation modules. This effectively addresses the limitations of static RAG structures in context adaptation and knowledge access. Experiments are conducted on the Natural Questions dataset. The proposed structure is thoroughly evaluated across different large models, including GPT-4, GPT-4o, and DeepSeek. Comparative and ablation experiments from multiple perspectives confirm the significant improvements in BLEU and ROUGE-L scores. The approach also demonstrates stronger robustness and generation consistency in tasks involving semantic ambiguity and multi-document fusion. These results highlight its broad application potential and practical value in building high-quality language generation systems.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 4 Pith papers

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

Time-Series Learning for Proactive Fault Prediction in Distributed Systems with Deep Neural Structures
cs.DC 2025-05 conditional novelty 3.0 of 10

A GRU plus attention plus feedforward classifier outperforms transformer baselines on Azure telemetry fault prediction in the reported metrics, but without code, error bars, or train/test details.
Collaborative Distillation Strategies for Parameter-Efficient Language Model Deployment
cs.CL 2025-07 reject novelty 2.0 of 10

The paper claims that combining weighted output fusion, entropy-driven teacher weighting, and feature alignment produces a better distilled language model.
Structured Memory Mechanisms for Stable Context Representation in Large Language Models
cs.CL 2025-05 reject novelty 2.0 of 10

A gated memory module with attention-based reading and forgetting is reported to improve NarrativeQA and dialogue consistency over GPT-2, BART, Longformer, and RETRO.
Deep Probabilistic Modeling of User Behavior for Anomaly Detection via Mixture Density Networks
cs.LG 2025-05 reject novelty 2.0 of 10

A mixture density network with negative log-likelihood scoring is claimed to outperform neural baselines on UNSW-NB15, but the experimental support is not auditable.

Reference graph

Works this paper leans on

23 extracted references · 4 canonical work pages · cited by 4 Pith papers

[1]

Dr-rag: Applying dynamic document relevance to retrieval-augmented generation for question-answering,

Z. Hei, et al., “Dr-rag: Applying dynamic document relevance to retrieval-augmented generation for question-answering,” arXiv preprint arXiv:2406.07348, 2024

arXiv 2024
[2]

DRAGIN: Dynamic Retrieval Augmented Generation based on the Information Needs of Large Language Models,

W. Su, et al., “DRAGIN: Dynamic Retrieval Augmented Generation based on the Information Needs of Large Language Models,” arXiv preprint arXiv:2403.10081, 2024

arXiv 2024
[3]

Retrieval Augmented Generation for Dynamic Graph Modeling

Y. Wu, Y. Fang and L. Liao, “Retrieval Augmented Generation for Dynamic Graph Modeling,” arXiv preprint arXiv:2408.14523, 2024

work page Pith review arXiv 2024
[4]

Context- Aware Adaptive Sampling for Intelligent Data Acquisition Systems Using DQN,

W. Huang, J. Zhan, Y. Sun, X. Han, T. An and N. Jiang, “Context- Aware Adaptive Sampling for Intelligent Data Acquisition Systems Using DQN,” arXiv preprint arXiv:2504.09344, 2025

arXiv 2025
[5]

A Self-Supervised Vision Transformer Approach for Dermatological Image Analysis,

F. Guo, X. Wu, L. Zhang, H. Liu and A. Kai, “A Self-Supervised Vision Transformer Approach for Dermatological Image Analysis,” Journal of Computer Science and Software Applications, vol. 5, no. 4, 2025

2025
[6]

A Reinforcement Learning Approach to Traffic Scheduling in Complex Data Center Topologies,

Y. Deng, “A Reinforcement Learning Approach to Traffic Scheduling in Complex Data Center Topologies,” Journal of Computer Technology and Software, vol. 4, no. 3, 2025

2025
[7]

Single-Device Human Activity Recognition Based on Spatiotemporal Feature Learning Networks,

J. Zhan, “Single-Device Human Activity Recognition Based on Spatiotemporal Feature Learning Networks,” Transactions on Computational and Scientific Methods, vol. 5, no. 3, 2025

2025
[8]

A Self-training Framework for Automated Medical Report Generation,

S. Wang, Z. Liu and B. Peng, "A Self-training Framework for Automated Medical Report Generation," Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pp. 16443-16449, December 2023

2023

Show all 23 references

[9]

Deep Learning for Cross-Domain Recommendation with Spatial-Channel Attention,

L. Zhu, “Deep Learning for Cross-Domain Recommendation with Spatial-Channel Attention,” Journal of Computer Science and Software Applications, vol. 5, no. 4, 2025

2025
[10]

Calibration Learning for Few-shot Novel Product Description,

Z. Liu, M. Wu, B. Peng, Y. Liu, Q. Peng and C. Zou, "Calibration Learning for Few-shot Novel Product Description," Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval, pp. 1864-1868, July 2023

2023
[11]

DynamicER: Resolving Emerging Mentions to Dynamic Entities for RAG,

J. Kim, D. Ko and G. Kim, “DynamicER: Resolving Emerging Mentions to Dynamic Entities for RAG,” arXiv preprint arXiv:2410.11494, 2024

2024 arXiv
[12]

The Synergistic Role of Deep Learning and Neural Architecture Search in Advancing Artificial Intelligence

X. Yan, J. Du, L. Wang, Y. Liang, J. Hu and B. Wang, "The Synergistic Role of Deep Learning and Neural Architecture Search in Advancing Artificial Intelligence", Proceedings of the 2024 International Conference on Electronics and Devices, Computational Science (ICEDCS), pp. 45...

2024
[13]

Enhancing retrieval and managing retrieval: A four- module synergy for improved quality and efficiency in rag systems,

Y. Shi, et al., “Enhancing retrieval and managing retrieval: A four- module synergy for improved quality and efficiency in rag systems,” arXiv preprint arXiv:2407.10670, 2024

2024 arXiv
[14]

Unsupervised Detection of Fraudulent Transactions in E-commerce Using Contrastive Learning,

X. Li, Y. Peng, X. Sun, Y. Duan, Z. Fang and T. Tang, “Unsupervised Detection of Fraudulent Transactions in E-commerce Using Contrastive Learning,” arXiv preprint arXiv:2503.18841, 2025

2025 arXiv
[15]

Social Network User Profiling for Anomaly Detection Based on Graph Neural Networks,

Y. Zhang, “Social Network User Profiling for Anomaly Detection Based on Graph Neural Networks,” arXiv preprint arXiv:2503.19380, 2025

2025 arXiv
[16]

A Graph Attention-Based Recommendation Framework for Sparse User-Item Interactions,

A. Liang, “A Graph Attention-Based Recommendation Framework for Sparse User-Item Interactions,” Journal of Computer Science and Software Applications, vol. 5, no. 4, 2025

2025
[17]

Investigating Hierarchical Term Relationships in Large Language Models,

G. Cai, J. Gong, J. Du, H. Liu and A. Kai, “Investigating Hierarchical Term Relationships in Large Language Models,” Journal of Computer Science and Software Applications, vol. 5, no. 4, 2025

2025
[18]

Improving Harmful Text Detection with Joint Retrieval and External Knowledge,

Z. Yu, S. Wang, N. Jiang, W. Huang, X. Han and J. Du, “Improving Harmful Text Detection with Joint Retrieval and External Knowledge,” arXiv preprint arXiv:2504.02310, 2025

2025 arXiv
[19]

Efficient Compression of Large Language Models with Distillation and Fine-Tuning,

A. Kai, L. Zhu and J. Gong, “Efficient Compression of Large Language Models with Distillation and Fine-Tuning,” Journal of Computer Science and Software Applications, vol. 3, no. 4, pp. 30–38, 2023

2023
[20]

Self- Supervised Graph Neural Networks for Enhanced Feature Extraction in Heterogeneous Information Networks

J. Wei, Y. Liu, X. Huang, X. Zhang, W. Liu and X. Yan, "Self- Supervised Graph Neural Networks for Enhanced Feature Extraction in Heterogeneous Information Networks", 2024 5th International Conference on Machine Learning and Computer Application (ICMLCA), pp. 272-276, 2024

2024
[21]

Comparison of ChatGPT–3.5, ChatGPT-4, and orthopaedic resident performance on orthopaedic assessment examinations,

P. A. Massey, C. Montgomery and A. S. Zhang, “Comparison of ChatGPT–3.5, ChatGPT-4, and orthopaedic resident performance on orthopaedic assessment examinations,” JAAOS - Journal of the American Academy of Orthopaedic Surgeons, vol. 31, no. 23, pp. 1173– 1179, 2023

2023
[22]

Qwen-IG: A Qwen-based Instruction Generation Model for LLM Fine-tuning,

L. Zhang, et al., “Qwen-IG: A Qwen-based Instruction Generation Model for LLM Fine-tuning,” Proceedings of the 2024 13th International Conference on Computing and Pattern Recognition, 2024

2024
[23]

Deepseek-v3 technical report,

A. Liu, et al., “Deepseek-v3 technical report,” arXiv preprint arXiv:2412.19437, 2024

2024 arXiv

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.

[1] [1]

Dr-rag: Applying dynamic document relevance to retrieval-augmented generation for question-answering,

Z. Hei, et al., “Dr-rag: Applying dynamic document relevance to retrieval-augmented generation for question-answering,” arXiv preprint arXiv:2406.07348, 2024

arXiv 2024

[2] [2]

DRAGIN: Dynamic Retrieval Augmented Generation based on the Information Needs of Large Language Models,

W. Su, et al., “DRAGIN: Dynamic Retrieval Augmented Generation based on the Information Needs of Large Language Models,” arXiv preprint arXiv:2403.10081, 2024

arXiv 2024

[3] [3]

Retrieval Augmented Generation for Dynamic Graph Modeling

Y. Wu, Y. Fang and L. Liao, “Retrieval Augmented Generation for Dynamic Graph Modeling,” arXiv preprint arXiv:2408.14523, 2024

work page Pith review arXiv 2024

[4] [4]

Context- Aware Adaptive Sampling for Intelligent Data Acquisition Systems Using DQN,

W. Huang, J. Zhan, Y. Sun, X. Han, T. An and N. Jiang, “Context- Aware Adaptive Sampling for Intelligent Data Acquisition Systems Using DQN,” arXiv preprint arXiv:2504.09344, 2025

arXiv 2025

[5] [5]

A Self-Supervised Vision Transformer Approach for Dermatological Image Analysis,

F. Guo, X. Wu, L. Zhang, H. Liu and A. Kai, “A Self-Supervised Vision Transformer Approach for Dermatological Image Analysis,” Journal of Computer Science and Software Applications, vol. 5, no. 4, 2025

2025

[6] [6]

A Reinforcement Learning Approach to Traffic Scheduling in Complex Data Center Topologies,

Y. Deng, “A Reinforcement Learning Approach to Traffic Scheduling in Complex Data Center Topologies,” Journal of Computer Technology and Software, vol. 4, no. 3, 2025

2025

[7] [7]

Single-Device Human Activity Recognition Based on Spatiotemporal Feature Learning Networks,

J. Zhan, “Single-Device Human Activity Recognition Based on Spatiotemporal Feature Learning Networks,” Transactions on Computational and Scientific Methods, vol. 5, no. 3, 2025

2025

[8] [8]

A Self-training Framework for Automated Medical Report Generation,

S. Wang, Z. Liu and B. Peng, "A Self-training Framework for Automated Medical Report Generation," Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pp. 16443-16449, December 2023

2023