Experiments with Large Language Models on Retrieval-Augmented Generation for Closed-Source Simulation Software

Andreas Baumann; Peter Eberhard

arxiv: 2502.03916 · v2 · submitted 2025-02-06 · 💻 cs.CL · cs.AI

Experiments with Large Language Models on Retrieval-Augmented Generation for Closed-Source Simulation Software

Andreas Baumann , Peter Eberhard This is my paper

Pith reviewed 2026-05-23 04:26 UTC · model grok-4.3

classification 💻 cs.CL cs.AI

keywords retrieval-augmented generationlarge language modelsclosed-source softwaresimulation softwarelocal LLMsinformation retrievalPasimodo

0 comments

The pith

Tailoring retrieved information to specific prompts significantly improves LLM performance on closed-source simulation tasks.

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

This paper tests retrieval-augmented generation with local large language models to support tasks in the closed-source mesh-free simulation software Pasimodo. Tasks include smart autocomplete, question answering, model summarization, component explanation, and creation of simulation components or input models. The systems deliver impressive results on many prompts but frequently fail when the retrieved information proves insufficient for the query. Making the supplied information dependent on the particular prompt yields a clear improvement in response quality while keeping all processing local to protect intellectual property.

Core claim

RAG systems using local LLMs for Pasimodo produce strong results across a range of tasks but often fail due to insufficient information; tailoring the information provided to the LLMs dependent on the prompts proves to be a significant improvement.

What carries the argument

Retrieval-augmented generation with prompt-dependent selection of information for local LLMs.

If this is right

Local LLMs can handle tasks from autocomplete to full model creation without sending data outside the institution.
Failures stem mainly from gaps in the retrieved context rather than inherent model limits.
Prompt-specific tailoring of context offers a practical route to higher response quality.
Smaller language models become viable for these tasks when retrieval is adjusted to the prompt.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The same tailoring approach could be applied to other proprietary engineering tools that rely on internal documentation.
Further gains might come from analyzing the prompt first to decide which retrieval method to use.
Integration into simulation workflows could reduce the time engineers spend searching documentation.

Load-bearing premise

The tasks, prompts, and documentation structure of Pasimodo are representative enough of other closed-source simulation frameworks for the observed failure modes and improvements to generalize.

What would settle it

Testing the same prompt-dependent RAG approach on a different closed-source simulation framework with its own documentation and checking whether the quality gains persist.

read the original abstract

Large Language Models (LLMs) are tools that have become indispensable in development and programming. However, they suffer from hallucinations, especially when dealing with unknown knowledge. This is particularly the case when LLMs are to be used to support closed-source software applications. Retrieval-Augmented Generation (RAG) offers an approach to use additional knowledge alongside the pre-trained knowledge of the LLM to respond to user prompts. Possible tasks range from a smart-autocomplete, text extraction for question answering, model summarization, component explaining, compositional reasoning, to creation of simulation components and complete input models. This work tests existing RAG systems for closed-source simulation frameworks, in our case the mesh-free simulation software Pasimodo. Since data protection and intellectual property rights are particularly important for problems solved with closed-source software, the tests focus on execution using local LLMs. In order to enable smaller institutions to use the systems, smaller language models will be tested first. The systems show impressive results, but often fail due to insufficient information. Different approaches for improving response quality are tested. In particular, tailoring the information provided to the LLMs dependent to the prompts proves to be a significant improvement. This demonstrates the great potential and the further work needed to improve information retrieval for closed-source simulation models.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper applies standard RAG to Pasimodo with local models and claims prompt-dependent retrieval helps, but reports only qualitative impressions.

read the letter

The core observation is that making retrieval depend on the prompt improves results over plain RAG when answering questions about Pasimodo components or building input models. The authors test this on local LLMs to keep proprietary documentation inside the institution, which matches a real constraint for smaller groups using closed-source engineering software. They cover a range of tasks from simple explanation to full model generation and note that standard retrieval often leaves the model short on needed details. That practical framing is the main thing the work contributes. The experiments stay within one mesh-free package and its documentation, so the tailoring trick is shown only in that setting. No accuracy numbers, no error breakdowns, and no description of corpus size or prompt count appear in the write-up, which leaves the size of the reported improvement unclear. The same limitation applies to generalization: nothing in the scope indicates the failure modes or the fix would hold for other closed-source simulators with different data layouts or docs. The paper is aimed at practitioners who want to bolt LLMs onto proprietary simulation tools without sending data outside. It is honest about the current limits and the need for better retrieval, but the evidence is too thin for strong claims. A serious editor should send it to review only after the authors add quantitative metrics and at least one cross-framework check; without those it reads more like an internal tech note than a finished result.

Referee Report

3 major / 2 minor

Summary. The manuscript reports experiments applying local LLMs with Retrieval-Augmented Generation (RAG) to tasks such as component explanation, compositional reasoning, and creation of simulation components and input models for the closed-source mesh-free simulation software Pasimodo. It observes that standard RAG frequently fails due to insufficient retrieved information and claims that tailoring the retrieved context to the specific user prompt yields a significant improvement in response quality, while emphasizing the use of local models to address data-protection concerns.

Significance. If the tailoring approach can be shown to produce measurable gains that hold across multiple frameworks, the work could offer a practical route for LLM-assisted development of proprietary simulation software. The focus on local execution and smaller models is a constructive choice for IP-sensitive settings. At present the evidence consists of qualitative impressions only, so the magnitude and reliability of the reported improvement remain unquantified.

major comments (3)

[Abstract / Results] Abstract and Results section: the statements that the systems 'show impressive results' and that prompt-dependent tailoring 'proves to be a significant improvement' rest entirely on unquantified qualitative impressions; no accuracy, success rate, error-rate, or inter-rater metrics are reported for any of the tested RAG variants or tasks.
[Experimental Setup] Experimental Setup (implied in Abstract and Scope): the manuscript provides no information on retrieval-corpus size, the exact test prompts employed, the concrete LLM versions tested, or the rubric used to judge response quality, rendering the central observations non-reproducible and preventing assessment of whether the observed failures are due to retrieval or to other factors.
[Conclusion / Scope] Conclusion and Scope: the paper frames its findings as relevant to 'closed-source simulation frameworks' in general, yet all experiments are performed on a single mesh-free code (Pasimodo) without any argument or cross-framework test establishing that its documentation style or data layout is representative; this makes the claimed generality of the tailoring benefit an untested extrapolation.

minor comments (2)

[Abstract] Abstract: 'dependent to the prompts' is grammatically incorrect and should read 'dependent on the prompts'.
[Results] The manuscript would benefit from an explicit list of the tasks, the number of prompts per task, and a table summarizing the qualitative outcomes for each RAG configuration.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive comments, which correctly identify the reliance on qualitative assessment and the need for greater detail and caution in scope. We address each major comment below and will revise the manuscript to strengthen these aspects.

read point-by-point responses

Referee: [Abstract / Results] Abstract and Results section: the statements that the systems 'show impressive results' and that prompt-dependent tailoring 'proves to be a significant improvement' rest entirely on unquantified qualitative impressions; no accuracy, success rate, error-rate, or inter-rater metrics are reported for any of the tested RAG variants or tasks.

Authors: We agree that the current claims rest on qualitative impressions. In the revision we will add quantitative evaluation: success rates for each task and RAG variant, obtained via independent rating by two authors using a predefined rubric, together with inter-rater agreement statistics. revision: yes
Referee: [Experimental Setup] Experimental Setup (implied in Abstract and Scope): the manuscript provides no information on retrieval-corpus size, the exact test prompts employed, the concrete LLM versions tested, or the rubric used to judge response quality, rendering the central observations non-reproducible and preventing assessment of whether the observed failures are due to retrieval or to other factors.

Authors: We will add a dedicated Experimental Setup subsection that reports the retrieval-corpus size, the full list of test prompts, the precise LLM versions and inference parameters, and the complete evaluation rubric (accuracy, completeness, relevance, and hallucination criteria). revision: yes
Referee: [Conclusion / Scope] Conclusion and Scope: the paper frames its findings as relevant to 'closed-source simulation frameworks' in general, yet all experiments are performed on a single mesh-free code (Pasimodo) without any argument or cross-framework test establishing that its documentation style or data layout is representative; this makes the claimed generality of the tailoring benefit an untested extrapolation.

Authors: The experiments are confined to Pasimodo. We will revise the conclusion and abstract to state that the observed benefit of prompt-dependent tailoring is demonstrated for this framework and to present cross-framework validation as necessary future work rather than an established general result. revision: partial

Circularity Check

0 steps flagged

No circularity: purely experimental evaluation with no derivations or self-referential claims

full rationale

The paper conducts empirical tests of RAG systems on Pasimodo documentation and prompts, reporting observed failures and improvements from prompt-dependent tailoring. No equations, fitted parameters, uniqueness theorems, or self-citations appear as load-bearing elements in any derivation chain. All claims rest on direct experimental outcomes rather than reducing to inputs by construction, satisfying the self-contained experimental criterion for score 0.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

The paper is an empirical feasibility study and introduces no new theoretical constructs, fitted parameters, or postulated entities.

pith-pipeline@v0.9.0 · 5753 in / 1000 out tokens · 31971 ms · 2026-05-23T04:26:29.899183+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

66 extracted references · 66 canonical work pages · 2 internal anchors

[1]

http://jalammar.github.io/illustrated-transformer/, last accessed January 10, 2025

Alammar, J.: The illustrated transformer (2018). http://jalammar.github.io/illustrated-transformer/, last accessed January 10, 2025

work page 2018
[2]

Ali, M., Fromm, M., Thellmann, K., et al.: Teuken-7B- Base & Teuken-7B-Instruct: Towards European LLMs (2024)

work page 2024
[3]

https://docs.useanything.com/llm-not- using-my-docs, last accessed January 10, 2025

AnythingLLM Docs: Why does the llm not use my doc- uments? (2024). https://docs.useanything.com/llm-not- using-my-docs, last accessed January 10, 2025

work page 2024
[4]

Baumann, A., Eberhard, P.: Investigation of chip jam- ming in deep-hole drilling. In: G. Fourtakas (ed.) Pro- ceedings of the 17th International SPHERIC Workshop, Rhodes, Greece, June 27-29, 2023 (2023)

work page 2023
[5]

Computational Par- ticle Mechanics11, 1851–1862 (2024)

Baumann, A., Gerken, J.F., Sollich, D., Rupasinghe, N., Biermann, D., Eberhard, P.: Modeling and mitigation of vortex formation in ejector deep hole drilling with smoothed particle hydrodynamics. Computational Par- ticle Mechanics11, 1851–1862 (2024)

work page 2024
[6]

Proceedings of the III International Conference on Particle-based Methods - Fundamentals and Applications PARTICLES (2013)

Beck, F., Fleissner, F., Eberhard, P.: A fully Lagrangian approach for modelling abrasive wear. Proceedings of the III International Conference on Particle-based Methods - Fundamentals and Applications PARTICLES (2013)

work page 2013
[7]

https://beebom.com/how-use-gemini-1-5-flash/, last ac- cessed January 10, 2025 15

Beebom: Gemini 1.5 flash is an underrated gem you need to try right now: Here’s how (2024). https://beebom.com/how-use-gemini-1-5-flash/, last ac- cessed January 10, 2025 15

work page 2024
[8]

Computer Methods in Applied Mechanics and Engineering180(1), 97–115 (1999)

Bonet, J., Lok, T.S.: Variational and momentum preser- vation aspects of smooth particle hydrodynamic formu- lations. Computer Methods in Applied Mechanics and Engineering180(1), 97–115 (1999)

work page 1999
[9]

Advances in Neural Information Processing Systems33, 1877–1901 (2020)

Brown, T., Mann, B., Ryder, N., Subbiah, M., Kaplan, J.D., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., Agarwal, S., Herbert-Voss, A., Krueger, G., Henighan, T., Child, R., Ramesh, A., Ziegler, D., Wu, J., Winter, C., Hesse, C., Chen, M., Sigler, E., Litwin, M., Gray, S., Chess, B., Clark, J., Berner, C., McCandlish, S., Radford, A., ...

work page 1901
[10]

Princeton Univ

Cole, R.H.: Underwater explosions. Princeton Univ. Press, Princeton, New Jersey (1948)

work page 1948
[11]

G´ eotechnique29(1), 47– 65 (1979)

Cundall, P.A., Strack, O.D.L.: A discrete numerical model for granular assemblies. G´ eotechnique29(1), 47– 65 (1979)

work page 1979
[12]

4171–4186

Devlin, J., Chang, M.W., Lee, K., Toutanova, K.: In: Pro- ceedings of NAACL-HLT 2019, pp. 4171–4186. Associa- tion for Computational Linguistics (2019)

work page 2019
[13]

Procedia IUTAM10, 161–179 (2014)

Eberhard, P., Seifried, R., Ergenzinger, C., St¨ uhler, S., Spreng, F., Beck, F., M¨ uller, A., Fleissner, F.: Particles - bridging the gap between solids and fluids. Procedia IUTAM10, 161–179 (2014)

work page 2014
[14]

Eberhard, P., Spreng, F.: Smoothed particle hydrody- namics for ductile solid continua. In: C.H. Hsueh, S. Schmauder, C.S. Chen, K.K. Chawla, N. Chawla, W. Chen, Y. Kagawa (eds.) Handbook of Mechanics of Materials, pp. 1415–1463. Springer, Singapore (2019)

work page 2019
[15]

https://explodingtopics.com/blog/gpt- parameters, last accessed January 10, 2025

Exploring Topics: Number of parameters in gpt-4 (lat- est data) (2024). https://explodingtopics.com/blog/gpt- parameters, last accessed January 10, 2025

work page 2024
[16]

URL https://arxiv.org/abs/2504.19338

Feng, J., Xu, R., Chu, X.: Openfoamgpt 2.0: end-to-end, trustworthy automation for computational fluid dynam- ics (2025). URL https://arxiv.org/abs/2504.19338

work page arXiv 2025
[17]

Fleissner, F.: Parallel Object Oriented Simulation with Lagrangian Particle Methods,Schriften aus dem Institut f¨ ur Technische und Numerische Mechanik der Univer- sit¨ at Stuttgart, vol. 16. Shaker Verlag, Aachen (2010)

work page 2010
[18]

Gao, Y., Xiong, Y., Gao, X., Jia, K., Pan, J., Bi, Y., Dai, Y., Sun, J., Wang, M., Wang, H.: Retrieval-augmented generation for large language models: A survey (2023)

work page 2023
[19]

Gemma 3 Technical Report

Gemma Team, et al.: Gemma 3 technical report (2025). URL https://arxiv.org/abs/2503.19786

work page internal anchor Pith review Pith/arXiv arXiv 2025
[20]

Multibody System Dy- namics60(4), 533–561 (2023)

Gerstmayr, J.: Exudyn - a C++-based Python package for flexible multibody systems. Multibody System Dy- namics60(4), 533–561 (2023)

work page 2023
[21]

Multibody System Dynamics62(2), 249–271 (2024)

Gerstmayr, J., Manzl, P., Pieber, M.: Multibody mod- els generated from natural language. Multibody System Dynamics62(2), 249–271 (2024)

work page 2024
[22]

Gnanasambandham, C., Eberhard, P.: Modelling a par- tially liquid-filled particle damper using coupled La- grangian methods. In: E. O˜ nate, M. Bischoff, D. Owen, P. Wriggers, T. Zohdi (eds.) Proceedings of the VI Inter- national Conference on Particle-based Methods - Fun- damentals and Applications PARTICLES, Barcelona, Spain, October, 28-30, 2019 (2019)

work page 2019
[23]

Journal of Hydraulic Research48, 6–27 (2010)

Gomez-Gesteira, M., Rogers, B.D., Dalrymple, R.A., Crespo, A.J.: State-of-the-art of classical sph for free- surface flows. Journal of Hydraulic Research48, 6–27 (2010)

work page 2010
[24]

https://ai.google.dev/gemini- api/docs/models/gemini#gemini-2.5-flash, last accessed December 1st, 2025

Google: Gemini models: Gemini 2.5 flash (2025). https://ai.google.dev/gemini- api/docs/models/gemini#gemini-2.5-flash, last accessed December 1st, 2025

work page 2025
[25]

https://notebooklm.google/, last accessed December 1st, 2025

Google: Notebooklm (2025). https://notebooklm.google/, last accessed December 1st, 2025

work page 2025
[26]

https://ai.google.dev/gemma/docs/core/model card 3, last accessed December 1st, 2025

Google DeepMind: Gemma 3 model card (2025). https://ai.google.dev/gemma/docs/core/model card 3, last accessed December 1st, 2025

work page 2025
[27]

Grattafiori, A., Dubey, A., Jauhri, A., et al.: The llama 3 herd of models (2024)

work page 2024
[28]

IEEE Access13, 46171–46189 (2025)

Hindi, M., Mohammed, L., Maaz, O., Alwarafy, A.: En- hancing the precision and interpretability of retrieval- augmented generation (rag) in legal technology: A survey. IEEE Access13, 46171–46189 (2025)

work page 2025
[29]

ACM Computing Surveys55(12), 1–38 (2023)

Ji, Z., Lee, N., Frieske, R., Yu, T., Su, D., Xu, Y., Ishii, E., Bang, Y.J., Madotto, A., Fung, P.: Survey of halluci- nation in natural language generation. ACM Computing Surveys55(12), 1–38 (2023)

work page 2023
[30]

URL http://arxiv.org/abs/2211.08411

Kandpal, N., Deng, H., Roberts, A., Wallace, E., Raffel, C.: Large language models strug- gle to learn long-tail knowledge (2023). URL http://arxiv.org/abs/2211.08411

work page arXiv 2023
[31]

In: Proceedings of the 34th International Conference on Neural Information Processing Systems, NIPS ’20

Lewis, P., Perez, E., Piktus, A., Petroni, F., Karpukhin, V., Goyal, N., K¨ uttler, H., Lewis, M., Yih, W.t., Rockt¨ aschel, T., Riedel, S., Kiela, D.: Retrieval- augmented generation for knowledge-intensive nlp tasks. In: Proceedings of the 34th International Conference on Neural Information Processing Systems, NIPS ’20. Cur- ran Associates Inc., Red Hoo...

work page 2020
[32]

Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sci- ences476(2241) (2020)

Lind, S.J., Rogers, B.D., Stansby, P.K.: Review of smoothed particle hydrodynamics: towards converged Lagrangian flow modelling. Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sci- ences476(2241) (2020)

work page 2020
[33]

Archives of Computational Methods in Engineering17(1), 25–76 (2010)

Liu, M.B., Liu, G.R.: Smoothed particle hydrodynamics (sph): an overview and recent developments. Archives of Computational Methods in Engineering17(1), 25–76 (2010)

work page 2010
[34]

Liu, Y., He, H., Han, T., Zhang, X., Liu, M., Tian, J., Zhang, Y., Wang, J., Gao, X., Zhong, T., Pan, Y., Xu, S., Wu, Z., Liu, Z., Zhang, X., Zhang, S., Hu, X., Zhang, T., Qiang, N., Liu, T., Ge, B.: Understanding llms: A comprehensive overview from training to inference (2024)

work page 2024
[35]

https://ai.meta.com/blog/llama-4-multimodal- intelligence, last accessed November 30th, 2025

Meta AI: The llama 4 herd: The beginning of a new era of natively multimodal ai innovation (2025). https://ai.meta.com/blog/llama-4-multimodal- intelligence, last accessed November 30th, 2025

work page 2025
[36]

https://github.com/meta-llama/llama- models/blob/main/models/llama3 2/MODEL CARD.md, last accessed December 1st, 2025

Meta Llama: Model information (2024). https://github.com/meta-llama/llama- models/blob/main/models/llama3 2/MODEL CARD.md, last accessed December 1st, 2025

work page 2024
[37]

https://www.llama.com/docs/model-cards-and-prompt- formats/llama3 1#-tool-calling-(8b/70b/405b)-, last accessed January 10, 2025

Meta Llama: Tool calling (8b/70b/405b) (2024). https://www.llama.com/docs/model-cards-and-prompt- formats/llama3 1#-tool-calling-(8b/70b/405b)-, last accessed January 10, 2025

work page 2024
[38]

Journal of Medical Internet Re- search27, e80557 (2025)

Miao, Y., Zhao, Y., Luo, Y., Wang, H., Wu, Y.: Im- proving large language model applications in the medical and nursing domains with retrieval-augmented genera- tion: Scoping review. Journal of Medical Internet Re- search27, e80557 (2025)

work page 2025
[39]

https://github.com/Mintplex-Labs/anything-llm/, last accessed December 1st, 2025

Mintplex Labs: Anythingllm (2025). https://github.com/Mintplex-Labs/anything-llm/, last accessed December 1st, 2025

work page 2025
[40]

M¨ oltner, T., Manzl, P., Pieber, M., Gerstmayr, J.: Cre- ation, evaluation and self-validation of simulation models with large language models (2025)

work page 2025
[41]

Journal of Computational Physics159(2), 290–311 (2000) 16

Monaghan, J.: Sph without a tensile instability. Journal of Computational Physics159(2), 290–311 (2000) 16

work page 2000
[42]

Journal of Computational Physics110(2), 399–406 (1994)

Monaghan, J.J.: Simulating free surface flows with sph. Journal of Computational Physics110(2), 399–406 (1994)

work page 1994
[43]

Re- ports on Progress in Physics68(8), 1703–1759 (2005)

Monaghan, J.J.: Smoothed particle hydrodynamics. Re- ports on Progress in Physics68(8), 1703–1759 (2005)

work page 2005
[44]

Journal of Waterway, Port, Coastal, and Ocean Engineering129(6), 250–259 (2003)

Monaghan, J.J., Kos, A., Issa, N.: Fluid motion gener- ated by impact. Journal of Waterway, Port, Coastal, and Ocean Engineering129(6), 250–259 (2003)

work page 2003
[45]

https://github.com/ollama/ollama/, last accessed December 1st, 2025

Ollama: Ollama (2024). https://github.com/ollama/ollama/, last accessed December 1st, 2025

work page 2024
[46]

https://platform.openai.com/docs/models/gpt-4o#gpt- 4o, last accessed December 1st, 2025

OpenAi: Models (2024). https://platform.openai.com/docs/models/gpt-4o#gpt- 4o, last accessed December 1st, 2025

work page 2024
[47]

Physics of Fluids37(3) (2025)

Pandey, S., Xu, R., Wang, W., Chu, X.: Openfoamgpt: A retrieval-augmented large language model (llm) agent for openfoam-based computational fluid dynamics. Physics of Fluids37(3) (2025)

work page 2025
[48]

Peeperkorn, M., Kouwenhoven, T., Brown, D., Jor- danous, A.: Is temperature the creativity parameter of large language models? In: Proceedings of the 15th International Conference on Computational Creativity (ICCC’24) (2024)

work page 2024
[49]

perplexity: Chatgpt-3.5 vs. 4 vs. 4o: Key differences (2024). https://www.perplexity.ai/page/chatgpt-3-5-vs- 4-vs-4o-key-dif-nEXAloovT1iUx9Snare0lQ, last accessed January 10, 2025

work page 2024
[50]

https://towardsdatascience.com/illustrated-guide-to- transformers-step-by-step-explanation-f74876522bc0, last accessed January 10, 2025

Phi, M.: Illustrated guide to transformers- step by step explanation (2020). https://towardsdatascience.com/illustrated-guide-to- transformers-step-by-step-explanation-f74876522bc0, last accessed January 10, 2025

work page 2020
[51]

Ranaldi, L., Haddow, B., Birch, A.: Multilingual retrieval-augmented generation for knowledge-intensive task (2025)

work page 2025
[52]

URL https://huggingface.co/blog/moe

Sanseviero, O., Tunstall, L., Schmid, P., Mangrulkar, S., Belkada, Y., Cuenca, P.: Mixture of experts explained (2023). URL https://huggingface.co/blog/moe

work page 2023
[53]

In: Proceedings of the 12th SPHERIC World Conference, Ourense, Spain, June 13- 15, 2017 (2017)

Schnabel, D., Eberhard, P.: Sph simulation of single-lip deep-hole drilling processes. In: Proceedings of the 12th SPHERIC World Conference, Ourense, Spain, June 13- 15, 2017 (2017)

work page 2017
[54]

In: 5th International Conference on Learning Represen- tations, ICLR 2017, Toulon, France, April 24-26, 2017, Conference Track Proceedings (2017)

Shazeer, N., Mirhoseini, A., Maziarz, K., Davis, A., Le, Q.V., Hinton, G.E., Dean, J.: Outrageously large neural networks: The sparsely-gated mixture-of-experts layer. In: 5th International Conference on Learning Represen- tations, ICLR 2017, Toulon, France, April 24-26, 2017, Conference Track Proceedings (2017)

work page 2017
[55]

Computational Particle Mechanics6(3), 449–460 (2019)

Shishova, E., Spreng, F., Hamann, D., Eberhard, P.: Tracking of material orientation in updated Lagrangian sph. Computational Particle Mechanics6(3), 449–460 (2019)

work page 2019
[56]

Shuster, K., Poff, S., Chen, M., Kiela, D., Weston, J.: Retrieval augmentation reduces hallucination in conver- sation. In: M. Moens, X. Huang, L. Specia, S.W. tau Yih (eds.) Findings of the Association for Computational Lin- guistics: EMNLP 2021, pp. 3784–3803 (2021)

work page 2021
[57]

In: VIII International Conference on Particle-Based Methods, Milan, Italy, October 9-11, 2023, PARTICLES 2023

Sollich, D., Eberhard, P.: Comparison of incompress- ible and weakly-compressible sph for the simulation of laser beam welding. In: VIII International Conference on Particle-Based Methods, Milan, Italy, October 9-11, 2023, PARTICLES 2023. CIMNE (2023)

work page 2023
[58]

In: Proceedings of the 17th SPHERIC Inter- national Workshop, Rhodes, Greece, June 27-29, 2023 (2023)

Sollich, D., Eberhard, P.: Modeling of pore formation in deep penetration laser beam welding using the sph method. In: Proceedings of the 17th SPHERIC Inter- national Workshop, Rhodes, Greece, June 27-29, 2023 (2023)

work page 2023
[59]

European Journal of Me- chanics - B/Fluids96, 26–38 (2022)

Sollich, D., Reinheimer, E.N., Wagner, J., Berger, P., Eberhard, P.: An improved recoil pressure boundary con- dition for the simulation of deep penetration laser beam welding using the sph method. European Journal of Me- chanics - B/Fluids96, 26–38 (2022)

work page 2022
[60]

Computational Particle Mechanics 1(2), 131–145 (2014)

Spreng, F., Schnabel, D., Mueller, A., Eberhard, P.: A local adaptive discretization algorithm for smoothed par- ticle hydrodynamics. Computational Particle Mechanics 1(2), 131–145 (2014)

work page 2014
[61]

https://mistral.ai/news/mixtral-of-experts, last ac- cessed November 30th, 2025

team, M.A.: Mixtral of experts - a high quality sparse mixture-of-experts (2023). https://mistral.ai/news/mixtral-of-experts, last ac- cessed November 30th, 2025

work page 2023
[62]

Computational Particle Mechanics8(3), 575–588 (2020)

Vacondio, R., Altomare, C., De Leffe, M., Hu, X., Le Touz´ e, D., Lind, S., Marongiu, J.C., Marrone, S., Rogers, B.D., Souto-Iglesias, A.: Grand challenges for smoothed particle hydrodynamics numerical schemes. Computational Particle Mechanics8(3), 575–588 (2020)

work page 2020
[63]

Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A.N., Kaiser, L.u., Polosukhin, I.: At- tention is all you need. In: I. Guyon, U.V. Luxburg, S. Bengio, H. Wallach, R. Fergus, S. Vishwanathan, R. Garnett (eds.) Advances in Neural Information Pro- cessing Systems, vol. 30. Curran Associates, Inc. (2017)

work page 2017
[64]

Journal of Hydraulic Research54(1), 1–26 (2016)

Violeau, D., Rogers, B.D.: Smoothed particle hydrody- namics (sph) for free-surface flows: past, present and fu- ture. Journal of Hydraulic Research54(1), 1–26 (2016)

work page 2016
[65]

In: Proceedings of the 38th Conference on Neural Information Processing Systems (NeurIPS 2024) (2024)

Wu, K., Wu, E., Zou, J.: Clasheval: Quantifying the tug- of-war between an llm’s internal prior and external evi- dence. In: Proceedings of the 38th Conference on Neural Information Processing Systems (NeurIPS 2024) (2024)

work page 2024
[66]

Retrieval-Augmented Generation for AI-Generated Content: A Survey

Zhao, P., Zhang, H., Yu, Q., Wang, Z., Geng, Y., Fu, F., Yang, L., Zhang, W., Jiang, J., Cui, B.: Retrieval- augmented generation for ai-generated content: A survey (2024). URL https://arxiv.org/abs/2402.19473

work page internal anchor Pith review Pith/arXiv arXiv 2024

[1] [1]

http://jalammar.github.io/illustrated-transformer/, last accessed January 10, 2025

Alammar, J.: The illustrated transformer (2018). http://jalammar.github.io/illustrated-transformer/, last accessed January 10, 2025

work page 2018

[2] [2]

Ali, M., Fromm, M., Thellmann, K., et al.: Teuken-7B- Base & Teuken-7B-Instruct: Towards European LLMs (2024)

work page 2024

[3] [3]

https://docs.useanything.com/llm-not- using-my-docs, last accessed January 10, 2025

AnythingLLM Docs: Why does the llm not use my doc- uments? (2024). https://docs.useanything.com/llm-not- using-my-docs, last accessed January 10, 2025

work page 2024

[4] [4]

Baumann, A., Eberhard, P.: Investigation of chip jam- ming in deep-hole drilling. In: G. Fourtakas (ed.) Pro- ceedings of the 17th International SPHERIC Workshop, Rhodes, Greece, June 27-29, 2023 (2023)

work page 2023

[5] [5]

Computational Par- ticle Mechanics11, 1851–1862 (2024)

Baumann, A., Gerken, J.F., Sollich, D., Rupasinghe, N., Biermann, D., Eberhard, P.: Modeling and mitigation of vortex formation in ejector deep hole drilling with smoothed particle hydrodynamics. Computational Par- ticle Mechanics11, 1851–1862 (2024)

work page 2024

[6] [6]

Proceedings of the III International Conference on Particle-based Methods - Fundamentals and Applications PARTICLES (2013)

Beck, F., Fleissner, F., Eberhard, P.: A fully Lagrangian approach for modelling abrasive wear. Proceedings of the III International Conference on Particle-based Methods - Fundamentals and Applications PARTICLES (2013)

work page 2013

[7] [7]

https://beebom.com/how-use-gemini-1-5-flash/, last ac- cessed January 10, 2025 15

Beebom: Gemini 1.5 flash is an underrated gem you need to try right now: Here’s how (2024). https://beebom.com/how-use-gemini-1-5-flash/, last ac- cessed January 10, 2025 15

work page 2024

[8] [8]

Computer Methods in Applied Mechanics and Engineering180(1), 97–115 (1999)

Bonet, J., Lok, T.S.: Variational and momentum preser- vation aspects of smooth particle hydrodynamic formu- lations. Computer Methods in Applied Mechanics and Engineering180(1), 97–115 (1999)

work page 1999

[9] [9]

Advances in Neural Information Processing Systems33, 1877–1901 (2020)

Brown, T., Mann, B., Ryder, N., Subbiah, M., Kaplan, J.D., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., Agarwal, S., Herbert-Voss, A., Krueger, G., Henighan, T., Child, R., Ramesh, A., Ziegler, D., Wu, J., Winter, C., Hesse, C., Chen, M., Sigler, E., Litwin, M., Gray, S., Chess, B., Clark, J., Berner, C., McCandlish, S., Radford, A., ...

work page 1901

[10] [10]

Princeton Univ

Cole, R.H.: Underwater explosions. Princeton Univ. Press, Princeton, New Jersey (1948)

work page 1948

[11] [11]

G´ eotechnique29(1), 47– 65 (1979)

Cundall, P.A., Strack, O.D.L.: A discrete numerical model for granular assemblies. G´ eotechnique29(1), 47– 65 (1979)

work page 1979

[12] [12]

4171–4186

Devlin, J., Chang, M.W., Lee, K., Toutanova, K.: In: Pro- ceedings of NAACL-HLT 2019, pp. 4171–4186. Associa- tion for Computational Linguistics (2019)

work page 2019

[13] [13]

Procedia IUTAM10, 161–179 (2014)

Eberhard, P., Seifried, R., Ergenzinger, C., St¨ uhler, S., Spreng, F., Beck, F., M¨ uller, A., Fleissner, F.: Particles - bridging the gap between solids and fluids. Procedia IUTAM10, 161–179 (2014)

work page 2014

[14] [14]

Eberhard, P., Spreng, F.: Smoothed particle hydrody- namics for ductile solid continua. In: C.H. Hsueh, S. Schmauder, C.S. Chen, K.K. Chawla, N. Chawla, W. Chen, Y. Kagawa (eds.) Handbook of Mechanics of Materials, pp. 1415–1463. Springer, Singapore (2019)

work page 2019

[15] [15]

https://explodingtopics.com/blog/gpt- parameters, last accessed January 10, 2025

Exploring Topics: Number of parameters in gpt-4 (lat- est data) (2024). https://explodingtopics.com/blog/gpt- parameters, last accessed January 10, 2025

work page 2024

[16] [16]

URL https://arxiv.org/abs/2504.19338

Feng, J., Xu, R., Chu, X.: Openfoamgpt 2.0: end-to-end, trustworthy automation for computational fluid dynam- ics (2025). URL https://arxiv.org/abs/2504.19338

work page arXiv 2025

[17] [17]

Fleissner, F.: Parallel Object Oriented Simulation with Lagrangian Particle Methods,Schriften aus dem Institut f¨ ur Technische und Numerische Mechanik der Univer- sit¨ at Stuttgart, vol. 16. Shaker Verlag, Aachen (2010)

work page 2010

[18] [18]

Gao, Y., Xiong, Y., Gao, X., Jia, K., Pan, J., Bi, Y., Dai, Y., Sun, J., Wang, M., Wang, H.: Retrieval-augmented generation for large language models: A survey (2023)

work page 2023

[19] [19]

Gemma 3 Technical Report

Gemma Team, et al.: Gemma 3 technical report (2025). URL https://arxiv.org/abs/2503.19786

work page internal anchor Pith review Pith/arXiv arXiv 2025

[20] [20]

Multibody System Dy- namics60(4), 533–561 (2023)

Gerstmayr, J.: Exudyn - a C++-based Python package for flexible multibody systems. Multibody System Dy- namics60(4), 533–561 (2023)

work page 2023

[21] [21]

Multibody System Dynamics62(2), 249–271 (2024)

Gerstmayr, J., Manzl, P., Pieber, M.: Multibody mod- els generated from natural language. Multibody System Dynamics62(2), 249–271 (2024)

work page 2024

[22] [22]

Gnanasambandham, C., Eberhard, P.: Modelling a par- tially liquid-filled particle damper using coupled La- grangian methods. In: E. O˜ nate, M. Bischoff, D. Owen, P. Wriggers, T. Zohdi (eds.) Proceedings of the VI Inter- national Conference on Particle-based Methods - Fun- damentals and Applications PARTICLES, Barcelona, Spain, October, 28-30, 2019 (2019)

work page 2019

[23] [23]

Journal of Hydraulic Research48, 6–27 (2010)

Gomez-Gesteira, M., Rogers, B.D., Dalrymple, R.A., Crespo, A.J.: State-of-the-art of classical sph for free- surface flows. Journal of Hydraulic Research48, 6–27 (2010)

work page 2010

[24] [24]

https://ai.google.dev/gemini- api/docs/models/gemini#gemini-2.5-flash, last accessed December 1st, 2025

Google: Gemini models: Gemini 2.5 flash (2025). https://ai.google.dev/gemini- api/docs/models/gemini#gemini-2.5-flash, last accessed December 1st, 2025

work page 2025

[25] [25]

https://notebooklm.google/, last accessed December 1st, 2025

Google: Notebooklm (2025). https://notebooklm.google/, last accessed December 1st, 2025

work page 2025

[26] [26]

https://ai.google.dev/gemma/docs/core/model card 3, last accessed December 1st, 2025

Google DeepMind: Gemma 3 model card (2025). https://ai.google.dev/gemma/docs/core/model card 3, last accessed December 1st, 2025

work page 2025

[27] [27]

Grattafiori, A., Dubey, A., Jauhri, A., et al.: The llama 3 herd of models (2024)

work page 2024

[28] [28]

IEEE Access13, 46171–46189 (2025)

Hindi, M., Mohammed, L., Maaz, O., Alwarafy, A.: En- hancing the precision and interpretability of retrieval- augmented generation (rag) in legal technology: A survey. IEEE Access13, 46171–46189 (2025)

work page 2025

[29] [29]

ACM Computing Surveys55(12), 1–38 (2023)

Ji, Z., Lee, N., Frieske, R., Yu, T., Su, D., Xu, Y., Ishii, E., Bang, Y.J., Madotto, A., Fung, P.: Survey of halluci- nation in natural language generation. ACM Computing Surveys55(12), 1–38 (2023)

work page 2023

[30] [30]

URL http://arxiv.org/abs/2211.08411

Kandpal, N., Deng, H., Roberts, A., Wallace, E., Raffel, C.: Large language models strug- gle to learn long-tail knowledge (2023). URL http://arxiv.org/abs/2211.08411

work page arXiv 2023

[31] [31]

In: Proceedings of the 34th International Conference on Neural Information Processing Systems, NIPS ’20

Lewis, P., Perez, E., Piktus, A., Petroni, F., Karpukhin, V., Goyal, N., K¨ uttler, H., Lewis, M., Yih, W.t., Rockt¨ aschel, T., Riedel, S., Kiela, D.: Retrieval- augmented generation for knowledge-intensive nlp tasks. In: Proceedings of the 34th International Conference on Neural Information Processing Systems, NIPS ’20. Cur- ran Associates Inc., Red Hoo...

work page 2020

[32] [32]

Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sci- ences476(2241) (2020)

Lind, S.J., Rogers, B.D., Stansby, P.K.: Review of smoothed particle hydrodynamics: towards converged Lagrangian flow modelling. Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sci- ences476(2241) (2020)

work page 2020

[33] [33]

Archives of Computational Methods in Engineering17(1), 25–76 (2010)

Liu, M.B., Liu, G.R.: Smoothed particle hydrodynamics (sph): an overview and recent developments. Archives of Computational Methods in Engineering17(1), 25–76 (2010)

work page 2010

[34] [34]

Liu, Y., He, H., Han, T., Zhang, X., Liu, M., Tian, J., Zhang, Y., Wang, J., Gao, X., Zhong, T., Pan, Y., Xu, S., Wu, Z., Liu, Z., Zhang, X., Zhang, S., Hu, X., Zhang, T., Qiang, N., Liu, T., Ge, B.: Understanding llms: A comprehensive overview from training to inference (2024)

work page 2024

[35] [35]

https://ai.meta.com/blog/llama-4-multimodal- intelligence, last accessed November 30th, 2025

Meta AI: The llama 4 herd: The beginning of a new era of natively multimodal ai innovation (2025). https://ai.meta.com/blog/llama-4-multimodal- intelligence, last accessed November 30th, 2025

work page 2025

[36] [36]

https://github.com/meta-llama/llama- models/blob/main/models/llama3 2/MODEL CARD.md, last accessed December 1st, 2025

Meta Llama: Model information (2024). https://github.com/meta-llama/llama- models/blob/main/models/llama3 2/MODEL CARD.md, last accessed December 1st, 2025

work page 2024

[37] [37]

https://www.llama.com/docs/model-cards-and-prompt- formats/llama3 1#-tool-calling-(8b/70b/405b)-, last accessed January 10, 2025

Meta Llama: Tool calling (8b/70b/405b) (2024). https://www.llama.com/docs/model-cards-and-prompt- formats/llama3 1#-tool-calling-(8b/70b/405b)-, last accessed January 10, 2025

work page 2024

[38] [38]

Journal of Medical Internet Re- search27, e80557 (2025)

Miao, Y., Zhao, Y., Luo, Y., Wang, H., Wu, Y.: Im- proving large language model applications in the medical and nursing domains with retrieval-augmented genera- tion: Scoping review. Journal of Medical Internet Re- search27, e80557 (2025)

work page 2025

[39] [39]

https://github.com/Mintplex-Labs/anything-llm/, last accessed December 1st, 2025

Mintplex Labs: Anythingllm (2025). https://github.com/Mintplex-Labs/anything-llm/, last accessed December 1st, 2025

work page 2025

[40] [40]

M¨ oltner, T., Manzl, P., Pieber, M., Gerstmayr, J.: Cre- ation, evaluation and self-validation of simulation models with large language models (2025)

work page 2025

[41] [41]

Journal of Computational Physics159(2), 290–311 (2000) 16

Monaghan, J.: Sph without a tensile instability. Journal of Computational Physics159(2), 290–311 (2000) 16

work page 2000

[42] [42]

Journal of Computational Physics110(2), 399–406 (1994)

Monaghan, J.J.: Simulating free surface flows with sph. Journal of Computational Physics110(2), 399–406 (1994)

work page 1994

[43] [43]

Re- ports on Progress in Physics68(8), 1703–1759 (2005)

Monaghan, J.J.: Smoothed particle hydrodynamics. Re- ports on Progress in Physics68(8), 1703–1759 (2005)

work page 2005

[44] [44]

Journal of Waterway, Port, Coastal, and Ocean Engineering129(6), 250–259 (2003)

Monaghan, J.J., Kos, A., Issa, N.: Fluid motion gener- ated by impact. Journal of Waterway, Port, Coastal, and Ocean Engineering129(6), 250–259 (2003)

work page 2003

[45] [45]

https://github.com/ollama/ollama/, last accessed December 1st, 2025

Ollama: Ollama (2024). https://github.com/ollama/ollama/, last accessed December 1st, 2025

work page 2024

[46] [46]

https://platform.openai.com/docs/models/gpt-4o#gpt- 4o, last accessed December 1st, 2025

OpenAi: Models (2024). https://platform.openai.com/docs/models/gpt-4o#gpt- 4o, last accessed December 1st, 2025

work page 2024

[47] [47]

Physics of Fluids37(3) (2025)

Pandey, S., Xu, R., Wang, W., Chu, X.: Openfoamgpt: A retrieval-augmented large language model (llm) agent for openfoam-based computational fluid dynamics. Physics of Fluids37(3) (2025)

work page 2025

[48] [48]

Peeperkorn, M., Kouwenhoven, T., Brown, D., Jor- danous, A.: Is temperature the creativity parameter of large language models? In: Proceedings of the 15th International Conference on Computational Creativity (ICCC’24) (2024)

work page 2024

[49] [49]

perplexity: Chatgpt-3.5 vs. 4 vs. 4o: Key differences (2024). https://www.perplexity.ai/page/chatgpt-3-5-vs- 4-vs-4o-key-dif-nEXAloovT1iUx9Snare0lQ, last accessed January 10, 2025

work page 2024

[50] [50]

https://towardsdatascience.com/illustrated-guide-to- transformers-step-by-step-explanation-f74876522bc0, last accessed January 10, 2025

Phi, M.: Illustrated guide to transformers- step by step explanation (2020). https://towardsdatascience.com/illustrated-guide-to- transformers-step-by-step-explanation-f74876522bc0, last accessed January 10, 2025

work page 2020

[51] [51]

Ranaldi, L., Haddow, B., Birch, A.: Multilingual retrieval-augmented generation for knowledge-intensive task (2025)

work page 2025

[52] [52]

URL https://huggingface.co/blog/moe

Sanseviero, O., Tunstall, L., Schmid, P., Mangrulkar, S., Belkada, Y., Cuenca, P.: Mixture of experts explained (2023). URL https://huggingface.co/blog/moe

work page 2023

[53] [53]

In: Proceedings of the 12th SPHERIC World Conference, Ourense, Spain, June 13- 15, 2017 (2017)

Schnabel, D., Eberhard, P.: Sph simulation of single-lip deep-hole drilling processes. In: Proceedings of the 12th SPHERIC World Conference, Ourense, Spain, June 13- 15, 2017 (2017)

work page 2017

[54] [54]

In: 5th International Conference on Learning Represen- tations, ICLR 2017, Toulon, France, April 24-26, 2017, Conference Track Proceedings (2017)

Shazeer, N., Mirhoseini, A., Maziarz, K., Davis, A., Le, Q.V., Hinton, G.E., Dean, J.: Outrageously large neural networks: The sparsely-gated mixture-of-experts layer. In: 5th International Conference on Learning Represen- tations, ICLR 2017, Toulon, France, April 24-26, 2017, Conference Track Proceedings (2017)

work page 2017

[55] [55]

Computational Particle Mechanics6(3), 449–460 (2019)

Shishova, E., Spreng, F., Hamann, D., Eberhard, P.: Tracking of material orientation in updated Lagrangian sph. Computational Particle Mechanics6(3), 449–460 (2019)

work page 2019

[56] [56]

Shuster, K., Poff, S., Chen, M., Kiela, D., Weston, J.: Retrieval augmentation reduces hallucination in conver- sation. In: M. Moens, X. Huang, L. Specia, S.W. tau Yih (eds.) Findings of the Association for Computational Lin- guistics: EMNLP 2021, pp. 3784–3803 (2021)

work page 2021

[57] [57]

In: VIII International Conference on Particle-Based Methods, Milan, Italy, October 9-11, 2023, PARTICLES 2023

Sollich, D., Eberhard, P.: Comparison of incompress- ible and weakly-compressible sph for the simulation of laser beam welding. In: VIII International Conference on Particle-Based Methods, Milan, Italy, October 9-11, 2023, PARTICLES 2023. CIMNE (2023)

work page 2023

[58] [58]

In: Proceedings of the 17th SPHERIC Inter- national Workshop, Rhodes, Greece, June 27-29, 2023 (2023)

Sollich, D., Eberhard, P.: Modeling of pore formation in deep penetration laser beam welding using the sph method. In: Proceedings of the 17th SPHERIC Inter- national Workshop, Rhodes, Greece, June 27-29, 2023 (2023)

work page 2023

[59] [59]

European Journal of Me- chanics - B/Fluids96, 26–38 (2022)

Sollich, D., Reinheimer, E.N., Wagner, J., Berger, P., Eberhard, P.: An improved recoil pressure boundary con- dition for the simulation of deep penetration laser beam welding using the sph method. European Journal of Me- chanics - B/Fluids96, 26–38 (2022)

work page 2022

[60] [60]

Computational Particle Mechanics 1(2), 131–145 (2014)

Spreng, F., Schnabel, D., Mueller, A., Eberhard, P.: A local adaptive discretization algorithm for smoothed par- ticle hydrodynamics. Computational Particle Mechanics 1(2), 131–145 (2014)

work page 2014

[61] [61]

https://mistral.ai/news/mixtral-of-experts, last ac- cessed November 30th, 2025

team, M.A.: Mixtral of experts - a high quality sparse mixture-of-experts (2023). https://mistral.ai/news/mixtral-of-experts, last ac- cessed November 30th, 2025

work page 2023

[62] [62]

Computational Particle Mechanics8(3), 575–588 (2020)

Vacondio, R., Altomare, C., De Leffe, M., Hu, X., Le Touz´ e, D., Lind, S., Marongiu, J.C., Marrone, S., Rogers, B.D., Souto-Iglesias, A.: Grand challenges for smoothed particle hydrodynamics numerical schemes. Computational Particle Mechanics8(3), 575–588 (2020)

work page 2020

[63] [63]

Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A.N., Kaiser, L.u., Polosukhin, I.: At- tention is all you need. In: I. Guyon, U.V. Luxburg, S. Bengio, H. Wallach, R. Fergus, S. Vishwanathan, R. Garnett (eds.) Advances in Neural Information Pro- cessing Systems, vol. 30. Curran Associates, Inc. (2017)

work page 2017

[64] [64]

Journal of Hydraulic Research54(1), 1–26 (2016)

Violeau, D., Rogers, B.D.: Smoothed particle hydrody- namics (sph) for free-surface flows: past, present and fu- ture. Journal of Hydraulic Research54(1), 1–26 (2016)

work page 2016

[65] [65]

In: Proceedings of the 38th Conference on Neural Information Processing Systems (NeurIPS 2024) (2024)

Wu, K., Wu, E., Zou, J.: Clasheval: Quantifying the tug- of-war between an llm’s internal prior and external evi- dence. In: Proceedings of the 38th Conference on Neural Information Processing Systems (NeurIPS 2024) (2024)

work page 2024

[66] [66]

Retrieval-Augmented Generation for AI-Generated Content: A Survey

Zhao, P., Zhang, H., Yu, Q., Wang, Z., Geng, Y., Fu, F., Yang, L., Zhang, W., Jiang, J., Cui, B.: Retrieval- augmented generation for ai-generated content: A survey (2024). URL https://arxiv.org/abs/2402.19473

work page internal anchor Pith review Pith/arXiv arXiv 2024