Pith. sign in

REVIEW 5 major objections 6 minor 2 cited by

Edge Agentic AI Framework for Autonomous Network Optimisation in O-RAN

T0 review · 5 major / 6 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read An agentic AI framework inside the RIC claims zero outages during special-event 5G surges.

desk verdict Zero-outage claim looks like an artifact of the control rule; the architecture is a plausible integration but needs a proper baseline and released code. read the letter →

arxiv 2507.21696 v4 pith:ZKVPIEUB submitted 2025-07-29 eess.SP

Abdelaziz Salama , Zeinab Nezami , Mohammed M. H. Qazzaz , Maryam Hafeez , Syed Ali Raza Zaidi This is my paper
classification eess.SP
keywords O-RANagenticAInetworkoptimisationoutagepreventionLSTMtrafficpredictionpowercontrollargelanguagemodelsnear-real-timeRIC
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 claims that an agentic AI framework running inside the near-real-time RAN Intelligent Controller can prevent network outages entirely in high-stress 5G scenarios. It combines persona-based large language model agents, external event and weather data, an LSTM traffic predictor, and tiered transmission-power control. In a simulated urban 5G environment with special-event crowd surges, the framework reports 0% outage time, against 8.4% for a fixed-power baseline and 3.3% for a reactive LLM agent. The case for caring is that safe, autonomous network optimisation would let operators handle predictable demand spikes without dropping connections.

What carries the argument

The load-bearing mechanism is a closed-loop ReAct agent with a persona-based controller, a four-tier power-adjustment policy indexed by SINR thresholds ($\gamma_1=15$ dB, $\gamma_2=18$ dB, $\gamma_3=20$ dB), and a reward function $R_{\text{total}} = 10\Delta\text{SINR} + R_{\text{threshold}} + R_{\text{action}} - 2\Delta\text{power}$. An embedded two-layer LSTM xApp forecasts traffic and SINR trends, while search, event, and weather tools supply context; the reward aligns the agent's choices with outage prevention and energy efficiency.

What would settle it

Run the same agent in a live or high-fidelity simulated stadium scenario with independent mmWave propagation, interference coupling, and body blockage, and count the fraction of time SINR falls below 15 dB; any nonzero outage rate during the event window would falsify the 100% outage-prevention claim.

Watch

Extended reading notes

Core claim

The central claim is that proactive, context-aware agents, not faster reactive control, are what make AI safe enough for RAN operations. The agent forecasts demand from LSTM predictions and external event information, then chooses among four SINR-based power-control tiers before degradation occurs. On the paper's simulated event days, this yields zero outages, stable SINR, and a 7.1% action rate, while the fixed-power network suffers 8.4% outage time and a reactive LLM agent suffers 3.3%. The authors conclude that LLM agents with the right tools and reward alignment can be deployed in critical network infrastructure.

Load-bearing premise

The central claim rests on the assumption that raising transmission power and allocating extra resource blocks can actually hold links together during crowd surges and special events; if real mmWave environments are interference-limited or human blockage cannot be compensated by power, the zero-outage result would not transfer.

Editorial extensions

If this is right

  • If the result holds in real deployments, operators can run outage prevention as an autonomous closed loop in the near-RT RIC rather than relying on static configurations.
  • Zero outage time under special-event surges means the combination of LSTM prediction plus external event context is enough to act before degradation, not after.
  • The 7.1% action rate suggests the framework intervenes sparingly, so the safety benefit does not come from constant retuning.
  • The coefficient of variation below 0.15 across validation days indicates the framework's behaviour is stable across different event scenarios.
  • A practical corollary is that RIC-integrated LLM agents can satisfy near-real-time control-loop timescales.

Reading between the lines

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

  • The zero-outage figure is only as strong as the simulator's assumption that power increases and extra PRBs can restore link quality; in interference-limited or blockage-dominated mmWave settings, the same actions could raise interference without preventing outage.
  • Because the event APIs supply the demand spike information, a controlled ablation without external event data could separate how much of the gain comes from the LSTM forecast versus the agent's event awareness.
  • The tiered reward and tool set could transfer to other near-RT RIC controls, such as beam management or handover optimisation, without retraining the LLM.
  • A field trial during a large public event would be the natural test: count SINR samples below 15 dB across cells and compare with the 0% figure.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

5 major / 6 minor

Summary. The paper proposes an agentic AI framework for autonomous network optimisation in O-RAN, combining persona-based LLM agents, an LSTM traffic-prediction xApp, a tiered power-control policy, and a reward mechanism. The framework is integrated in a near-real-time RIC and evaluated in a simulated urban 5G environment. The central reported result is a zero-outage performance under high-demand special-event scenarios, compared to 8.4% outage for a fixed-power baseline and 3.3% for a reactive LLM baseline. The paper also claims 98% LSTM traffic-prediction accuracy and near-real-time responsiveness.

Significance. If the zero-outage result were properly supported, the framework would be a useful contribution to O-RAN intelligent control, particularly for proactive outage prevention through tool-augmented LLM agents and LSTM forecasting. The paper addresses a timely problem and presents a clear system architecture, with explicit equations for outage and reward. It also acknowledges limitations such as scalability validation. However, the central claim is currently not established: the evaluation metric is definitionally coupled to the control rule, and the reward function directly rewards the metric being reported. The paper's strengths are the relevant problem statement, the integration of predictive and agentic components, and the explicit statement of system assumptions; the weakness is that the experimental design does not isolate the contribution of those components.

major comments (5)
  1. [Section III-C, Step 3 and Section III-E, Eq. (2)] The zero-outage result is definitionally coupled to the control policy. Outage is defined in Eq. (2) as the fraction of samples with SINR below γ1 (15 dB), and the Tier 1 rule in Section III-C states that whenever SINR < γ1, the agent increases transmission power by up to 3 dB and allocates additional PRBs. Combined with the System Assumptions in Section III-C, which state that dynamic PRB allocation and adaptive transmission power control are sufficient to maintain link reliability, this means any policy that applies the Tier 1 action whenever SINR < γ1 will achieve 0% outage by construction. To support the paper's claim that the agentic, predictive, and multi-tool components are responsible for the result, the authors must report a non-agentic baseline that applies the same threshold-triggered action and compare its outage rate with that of the proposed framework.
  2. [Section III-C, Step 5, Eq. (1)] The reward function in Eq. (1) is not fully specified and is directly tied to the evaluation metric. The terms Rthreshold and Raction are never defined, and ΔSINR explicitly rewards SINR improvement, which is the same quantity used to compute outage rate in Eq. (2). This means the controller is optimized against the metric used to evaluate it, and the reported 0% outage may reflect reward shaping rather than genuine network-management capability. The authors should define all terms in Eq. (1), specify their numerical values, and include an ablation in which the reward does not contain a threshold-compliance term.
  3. [Section IV-A, Baseline Comparison] The comparison with baselines is incomplete. The fixed-power and reactive LLM baselines do not include the same tiered power-control rule, so the reported difference in outage rates (0% vs. 8.4% and 3.3%) may be entirely explained by the presence of a simple threshold controller rather than by the LSTM forecasts, persona-based reasoning, or multi-tool integration. The authors should add a baseline that uses the identical Tier 1–Tier 4 action rules without the agentic or predictive components, and a further baseline with the LSTM but without the LLM agent, to isolate the contribution of each component.
  4. [Section IV-A and Figure 3] The reported LSTM prediction accuracy is internally inconsistent. The text in Section IV-A states 98% prediction accuracy, while Figure 3 labels the event-day panel as 'Accuracy: 0.98%' and reports 'MAPE: 5.30%'. If accuracy is defined as (100 − MAPE), the MAPE of 5.30% corresponds to 94.7%, not 98%. The authors must clarify the definition of accuracy, correct the label, and report the actual prediction error metric at the time granularity used in the experiments. This matters because the proactive anomaly-detection claim depends on the reliability of the traffic forecasts.
  5. [Section IV-A and IV-B] The evaluation protocol is not sufficiently described for reproducibility. The paper reports a single high-traffic day in Section IV-A, then states in Section IV-B that evaluation was performed on 15 randomly selected event days over a 2-month period, but no details are given on how the simulator generates the network traces, what propagation and traffic models are used beyond the qualitative list, or how the agent actions affect the simulator state. Without the simulator code or a precise specification of the simulation environment, the central numerical claims (0%, 8.4%, 3.3%) cannot be independently checked. At minimum, the authors should provide the simulation parameters, the data-generation procedure, and the code or a detailed pseudo-code implementation.
minor comments (6)
  1. [Figure 3] The accuracy label in the event-day panel reads 'Accuracy: 0.98%', which should likely be '98%' or '0.98' depending on the intended definition; please correct this typo.
  2. [Section III-C] Four thresholds γ1–γ4 are listed, with γ4 = 25 dB, but only three threshold boundaries are needed for a four-tier decision rule; please clarify the role of γ4 or remove it.
  3. [Figure 2] The decision-tier diagram lists 'Tier 3' twice, once for optimization and once for efficiency; this likely should be 'Tier 4' in the second instance.
  4. [Section I, Index Terms] The index terms appear without spaces after commas in the PDF; please format them as separate terms.
  5. [Section III-A] The description of channel effects mentions 'comprehensive weather-induced variations' but does not state how these are combined with mmWave blockage and shadow fading; please specify the signal-to-interference-plus-noise computation model used in the simulator.
  6. [Section IV-B] The claim that the 'model maintained a coefficient of variation below 0.15 across all metrics' is not accompanied by confidence intervals or a statistical test; please provide variability measures for the reported outage and action rates.

Circularity Check

1 steps flagged · score 8.0 of 10

Headline zero-outage result is guaranteed by construction: the outage metric (Eq. 2) is defined by the same SINR<γ1 predicate that triggers Tier 1, and the system assumption states the Tier 1 actions are sufficient to restore link reliability.

  1. self definitional [Section III-C Step 3 and System Assumptions; Section III-E Eq. (2); Section IV-A]
    "Outage Rate = SINR < γ1 dB samples / Total samples × 100%. ... Tier 1 Critical (SINR < γ1 dB): Apply emergency power boost (increase transmission power by up to 3 dB) and allocate additional Physical Resource Blocks (PRBs) to maintain link stability and prevent imminent outage. ... The proposed system operates under the assumption that dynamic allocation of available PRBs and adaptive transmission power control are sufficient to maintain link reliability and service continuity, even under degraded network conditions."

    The evaluation metric (Eq. 2) counts a sample as an outage exactly when SINR < γ1. The Tier 1 control rule is triggered by exactly the same condition, SINR < γ1, and the System Assumptions state that the Tier 1 actions (PRB allocation plus adaptive transmission power control) are sufficient to maintain link reliability even under degraded conditions. Therefore, in the simulator, every potential outage sample triggers an action that the paper assumes is sufficient to restore reliability; a 0% outage rate follows from the trigger plus the assumption for any controller implementing Tier 1, independent of the LSTM forecasts, persona-based reasoning, or multi-tool integration. The reward in Eq.

full rationale

The central quantitative claim — 100% outage prevention under high-demand scenarios — reduces to the paper's own definitions and assumptions. Outage is defined as SINR below γ1 (Eq. 2); Tier 1 is triggered by SINR below γ1; and the System Assumptions declare that the Tier 1 actions are sufficient to maintain link reliability. Hence the zero-outage result is structurally guaranteed in the simulator and cannot validate the LSTM prediction, persona agents, or tool integration. The reward function's Rthreshold term further trains the controller toward the exact threshold used as the KPI. This is a genuine internal-validity circularity, not merely a missing comparison, because the result follows from the evaluation setup by construction. No load-bearing self-citation chain was found: the authors' prior works are cited for O-RAN power optimization and federated learning but are not the basis of the zero-outage claim. The traffic-prediction accuracy, action-rate, and stability results are not circular, but they do not rescue the headline outage claim. Score 8 reflects that the paper's principal validation metric is forced by definition rather than empirically established.

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

The central claims rest mainly on hand-set thresholds, an unspecified reward component, and an unverified simulator. No new physical entities are introduced. The free parameters are not fitted to external benchmarks, and the environment assumptions are untested outside the paper.

free parameters (3)
  • SINR thresholds gamma1..gamma4 = 15, 18, 20, 25 dB
    Hand-set in Section III-C to satisfy QoS requirements. The tiered actions and the outage metric are defined against these thresholds, so they directly shape the headline results.
  • Reward weights in Eq. (1) = 10 per dB SINR improvement, 2 per dB power penalty
    Chosen without a tuning procedure or sensitivity analysis. Rthreshold and Raction are mentioned but never defined, so the reward is not fully specified.
  • Power adjustment magnitudes = up to 3 dB increases, 1-3 dB decreases
    Action magnitudes in the tiered policy are fixed without justification or ablation, even though they determine how quickly outages can be prevented.
assumptions (3)
  • domain assumption Dynamic PRB and Tx power allocation are sufficient to maintain link reliability under degraded conditions
    Stated verbatim in Section III-C as a System Assumption. If false in a real deployment, outages cannot be prevented by the agent's actions regardless of prediction quality.
  • domain assumption The simulated environment reproduces real 5G mmWave propagation, traffic, and interference dynamics
    All quantitative claims are evaluated in the authors' simulator, described in Section IV. There is no validation against measurements from a real RAN or testbed.
  • domain assumption External event, weather, and social media APIs provide accurate and timely context
    Section III-B describes these tools as key enablers, but the results contain no ablation showing whether they were active or how their accuracy affects the outcome.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Edge Agentic AI Framework for Autonomous Network Optimisation in O-RAN." pith.science (2026). https://pith.science/paper/ZKVPIEUB

@misc{pith2026250721696,
  author       = {Pith},
  title        = {Pith review of: Edge Agentic AI Framework for Autonomous Network Optimisation in O-RAN},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZKVPIEUB}},
  note         = {Machine review of arXiv:2507.21696}
}
read the original abstract

The deployment of AI agents within legacy Radio Access Network (RAN) infrastructure poses significant safety and reliability challenges for future 6G networks. This paper presents a novel Edge AI framework for autonomous network optimisation in Open RAN environments, addressing these challenges through three core innovations: (1) a persona-based multi-tools architecture enabling distributed, context-aware decision-making; (2) proactive anomaly detection agent powered by traffic predictive tool; and (3) a safety, aligned reward mechanism that balances performance with operational stability. Integrated into the RAN Intelligent Controller (RIC), our framework leverages multimodal data fusion, including network KPIs, a traffic prediction model, and external information sources, to anticipate and respond to dynamic network conditions. Extensive evaluation using realistic 5G scenarios demonstrates that the edge framework achieves zero network outages under high-stress conditions, compared to 8.4% for traditional fixed-power networks and 3.3% for large language model (LLM) agent-based approaches, while maintaining near real-time responsiveness and consistent QoS. These results establish that, when equipped with the right tools and contextual awareness, AI agents can be safely and effectively deployed in critical network infrastructure, laying the framework for intelligent and autonomous 5G and beyond network operations.

Figures

Figures reproduced from arXiv: 2507.21696 by the authors.

Figure 1
Figure 1. An O-RAN architecture featuring an xApp powered by an agent-based [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. System Model Layout - AI Agent Framework for AI-RAN Network Optimisation. [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. LSTM model prediction results in wireless network traffic forecasting, [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Performance comparison of models for anomaly detection and outage [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: Radar chart comparing the performance of our proposed model against [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]

Discussion (0). Sign in to comment.

Forward citations

Cited by 2 Pith papers

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

  1. Cost and Accuracy of Long-Term Memory in Distributed Multi-Agent Systems Based on Large Language Models

    cs.IR 2026-01 reject novelty 5.0 of 10

    A two-framework testbed comparison claims mem0 is Pareto-optimal over Graphiti for distributed LLM agents because its lower cost is paired with accuracy that is not significantly different.

  2. Toward Edge General Intelligence with Agentic AI and Agentification: Concepts, Technologies, and Future Directions

    cs.NI 2025-08 conditional novelty 4.0 of 10

    A survey that organizes agentic AI for 6G edge networks into four pillars, compactness, efficiency, knowledge and reasoning, and migration, and illustrates them with prior case studies.

Reference graph

Works this paper leans on

29 extracted references · 20 canonical work pages · cited by 2 Pith papers

  1. [1]

    Transition technologies towards 6g networks

    Thiago R Raddo, Simon Rommel, Bruno Cimoli, Chris Vagionas, Diego Perez-Galacho, Evangelos Pikasis, Evangelos Grivas, Konstantinos Nton- tin, Michael Katsikis, Dimitrios Kritharidis, et al. Transition technologies towards 6g networks. EURASIP Journal on Wireless Communications and Networking, 2021(1):100, 2021

    work page 2021

  2. [2]

    Oran-map: A hybrid approach to mobility-aware power optimisation in open radio access networks (oran)

    Mohammed MH Qazzaz, Abdelaziz Salama, Maryam Hafeez, and Syed AR Zaidi. Oran-map: A hybrid approach to mobility-aware power optimisation in open radio access networks (oran). In IEEE Infocom. Proceedings. IEEE, 2025

    work page 2025

  3. [3]

    Autogen: Enabling next-gen llm applications via multi-agent conversation framework

    Qingyun Wu, Gagan Bansal, Jieyu Zhang, Yiran Wu, Shaokun Zhang, Erkang Zhu, Beibin Li, Li Jiang, Xiaoyun Zhang, and Chi Wang. Autogen: Enabling next-gen llm applications via multi-agent conversation framework. arXiv preprint arXiv:2308.08155 , 2023

    arXiv 2023

  4. [4]

    Llm-driven agentic ai approach to enhanced o-ran resilience in next-generation networks, May 2025

    Xingqi Wu, Juntao Chen, and Li Qian. Llm-driven agentic ai approach to enhanced o-ran resilience in next-generation networks, May 2025. Authorea Preprints. Accessed: June 2, 2025

    work page 2025

  5. [5]

    Personagym: Evaluating persona agents and llms

    Vinay Samuel, Vladislav Lialin, Ratish Shah, Chen Li, Tanya Zhao, Hannah Rashkin, Mo Yu, Aston Dcunha, Shreyas Ghandi, Yejin Peggy Wang, Karthik Narasimhan, and Mohit Iyyer. Personagym: Evaluating persona agents and llms. arXiv preprint arXiv:2407.18416 , Jul 2024

    arXiv 2024

  6. [6]

    Chatharuhi: Reviving anime character in reality via large language model

    Cheng Li, Ziang Leng, Chenxi Yan, Junyi Shen, Hao Wang, Weishi Mi, Yaying Fei, Xiaoyang Feng, Song Yan, HaoSheng Wang, et al. Chatharuhi: Reviving anime character in reality via large language model. arXiv preprint arXiv:2308.09597 , 2023

    arXiv 2023

  7. [7]

    Better zero-shot reasoning with role-play prompting

    Aobo Kong, Shiwan Zhao, Hao Chen, Qicheng Li, Yong Qin, Ruiqi Sun, Xin Zhou, Enzhi Wang, and Xiaohang Dong. Better zero-shot reasoning with role-play prompting. arXiv preprint arXiv:2308.07702 , 2023

    arXiv 2023

  8. [8]

    Softbank corp

    SoftBank. Softbank corp. and red hat develop solution to optimize power consumption in ai-ran data centers. Press Release, Mar 2025

    work page 2025

Show all 29 references
  1. [9]

    Nvidia ai aerial launches to optimize wireless networks, deliver new generative ai experiences on one platform

    Ronnie Vasishta. Nvidia ai aerial launches to optimize wireless networks, deliver new generative ai experiences on one platform. NVIDIA Blog, Sep 2024

    2024

  2. [10]

    Evaluating generative ai for telecom

    Ericsson. Evaluating generative ai for telecom. Ericsson, May 2025

    2025

  3. [11]

    Deploy ai-ran at cell sites with nvidia arc-compact

    NVIDIA. Deploy ai-ran at cell sites with nvidia arc-compact. NVIDIA Developer, May 2025

    2025

  4. [12]

    Joint admission control and resource provisioning for urllc traffic in o-ran: A constrained multi- agent reinforcement learning approach, May 2025

    Xingqi Wu, Junaid Farooq, and Juntao Chen. Joint admission control and resource provisioning for urllc traffic in o-ran: A constrained multi- agent reinforcement learning approach, May 2025. Authorea Preprints. Accessed: June 2, 2025

    2025

  5. [13]

    Explainable ai in 6g o-ran: A tutorial and survey on architecture, use cases, challenges, and future research

    Bouziane Brik, Hatim Chergui, Lanfranco Zanzi, Francesco Devoti, Adlen Ksentini, Muhammad Shuaib Siddiqui, Xavier Costa-P `erez, and Christos Verikoukis. Explainable ai in 6g o-ran: A tutorial and survey on architecture, use cases, challenges, and future research. IEEE Communi...

    2024

  6. [14]

    Fedora: Resource allocation for federated learning in oran using radio intelligent controllers

    Abdelaziz Salama, Mohammed MH Qazzaz, Syed Danial Ali Shah, Maryam Hafeez, and Syed Ali Zaidi. Fedora: Resource allocation for federated learning in oran using radio intelligent controllers. arXiv preprint arXiv:2505.19211, 2025

    2025 arXiv

  7. [15]

    Alympics: Llm agents meet game theory

    Shaoguang Mao, Yao Fu, Chuan Zhang, Yilun Shen, Yeqi Sun, Yu Wu, and Kun Li. Alympics: Llm agents meet game theory. In Proceedings of the 31st International Conference on Computational Linguistics (COL- ING 2025), Abu Dhabi, UAE, Jan 2025. Forthcoming

    2025

  8. [16]

    Teleqna: A benchmark dataset to assess large language models telecommunications knowledge

    Ali Maatouk, Fadhel Ayed, Nicola Piovesan, Antonio De Domenico, Merouane Debbah, and Zhi-Quan Luo. Teleqna: A benchmark dataset to assess large language models telecommunications knowledge. arXiv preprint arXiv:2310.15051, 2023

    arXiv

  9. [17]

    Advanced architectures integrated with agentic ai for next-generation wireless networks

    Kapal Dev, Sunder Ali Khowaja, Keshav Singh, Engin Zeydan, and Merouane Debbah. Advanced architectures integrated with agentic ai for next-generation wireless networks. arXiv preprint arXiv:2502.01089, 2025

    2025

  10. [18]

    The power of large language models for wireless communication system development: A case study on fpga platforms

    Yuyang Du, Hongyu Deng, Soung Chang Liew, Kexin Chen, Yulin Shao, and He Chen. The power of large language models for wireless communication system development: A case study on fpga platforms. arXiv preprint arXiv:2307.07319 , 2023

    2023 arXiv

  11. [19]

    Llm-based policy generation for intent-based management of applications

    Kristina Dzeparoska, Jieyu Lin, Ali Tizghadam, and Alberto Leon- Garcia. Llm-based policy generation for intent-based management of applications. In 2023 19th International Conference on Network and Service Management (CNSM) , pages 1–7. IEEE, 2023

    2023

  12. [20]

    What do llms need to synthesize correct router configurations? In Proceedings of the 22nd ACM Workshop on Hot Topics in Networks , pages 189–195, 2023

    Rajdeep Mondal, Alan Tang, Ryan Beckett, Todd Millstein, and George Varghese. What do llms need to synthesize correct router configurations? In Proceedings of the 22nd ACM Workshop on Hot Topics in Networks , pages 189–195, 2023

    2023

  13. [21]

    Toward reproducing network research results using large language models

    Qiao Xiang, Yuling Lin, Mingjun Fang, Bang Huang, Siyong Huang, Ridi Wen, Franck Le, Linghe Kong, and Jiwu Shu. Toward reproducing network research results using large language models. In Proceedings of the 22nd ACM Workshop on Hot Topics in Networks , 2023

    2023

  14. [22]

    Wireless multi-agent generative ai: From connected intelligence to collective intelligence

    Hang Zou, Qiyang Zhao, Lina Bariah, Mehdi Bennis, and Merouane Debbah. Wireless multi-agent generative ai: From connected intelligence to collective intelligence. arXiv preprint arXiv:2307.02757 , 2023

    2023 arXiv

  15. [23]

    Edgefm: Leveraging foundation model for open-set learning on the edge

    Bufang Yang, Lixing He, Neiwen Ling, Zhenyu Yan, Guoliang Xing, Xian Shuai, Xiaozhe Ren, and Xin Jiang. Edgefm: Leveraging foundation model for open-set learning on the edge. In Proceedings of the 21st ACM Conference on Embedded Networked Sensor Systems , pages 111–124, 2023

    2023

  16. [24]

    Et-bert: A contextualized datagram representation with pre-training transformers for encrypted traffic classification

    Xinjie Lin, Gang Xiong, Gaopeng Gou, Zhen Li, Junzheng Shi, and Jing Yu. Et-bert: A contextualized datagram representation with pre-training transformers for encrypted traffic classification. In Proceedings of the ACM Web Conference 2022 , pages 633–642, 2022

    2022

  17. [25]

    Reward design with language models

    Minae Kwon, Sang Michael Xie, Kalesha Bullard, and Dorsa Sadigh. Reward design with language models. arXiv preprint arXiv:2303.00001, 2023

    2023 arXiv

  18. [26]

    Diagnosing infeasible optimization problems using large language models

    Hao Chen, Gonzalo E Constante-Flores, and Can Li. Diagnosing infeasible optimization problems using large language models. INFOR: Information Systems and Operational Research , 62(4):573–587, 2024

    2024

  19. [27]

    Large language models empowered autonomous edge ai for connected intelligence

    Yifei Shen, Jiawei Shao, Xinjie Zhang, Zehong Lin, Hao Pan, Dongsheng Li, Jun Zhang, and Khaled B Letaief. Large language models empowered autonomous edge ai for connected intelligence. IEEE Communications Magazine, 2024

    2024

  20. [28]

    Parameter-efficient fine-tuning of large-scale pre-trained language models

    Ning Ding, Yujia Qin, Guang Yang, Fuchao Wei, Zonghan Yang, Yusheng Su, Shengding Hu, Yulin Chen, Chi-Min Chan, Weize Chen, et al. Parameter-efficient fine-tuning of large-scale pre-trained language models. Nature Machine Intelligence , 5(3):220–235, 2023

    2023

  21. [29]

    Mobile- llama: Instruction fine-tuning open-source llm for network analysis in 5g networks

    Khen Bo Kan, Hyunsu Mun, Guohong Cao, and Youngseok Lee. Mobile- llama: Instruction fine-tuning open-source llm for network analysis in 5g networks. IEEE Network, 2024

    2024

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.