pith. sign in

arxiv: 2606.06239 · v1 · pith:JX2GFTBCnew · submitted 2026-06-04 · 📡 eess.SP

Foundation Models for Wireless Communications: From PHY Intelligence to Network Autonomy

Le Liang , Jiajia Guo , Jun Zhang , Chan-Byoung Chae , Lu Lu , Shugong Xu , Octavia A. Dobre , Shi Jin
show 1 more author
Geoffrey Ye Li
This is my paper

Pith reviewed 2026-06-28 00:10 UTC · model grok-4.3

classification 📡 eess.SP
keywords foundation models6G networksphysical layerwireless resource managementnetwork autonomyintegrated sensing and communicationsagentic AIsemantic communications
0
0 comments X

The pith

Foundation models adapt pre-trained versions, build on wireless data, or act as agents to manage 6G networks from physical layer to full autonomy.

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

The paper surveys how large-scale foundation models, trained on massive data for general feature extraction, can be applied to wireless communications to handle the complexity of 6G. It describes three progressive approaches: adapting existing models to wireless tasks, constructing models directly from wireless data to capture physical characteristics, and using agentic models for autonomous reasoning and orchestration. These steps aim to shift network optimization and management toward more flexible, data-driven intelligence across physical-layer processing and system-level control. A reader would care because the work positions foundation models as a way to move beyond narrow AI solutions toward general-purpose adaptability in future wireless systems.

Core claim

Foundation models reshape physical-layer processing and wireless resource management across three paradigms: adaptation of off-the-shelf pre-trained models to wireless tasks, wireless-native models built from scratch on wireless data to bridge modality gaps and capture universal physical characteristics, and agentic models that turn static processing into autonomous, reasoning-driven network orchestration, with further effects on ISAC, MIMO, semantic communications, and system-level autonomy.

What carries the argument

Three progressive paradigms of foundation model application (off-the-shelf adaptation, wireless-native construction from wireless data, and agentic orchestration for autonomy).

If this is right

  • Adaptation of pre-trained models enables zero-shot or few-shot solutions for wireless tasks without full retraining.
  • Wireless-native models capture domain-specific physical traits and reduce gaps when transferring across wireless modalities.
  • Agentic models support reasoning-based decisions that move network management from reactive to proactive and autonomous.
  • Applications extend to integrated sensing and communications, advanced MIMO, semantic communications, and overall network autonomy.
  • The approach charts a path to fully intelligent and adaptive wireless networks for 6G.

Where Pith is reading between the lines

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

  • If universal wireless characteristics exist, the same foundation model could handle tasks across different frequencies or standards with minimal changes.
  • Agentic models might enable networks that continuously self-optimize based on live conditions without human intervention.
  • Training wireless-native models would require large, diverse wireless datasets, potentially shifting research focus toward data collection methods.
  • This framework could reduce reliance on hand-crafted algorithms for each new wireless problem.

Load-bearing premise

Wireless data contains universal physical characteristics that foundation models can learn to overcome differences between data types.

What would settle it

A controlled test showing that models trained from scratch on wireless data perform no better than generic pre-trained models or task-specific networks on downstream wireless tasks such as channel estimation or resource allocation.

Figures

Figures reproduced from arXiv: 2606.06239 by Chan-Byoung Chae, Geoffrey Ye Li, Jiajia Guo, Jun Zhang, Le Liang, Lu Lu, Octavia A. Dobre, Shi Jin, Shugong Xu.

Figure 1
Figure 1. Figure 1: Architectural illustration of encoder-decoder and decoder-only Trans [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: The architecture of BP-LLM [6], which employs explicit modality alignment via patch reprogramming to map historical sequences of beam indices [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Framework of developing wireless-native foundation models for [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: NMSE versus feedback overhead for CSI feedback in unseen [PITH_FULL_IMAGE:figures/full_fig_p008_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: The architecture of agentic AI in physical-layer processing, adapted from [49]. The framework is composed of foundational capabilities, the agentic [PITH_FULL_IMAGE:figures/full_fig_p010_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Paradigms of foundation model adaptation for wireless resource [PITH_FULL_IMAGE:figures/full_fig_p011_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Max-sum-rate performance of adapting a pre-trained LLM to wireless [PITH_FULL_IMAGE:figures/full_fig_p012_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Coexistence between LLM-driven APs and legacy CSMA/CA APs in [PITH_FULL_IMAGE:figures/full_fig_p014_8.png] view at source ↗
read the original abstract

6G networks will introduce unprecedented complexity, which calls for a paradigm shift in network optimization and management. Artificial intelligence (AI)-based solutions, especially those enabled by the recently developed foundation models, have been recognized as promising candidates. Foundation models are large-scale AI models with general-purpose feature extraction capabilities, and once trained on massive amounts of data, they can be adapted to solve a wide range of downstream tasks, either in a zero-shot manner or with few-shot fine-tuning. This article provides a comprehensive overview of how foundation models are reshaping physical-layer processing and wireless resource management across three progressive paradigms. First, we examine the adaptation of off-the-shelf pre-trained foundation models to various wireless tasks. Second, we explore wireless-native foundation models, built from scratch on wireless data to bridge cross-domain modality gaps and capture universal wireless-domain physical characteristics. Third, we highlight agentic foundation models, which elevate static data processing into autonomous, reasoning-driven network orchestration. Furthermore, we discuss the impact of applying foundation models to emerging 6G frontiers, including integrated sensing and communications (ISAC), new multiple-input multiple-output (MIMO) architectures, semantic communications, and system-level network autonomy. Finally, we identify critical open challenges and opportunities, charting a promising path toward fully intelligent and adaptive wireless networks.

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.

Referee Report

0 major / 1 minor

Summary. The manuscript is a survey providing a comprehensive overview of foundation models for wireless communications. It organizes the topic into three progressive paradigms—adaptation of off-the-shelf pre-trained models to wireless tasks, wireless-native foundation models constructed from scratch on wireless data, and agentic foundation models for reasoning-driven autonomous orchestration—while mapping these to 6G use cases including ISAC, new MIMO architectures, semantic communications, and system-level network autonomy, and outlining open challenges and opportunities.

Significance. If the taxonomy and literature mapping are accurate and balanced, the paper would offer a useful organizational framework for an emerging area, helping researchers navigate the shift toward general-purpose AI in wireless systems. The explicit treatment of the wireless-native paradigm as a prospective research direction (rather than a demonstrated result) is a strength that avoids overclaiming. No new theorems, datasets, or experiments are advanced, which is appropriate for a survey but means the work's value rests on the quality of its synthesis and coverage.

minor comments (1)
  1. [Abstract] Abstract: the phrase 'capture universal wireless-domain physical characteristics' is used to motivate the wireless-native paradigm but is not accompanied by concrete examples or citations in the provided abstract; expanding this briefly in the introduction or §2 would improve clarity without altering the survey's descriptive claims.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for the positive summary, significance assessment, and recommendation of minor revision. The report does not list any specific major comments under the MAJOR COMMENTS section, so we have no individual points to address.

Circularity Check

0 steps flagged

No significant circularity; survey with no derivations or fitted quantities

full rationale

The manuscript is a literature survey categorizing existing and prospective uses of foundation models into three paradigms (off-the-shelf adaptation, wireless-native construction, agentic orchestration) and mapping them to 6G use cases. It advances no new theorems, datasets, or experimental results. No equations appear, and the wireless-native paradigm is explicitly framed as an open research direction rather than a demonstrated result whose validity depends on the universality assumption. All claims are descriptive and externally supported by cited literature; the central discussion does not reduce to self-referential definitions or self-citation chains.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

This is a survey paper; the abstract introduces no new free parameters, mathematical axioms, or invented physical entities. All content draws from prior foundation-model and wireless-AI literature.

pith-pipeline@v0.9.1-grok · 5785 in / 1219 out tokens · 23810 ms · 2026-06-28T00:10:29.759677+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 1 Pith paper

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

  1. Physics-Informed Path-Parametric Learning for Efficient and Lightweight CSI Feedback

    eess.SP 2026-06 unverdicted novelty 6.0

    HS-PINNnet integrates multipath physics into a hierarchical neural network for efficient CSI feedback, outperforming prior methods with major reductions in computational cost.

Reference graph

Works this paper leans on

114 extracted references · 5 linked inside Pith · cited by 1 Pith paper

  1. [1]

    BERT: Pre-training of deep bidirectional transformers for language understanding,

    J. Devlin, M.-W. Chang, K. Lee, and K. Toutanova, “BERT: Pre-training of deep bidirectional transformers for language understanding,” inProc. Conf. N. Amer. Chapter Assoc. Com- put. Linguist. Hum. Lang. Technol. (NAACL-HLT), 2019, pp. 4171–4186

    2019

  2. [2]

    Language models are unsupervised multitask learners,

    A. Radford, J. Wu, R. Child, D. Luan, D. Amodei, and I. Sutskever, “Language models are unsupervised multitask learners,”OpenAI blog, vol. 1, no. 8, pp. 1–9, Feb. 2019

    2019

  3. [3]

    An image is worth 16x16 words: Transformers for image recognition at scale,

    A. Dosovitskiyet al., “An image is worth 16x16 words: Transformers for image recognition at scale,” inProc. Int. Conf. Learn. Represent. (ICLR), 2021, pp. 1–21

    2021

  4. [4]

    Large language models for wireless communi- cations: From adaptation to autonomy,

    L. Lianget al., “Large language models for wireless communi- cations: From adaptation to autonomy,”IEEE Commun. Mag., vol. 64, no. 5, pp. 140–147, May 2026. 20

    2026

  5. [5]

    LLM4CP: Adapting large language models for channel prediction,

    B. Liu, X. Liu, S. Gao, X. Cheng, and L. Yang, “LLM4CP: Adapting large language models for channel prediction,”J. Commun. Inf. Netw., vol. 9, no. 2, pp. 113–125, Jun. 2024

    2024

  6. [6]

    Beam prediction based on large language models,

    Y . Sheng, K. Huang, L. Liang, P. Liu, S. Jin, and G. Y . Li, “Beam prediction based on large language models,”IEEE Wireless Commun. Lett., vol. 14, no. 5, pp. 1406–1410, May 2025

    2025

  7. [7]

    WiFo: Wireless foundation model for channel prediction,

    B. Liu, S. Gao, X. Liu, X. Cheng, and L. Yang, “WiFo: Wireless foundation model for channel prediction,”Sci. China Inf. Sci., vol. 68, no. 6, p. 162302, Jul. 2025

    2025

  8. [8]

    A wireless foundation model for multi-task prediction,

    Y . Shenget al., “A wireless foundation model for multi-task prediction,”arXiv preprint arXiv:2507.05938, Jul. 2025

    arXiv 2025

  9. [9]

    WirelessLLM: Empowering large language models towards wireless intelligence,

    J. Shaoet al., “WirelessLLM: Empowering large language models towards wireless intelligence,”J. Commun. Inf. Netw., vol. 9, no. 2, pp. 99–112, June 2024

    2024

  10. [10]

    Learning multi-access point coordination in agentic AI Wi-Fi with large language models,

    Y . Fanet al., “Learning multi-access point coordination in agentic AI Wi-Fi with large language models,” inProc. IEEE Int. Conf. Commun. (ICC), 2026, pp. 1–6

    2026

  11. [11]

    Language models are few-shot learners,

    T. Brownet al., “Language models are few-shot learners,” in Proc. Adv. Neural Inf. Process. Syst. (NeurIPS), vol. 33, 2020, pp. 1877–1901

    2020

  12. [12]

    A survey of state of the art large vision language models: Benchmark evaluations and challenges,

    Z. Li, X. Wu, H. Du, F. Liu, H. Nghiem, and G. Shi, “A survey of state of the art large vision language models: Benchmark evaluations and challenges,” inProc. Comput. Vis. Pattern Recog. Conf., 2025, pp. 1587–1606

    2025

  13. [13]

    On the opportunities and risks of foundation models,

    R. Bommasaniet al., “On the opportunities and risks of foundation models,”arXiv preprint arXiv:2108.07258, Jul. 2022

    Pith/arXiv arXiv 2022

  14. [14]

    Attention is all you need,

    A. Vaswaniet al., “Attention is all you need,” inProc. Adv. Neural Inf. Process. Syst. (NeurIPS), vol. 30, 2017, pp. 5998– 6008

    2017

  15. [15]

    Learning transferable visual models from natural language supervision,

    A. Radfordet al., “Learning transferable visual models from natural language supervision,” inProc. Int. Conf. Mach. Learn- ing (ICML), 2021, pp. 8748–8763

    2021

  16. [16]

    LLaV A-OneVision: Easy visual task transfer,

    B. Liet al., “LLaV A-OneVision: Easy visual task transfer,” Trans. Mach. Learn. Res., pp. 1–44, Feb. 2025

    2025

  17. [17]

    Representation learning with contrastive predictive coding,

    A. v. d. Oord, Y . Li, and O. Vinyals, “Representation learning with contrastive predictive coding,”arXiv preprint arXiv:1807.03748, 2018

    Pith/arXiv arXiv 2018

  18. [18]

    Chain-of-thought prompting elicits reasoning in large language models,

    J. Weiet al., “Chain-of-thought prompting elicits reasoning in large language models,” inProc. Adv. Neural Inf. Process. Syst. (NeurIPS), vol. 35, 2022, pp. 24 824–24 837

    2022

  19. [19]

    Retrieval-augmented generation for knowledge-intensive NLP tasks,

    P. Lewiset al., “Retrieval-augmented generation for knowledge-intensive NLP tasks,” inProc. Adv. Neural Inf. Process. Syst. (NeurIPS), vol. 33, 2020, pp. 9459–9474

    2020

  20. [20]

    ReAct: Synergizing reasoning and acting in language models,

    S. Yaoet al., “ReAct: Synergizing reasoning and acting in language models,” inProc. Int. Conf. Learn. Represent. (ICLR), 2023

    2023

  21. [21]

    LLM4XCE: Large language models for extremely large-scale massive MIMO channel estimation,

    R. Li, S. Li, and P. Dong, “LLM4XCE: Large language models for extremely large-scale massive MIMO channel estimation,” inProc. 17th Int. Conf. Wireless Commun. Signal Process. (WCSP), 2025, pp. 1–6

    2025

  22. [22]

    LVM4CSI: Enabling direct application of pre-trained large vision models for wireless channel tasks,

    J. Guo, P. Jiang, C.-K. Wen, S. Jin, and J. Zhang, “LVM4CSI: Enabling direct application of pre-trained large vision models for wireless channel tasks,”arXiv preprint arXiv:2507.05121, Jul. 2025

    arXiv 2025

  23. [23]

    Semantic pilot design for data- aided channel estimation using a large language model,

    S. Park and H. J. Yang, “Semantic pilot design for data- aided channel estimation using a large language model,”arXiv preprint arXiv:2602.04126, Feb. 2026

    Pith/arXiv arXiv 2026

  24. [24]

    Exploring the potential of large language models for massive MIMO CSI feedback,

    Y . Cui, J. Guo, C.-K. Wen, S. Jin, and E. Tong, “Exploring the potential of large language models for massive MIMO CSI feedback,” inProc. IEEE Global Commun. Conf. (GLOBE- COM), 2025, pp. 3511–3516

    2025

  25. [25]

    OLAMCF: Offline large AI models enhanced CSI feedback in FDD massive MIMO systems,

    J. Zhuang, Y . Wang, H. Hou, Y . Han, W. Wang, and S. Jin, “OLAMCF: Offline large AI models enhanced CSI feedback in FDD massive MIMO systems,” inProc. IEEE Global Commun. Conf. (GLOBECOM), 2025, pp. 3517–3522

    2025

  26. [26]

    Bridging the modality gap: Enhancing channel prediction with semantically aligned LLMs and knowledge distillation,

    Z. Li, Q. Yang, Z. Xiong, Z. Shi, and T. Q. S. Quek, “Bridging the modality gap: Enhancing channel prediction with semantically aligned LLMs and knowledge distillation,” IEEE J. Sel. Areas Commun., vol. 44, pp. 3382–3396, Dec. 2025

    2025

  27. [27]

    SCA-LLM: Spectral-attentive LLM-based wire- less world modeling for agentic communications,

    K. Heet al., “SCA-LLM: Spectral-attentive LLM-based wire- less world modeling for agentic communications,”arXiv preprint arXiv:2509.08139, Sep. 2025

    arXiv 2025

  28. [28]

    Sensing- assisted channel prediction in complex wireless environments: An LLM-based approach,

    J. He, Z. Ren, J. Yao, H. Hu, T. X. Han, and J. Xu, “Sensing- assisted channel prediction in complex wireless environments: An LLM-based approach,”IEEE Wireless Commun. Lett., vol. 14, no. 12, pp. 3857–3861, Dec. 2025

    2025

  29. [29]

    LLM- MM: End-to-end robust multimodal beam prediction for 6G V2X networks via MoE-LoRA adaptation,

    J. Lei, X. Li, C. Wu, Q. Fu, J. Liu, and N. Kato, “LLM- MM: End-to-end robust multimodal beam prediction for 6G V2X networks via MoE-LoRA adaptation,”IEEE J. Sel. Areas Commun., vol. 44, pp. 2964–2977, Dec. 2025

    2025

  30. [30]

    Large-model AI for near field beam prediction: A CNN- GPT2 framework for 6G XL-MIMO,

    W. Liu, C. Pan, H. Ren, W. Zhang, C.-X. Wang, and J. Wang, “Large-model AI for near field beam prediction: A CNN- GPT2 framework for 6G XL-MIMO,”IEEE Trans. Wireless Commun., vol. 25, pp. 15 149–15 165, Apr. 2026

    2026

  31. [31]

    BeamVLM for low-altitude economy: Generative beam prediction via vision-language models,

    C. Kou, C. You, M. Wu, D. Wen, Z. Zhang, and C. Xing, “BeamVLM for low-altitude economy: Generative beam prediction via vision-language models,”arXiv preprint arXiv:2602.19929, Feb. 2026

    arXiv 2026

  32. [32]

    Large sequence model for MIMO equalization in fully decoupled radio access network,

    K. Yu, H. Zhou, Y . Xu, Z. Liu, H. Du, and X. Shen, “Large sequence model for MIMO equalization in fully decoupled radio access network,”IEEE Open J. Commun. Soc., vol. 6, pp. 4491–4504, May 2025

    2025

  33. [33]

    Decision feedback in-context symbol detection over block-fading channels,

    L. Fan, J. Yang, C. Shen, and C. L. Brown, “Decision feedback in-context symbol detection over block-fading channels,” in Proc. IEEE Int. Conf. Commun. (ICC), 2025, pp. 3785–3790

    2025

  34. [34]

    Turbo- ICL: In-context learning-based turbo equalization,

    Z. Song, M. Zecchin, B. Rajendran, and O. Simeone, “Turbo- ICL: In-context learning-based turbo equalization,”arXiv preprint arXiv:2505.06175, May 2025

    arXiv 2025

  35. [35]

    Large language model enabled multi- task physical layer network,

    T. Zheng and L. Dai, “Large language model enabled multi- task physical layer network,”IEEE Trans. Commun., vol. 74, pp. 307–321, 2026

    2026

  36. [36]

    SignalLLM: A general purpose LLM agent framework for automated signal processing,

    J. Ke, Q. Hu, S. Yuan, Y . Xu, and J. Yang, “SignalLLM: A general purpose LLM agent framework for automated signal processing,”arXiv preprint arXiv:2509.17197, Sep. 2025

    arXiv 2025

  37. [37]

    Beam prediction based on multimodal large language models,

    T. Mao, L. Liang, J. Yang, X. Li, S. Jin, and G. Y . Li, “Beam prediction based on multimodal large language models,”arXiv preprint arXiv:2603.15093, 2026

    arXiv 2026

  38. [38]

    Reducing pilots in channel estimation with predictive foun- dation models,

    X. Zhou, L. Liang, H. Ye, J. Zhang, C.-K. Wen, and S. Jin, “Reducing pilots in channel estimation with predictive foun- dation models,”arXiv preprint arXiv:2512.15562, Dec. 2025

    arXiv 2025

  39. [39]

    Generative diffusion models for high dimensional channel estimation,

    X. Zhou, L. Liang, J. Zhang, P. Jiang, Y . Li, and S. Jin, “Generative diffusion models for high dimensional channel estimation,”IEEE Trans. Wireless Commun., vol. 24, no. 7, pp. 5840–5854, Jul. 2025

    2025

  40. [40]

    Prompt-enabled large AI models for CSI feedback,

    J. Guo, Y . Cui, C.-K. Wen, and S. Jin, “Prompt-enabled large AI models for CSI feedback,”IEEE J. Sel. Areas Commun., vol. 44, pp. 2654–2668, 2026

    2026

  41. [41]

    WiFo-CF: Wireless foundation model for CSI feedback,

    X. Liu, S. Gao, B. Liu, X. Cheng, and L. Yang, “WiFo-CF: Wireless foundation model for CSI feedback,”IEEE Trans. Wireless Commun., vol. 25, pp. 15 039–15 053, Apr. 2026

    2026

  42. [42]

    Signal compression for wireless communication and sensing: A general approach utilizing pretrained wireless foundation models,

    L. Jing, T. Yang, H. Zhang, Y . Shi, C. Zhang, and B. Zhang, “Signal compression for wireless communication and sensing: A general approach utilizing pretrained wireless foundation models,”IEEE Trans. Mobile Comput., 2026, early Access

    2026

  43. [43]

    Self-supervised radio pre-training: Toward foundational models for spectrogram learning,

    A. Aboulfotouh, A. Eshaghbeigi, D. Karslidis, and H. Abou- Zeid, “Self-supervised radio pre-training: Toward foundational models for spectrogram learning,” inProc. IEEE Global Com- mun. Conf. (GLOBECOM), 2024, pp. 1269–1274

    2024

  44. [44]

    SpectrumFM: A foundation model for intel- ligent spectrum management,

    F. Zhouet al., “SpectrumFM: A foundation model for intel- ligent spectrum management,”IEEE J. Sel. Areas Commun., vol. 44, pp. 4471–4488, Apr. 2026

    2026

  45. [45]

    Self-supervised and in- variant representations for wireless localization,

    A. Salihu, M. Rupp, and S. Schwarz, “Self-supervised and in- variant representations for wireless localization,”IEEE Trans. Wireless Commun., vol. 23, no. 8, pp. 8281–8296, 2024

    2024

  46. [46]

    WiFo-MUD: 21 Wireless foundation model for heterogeneous multi-user de- modulator,

    Z. Yang, S. Gao, X. Cai, X. Cheng, and L. Yang, “WiFo-MUD: 21 Wireless foundation model for heterogeneous multi-user de- modulator,”arXiv preprint arXiv:2601.00612, Jan. 2026

    arXiv 2026

  47. [47]

    A foundation model for massive MIMO precoding with an adaptive per-user rate-power tradeoff,

    J. Emery, A. H. Karkan, J.-F. Frigon, and F. Leduc-Primeau, “A foundation model for massive MIMO precoding with an adaptive per-user rate-power tradeoff,” inProc. IEEE Int. Symp. Pers., Indoor, Mobile Radio Commun. (PIMRC), 2025, pp. 1–6

    2025

  48. [48]

    Filter-and-attend: Wireless channel foundation model with noise-plus-interference suppression structure,

    Y . Wang, L. Sun, T. Yang, Y . Shi, M. Elkashlan, and X. Tang, “Filter-and-attend: Wireless channel foundation model with noise-plus-interference suppression structure,”arXiv preprint arXiv:2509.15993, Jan. 2026

    arXiv 2026

  49. [49]

    Wirelessagent: Large language model agents for intelligent wireless networks,

    J. Tong, J. Shao, Q. Wu, W. Guo, Z. Li, Z. Lin, and J. Zhang, “Wirelessagent: Large language model agents for intelligent wireless networks,”China Commun., vol. 23, no. 3, pp. 265– 285, Mar. 2024

    2024

  50. [50]

    Large wireless model (LWM): A foundation model for wireless channels,

    S. Alikhani, G. Charan, and A. Alkhateeb, “Large wireless model (LWM): A foundation model for wireless channels,” arXiv preprint arXiv:2411.08872, Nov. 2024

    arXiv 2024

  51. [51]

    WirelessGPT: A generative pre-trained multi- task learning framework for wireless communication,

    T. Yanget al., “WirelessGPT: A generative pre-trained multi- task learning framework for wireless communication,”IEEE Netw., vol. 39, no. 5, pp. 158–165, Sep. 2025

    2025

  52. [52]

    ICWLM: A multi-task wireless large model via in-context learning,

    Y . Wen, X. Chen, M. Zhang, Z. Yang, C. Huang, and Z. Zhang, “ICWLM: A multi-task wireless large model via in-context learning,”IEEE Trans. Commun., vol. 74, pp. 3646–3658, 2026

    2026

  53. [53]

    LLM4WM: Adapting LLM for wireless multi-tasking,

    X. Liu, S. Gao, B. Liu, X. Cheng, and L. Yang, “LLM4WM: Adapting LLM for wireless multi-tasking,”IEEE Trans. Mach. Learn. Commun. Netw., vol. 3, pp. 835–847, Jul. 2025

    2025

  54. [54]

    Large and small model collaboration for air interface,

    Y . Cui, J. Guo, X. Li, C.-K. Wen, and S. Jin, “Large and small model collaboration for air interface,”arXiv preprint arXiv:2512.12170, Dec. 2025

    arXiv 2025

  55. [55]

    BeamAgent: LLM- aided MIMO beamforming with decoupled intent parsing and alternating optimization for joint site selection and precoding,

    X. Wang, Y . Zhang, and N. Cheng, “BeamAgent: LLM- aided MIMO beamforming with decoupled intent parsing and alternating optimization for joint site selection and precoding,” arXiv preprint arXiv:2603.18855, Mar. 2026

    arXiv 2026

  56. [56]

    LLM agents as 6G orchestrator: A paradigm for task-oriented physical-layer automation,

    Z. Xiaoet al., “LLM agents as 6G orchestrator: A paradigm for task-oriented physical-layer automation,” inProc. IEEE GLOBECOM Workshops (GC Wkshps), 2024, pp. 1–6

    2024

  57. [57]

    Agentic wireless communication for 6G: Intent-aware and contin- uously evolving physical-layer intelligence,

    Z. Li, X. Jin, J. Pan, Q. Yang, and Z. Shi, “Agentic wireless communication for 6G: Intent-aware and contin- uously evolving physical-layer intelligence,”arXiv preprint arXiv:2602.17096, Feb. 2026

    arXiv 2026

  58. [58]

    LLM-integrated digital twins for hierarchical resource allocation in 6G networks,

    M. Haider, I. Ahmed, M. Z. Hassan, K. Hasan, and H. V . Poor, “LLM-integrated digital twins for hierarchical resource allocation in 6G networks,”IEEE Commun. Mag., pp. 1–10, Mar. 2026

    2026

  59. [59]

    AgentRAN: An agentic AI architecture for autonomous control of open 6G networks,

    M. Elkael, S. D’Oro, L. Bonati, M. Polese, Y . Lee, K. Furueda, and T. Melodia, “AgentRAN: An agentic AI architecture for autonomous control of open 6G networks,”IEEE Commun. Mag., pp. 1–7, May. 2026

    2026

  60. [60]

    ViWi: A deep learning dataset framework for vision-aided wireless communications,

    M. Alrabeiah, A. Hredzak, Z. Liu, and A. Alkhateeb, “ViWi: A deep learning dataset framework for vision-aided wireless communications,” inProc. IEEE Veh. Technol. Conf. (VTC- Fall), 2019, pp. 1–5

    2019

  61. [61]

    DeepMIMO: A generic deep learning dataset for millimeter wave and massive MIMO applications,

    A. Alkhateeb, “DeepMIMO: A generic deep learning dataset for millimeter wave and massive MIMO applications,” inProc. Inf. Theory Appl. Workshop (ITA), 2019, pp. 1–8

    2019

  62. [62]

    SynthSoM: A synthetic intelligent multi-modal sensing- communication dataset for synesthesia of machines (SoM),

    X. Cheng, H. Zhang, S. Gao, B. Liu, X. Liu, and L. Yang, “SynthSoM: A synthetic intelligent multi-modal sensing- communication dataset for synesthesia of machines (SoM),” Sci. Data, vol. 12, pp. 36–41, May 2025

    2025

  63. [63]

    Multimodal-Wireless: A large-scale dataset for sensing and communication,

    T. Mao, L. Liang, J. Yang, H. Ye, S. Jin, and G. Y . Li, “Multimodal-Wireless: A large-scale dataset for sensing and communication,” inProc. IEEE Int. Conf. Commun. (ICC), 2026, pp. 1–6

    2026

  64. [64]

    Foundation model for intelligent wireless communications,

    B. Liu, X. Liu, S. Gao, X. Cheng, and L. Yang, “Foundation model for intelligent wireless communications,”arXiv preprint arXiv:2511.22222, Nov. 2025

    Pith/arXiv arXiv 2025

  65. [65]

    Multimodal- ity in mmWave MIMO beam selection using deep learning: Datasets and challenges,

    J. Gu, B. Salehi, D. Roy, and K. R. Chowdhury, “Multimodal- ity in mmWave MIMO beam selection using deep learning: Datasets and challenges,”IEEE Commun. Mag., vol. 60, no. 11, pp. 36–41, Nov. 2022

    2022

  66. [66]

    DeepSense 6G: A large-scale real- world multi-modal sensing and communication dataset,

    A. Alkhateebet al., “DeepSense 6G: A large-scale real- world multi-modal sensing and communication dataset,”IEEE Commun. Mag., vol. 61, no. 9, pp. 122–128, Sep. 2023

    2023

  67. [67]

    CSI-Bench: A large-scale in-the-wild dataset for multi-task WiFi sensing,

    G. Zhuet al., “CSI-Bench: A large-scale in-the-wild dataset for multi-task WiFi sensing,” inProc. Adv. Neural Inf. Process. Syst. (NeurIPS), 2025

    2025

  68. [68]

    LLM-empowered resource allocation in wireless communications systems,

    W. Lee and J. Park, “LLM-empowered resource allocation in wireless communications systems,”IEEE Access, vol. 14, pp. 15 260–15 272, Jan. 2026

    2026

  69. [69]

    Wireless power control based on large language models,

    J. Wang, Y . Sheng, L. Liang, H. Ye, and S. Jin, “Wireless power control based on large language models,”arXiv preprint arXiv:2603.00474, Apr. 2026

    arXiv 2026

  70. [70]

    Adaptive resource allocation optimization using large language models in dynamic wireless environments,

    H. Noh, B. Shim, and H. J. Yang, “Adaptive resource allocation optimization using large language models in dynamic wireless environments,”IEEE Trans. Veh. Technol., vol. 74, no. 10, pp. 16 630–16 635, Oct. 2025

    2025

  71. [71]

    Weighted sum-rate maximization using weighted MMSE for MIMO-BC beamforming design,

    S. S. Christensen, R. Agarwal, E. de Carvalho, and J. M. Cioffi, “Weighted sum-rate maximization using weighted MMSE for MIMO-BC beamforming design,”IEEE Trans. Wireless Com- mun., vol. 7, no. 12, pp. 4868–4877, Dec. 2008

    2008

  72. [72]

    Graph neural networks for wireless communications: From theory to practice,

    Y . Shen, J. Zhang, S. Song, and K. B. Letaief, “Graph neural networks for wireless communications: From theory to practice,”IEEE Trans. Wireless Commun., vol. 22, no. 5, pp. 3554–3569, May 2023

    2023

  73. [73]

    Unfolding WMMSE using graph neural networks for efficient power allocation,

    A. Chowdhury, G. Verma, C. Rao, A. Swami, and S. Segarra, “Unfolding WMMSE using graph neural networks for efficient power allocation,”IEEE Trans. Wireless Commun., vol. 20, no. 9, pp. 6004–6017, Apr. 2021

    2021

  74. [74]

    LLM-augmented deep reinforcement learning for dynamic O-RAN network slicing,

    F. Lotfi, H. Rajoli, and F. Afghah, “LLM-augmented deep reinforcement learning for dynamic O-RAN network slicing,” inProc. IEEE Int. Conf. Commun. (ICC), 2025, pp. 3827– 3832

    2025

  75. [75]

    AI-driven decentralized network management: Leveraging multi-agent large language models for scalable optimization,

    H. Lee, M. Kim, S. Baek, W. Zhou, M. Debbah, and I. Lee, “AI-driven decentralized network management: Leveraging multi-agent large language models for scalable optimization,” IEEE Commun. Mag., vol. 63, no. 6, pp. 50–56, May 2025

    2025

  76. [76]

    Large language model-enabled reinforcement learning for wireless network optimization,

    J. Zhenget al., “Large language model-enabled reinforcement learning for wireless network optimization,”IEEE Commun. Mag., vol. 64, no. 4, pp. 82–89, Feb. 2026

    2026

  77. [77]

    LLM-hRIC: LLM- empowered hierarchical RAN intelligent control for O-RAN,

    L. Bao, S. Yun, J. Lee, and T. Q. Quek, “LLM-hRIC: LLM- empowered hierarchical RAN intelligent control for O-RAN,” IEEE Commun. Mag., pp. 1–7, Apr. 2026

    2026

  78. [78]

    Multi-LLM cooperation-based joint intent- driven network slicing and resource allocation,

    M. Hanet al., “Multi-LLM cooperation-based joint intent- driven network slicing and resource allocation,”IEEE Wireless Commun. Lett., vol. 15, pp. 1568–1572, Jan. 2026

    2026

  79. [79]

    ComAgent: Multi-LLM based agentic AI empowered intelligent wireless networks,

    H. Li, M. Xiao, K. Wang, R. Schober, D. I. Kim, and Y . L. Guan, “ComAgent: Multi-LLM based agentic AI empowered intelligent wireless networks,”arXiv preprint arXiv:2601.19607, Jan. 2026

    arXiv 2026

  80. [80]

    LAMeTA: Intent-aware agentic network op- timization via a large AI model-empowered two-stage ap- proach,

    Y . Liuet al., “LAMeTA: Intent-aware agentic network op- timization via a large AI model-empowered two-stage ap- proach,”IEEE J. Sel. Areas Commun., vol. 44, pp. 2462–2478, Feb. 2026

    2026

Showing first 80 references.