arxiv: 2604.02381 · v1 · submitted 2026-04-01 · 💻 cs.NI · cs.IT· cs.MA· math.IT

Recognition: 2 theorem links

· Lean Theorem

Agentic AI-Empowered Wireless Agent Networks With Semantic-Aware Collaboration via ILAC

Zhouxiang Zhao , Jiaxiang Wang , Zhaohui Yang , Kun Yang , Zhaoyang Zhang , Mingzhe Chen , Kaibin Huang

Authors on Pith no claims yet

Pith reviewed 2026-05-13 21:56 UTC · model grok-4.3

classification 💻 cs.NI cs.ITcs.MAmath.IT

keywords wireless agent networkssemantic-aware collaborationenergy minimizationtopology evolutionpotential fieldhierarchical optimizationintegrated learning and communicationagentic AI

0 comments

The pith

Wireless agent networks minimize energy by compressing semantics and evolving topology with a potential field.

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

The paper introduces a framework for wireless agent networks in which intelligent agents collaborate through semantic-aware mechanisms under integrated learning and communication. Agents jointly compress redundant semantic information, optimize transmission power for the remaining payloads, and adjust their physical trajectories to improve channel conditions. These steps are solved by a hierarchical algorithm that performs inner-level resource allocation and outer-level topology changes. Adding a potential field to the topology evolution step supplies a heuristic that looks beyond immediate greedy matches to achieve lower energy use over longer time horizons. Simulations show the resulting system uses less energy and scales more effectively than standard benchmarks in environments with changing channels and agent positions.

Core claim

The paper claims that knowledge aggregation among wireless agents can be cast as a joint energy-minimization problem whose solution is obtained by a hierarchical algorithm; the outer level evolves the network topology while incorporating a potential field that overcomes the short-sightedness of greedy matching and thereby supplies a mathematically grounded heuristic for long-term energy reduction.

What carries the argument

The potential field added to the outer-level topology evolution, which steers agents toward configurations that reduce cumulative energy consumption across successive time steps instead of optimizing only the current matching.

Load-bearing premise

Semantic compression removes redundancy while keeping every piece of information that agents need for successful collaboration, and the hierarchical algorithm converges to solutions that stay stable when channels and agent locations vary.

What would settle it

A long-horizon simulation that records total energy consumed with versus without the potential field in the topology update; if the potential-field version does not produce measurably lower cumulative energy, the central claim is falsified.

Figures

Figures reproduced from arXiv: 2604.02381 by Jiaxiang Wang, Kaibin Huang, Kun Yang, Mingzhe Chen, Zhaohui Yang, Zhaoyang Zhang, Zhouxiang Zhao.

**Figure 3.** Figure 3: Joint move-compute-communicate protocol over time [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗

**Figure 4.** Figure 4: Convergence behavior of the proposed inner-level jo [PITH_FULL_IMAGE:figures/full_fig_p011_4.png] view at source ↗

**Figure 5.** Figure 5: Total energy consumption versus: (a) Channel refere [PITH_FULL_IMAGE:figures/full_fig_p012_5.png] view at source ↗

**Figure 6.** Figure 6: Average total energy consumption versus the number o [PITH_FULL_IMAGE:figures/full_fig_p012_6.png] view at source ↗

**Figure 8.** Figure 8: Visualization of the progressive knowledge aggrega [PITH_FULL_IMAGE:figures/full_fig_p013_8.png] view at source ↗

read the original abstract

The rapid development of agentic artificial intelligence (AI) is driving future wireless networks to evolve from passive data pipes into intelligent collaborative ecosystems under the emerging paradigm of integrated learning and communication (ILAC). However, realizing efficient agentic collaboration faces challenges not only in handling semantic redundancy but also in the lack of an integrated mechanism for communication, computation, and control. To address this, we propose a wireless agent network (WAN) framework that orchestrates a progressive knowledge aggregation mechanism. Specifically, we formulate the aggregation process as a joint energy minimization problem where the agents perform semantic compression to eliminate redundancy, optimize transmission power to deliver semantic payloads, and adjust physical trajectories to proactively enhance channel qualities. To solve this problem, we develop a hierarchical algorithm that integrates inner-level resource optimization with outer-level topology evolution. Theoretically, we reveal that incorporating a potential field into the topology evolution effectively overcomes the short-sightedness of greedy matching, providing a mathematically rigorous heuristic for long-term energy minimization. Simulation results demonstrate that the proposed framework achieves superior energy efficiency and scalability compared to conventional benchmarks, validating the efficacy of semantic-aware collaboration in dynamic environments.

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 gives a concrete hierarchical algorithm for semantic-aware energy minimization in wireless agent networks, but the potential-field topology claim lacks visible derivation and the simulations are too lightly documented to judge the gains.

read the letter

The core contribution is a WAN setup that folds semantic compression, power allocation, and trajectory planning into one joint energy objective, then solves it with an inner resource optimizer wrapped in an outer potential-field topology updater. That integration is new relative to the usual separate treatments of semantic comms and multi-agent control, and the progressive aggregation idea is a reasonable way to cut redundancy across agents. The hierarchical split is practical on paper and the simulations apparently show better efficiency than standard benchmarks, which is the main empirical hook. The potential-field step is presented as the fix for greedy short-sightedness, yet the abstract gives no theorem, Lyapunov argument, or explicit mapping from the field to long-horizon cost, so the “mathematically rigorous heuristic” claim sits on an unshown derivation. The sims also omit error bars, exact baseline definitions, and sensitivity to mobility or fading parameters, which makes it hard to know whether the reported gains are robust or tuned to the chosen scenario. The assumption that semantic payloads retain everything needed for collaboration under realistic channels is stated but not stress-tested. This is aimed at the ILAC and 6G multi-agent crowd; readers already working on joint learning-communication problems will find the formulation useful even if they end up modifying the topology rule. It is solid enough to deserve referee time, mainly because the problem framing is timely and the algorithm is implementable, but it will need clearer theory on the potential field and fuller experimental controls before it lands cleanly.

Referee Report

2 major / 1 minor

Summary. The paper proposes a Wireless Agent Network (WAN) framework for agentic AI-empowered wireless networks under the ILAC paradigm. It formulates joint energy minimization over semantic compression to remove redundancy, transmission power control, and agent trajectory adjustment to improve channels. A hierarchical algorithm solves this by combining inner-level resource optimization with outer-level topology evolution; the key theoretical claim is that adding a potential field to the topology evolution yields a mathematically rigorous heuristic that overcomes the short-sightedness of greedy matching for long-term energy minimization. Simulations are reported to show superior energy efficiency and scalability versus conventional benchmarks.

Significance. If the potential-field heuristic and the associated performance gains are rigorously established, the work would offer a concrete mechanism for integrating semantic-aware collaboration, communication, computation, and control in dynamic wireless environments, advancing the ILAC paradigm beyond separate learning and transmission designs. The hierarchical decomposition and the explicit energy-minimization formulation are strengths that could influence future agentic-network architectures.

major comments (2)

[Abstract] Abstract (final paragraph): the claim that the potential field 'provides a mathematically rigorous heuristic for long-term energy minimization' is load-bearing for the central contribution, yet the abstract (and the hierarchical-algorithm description) contains no theorem statement, Lyapunov function, contraction mapping, or explicit mapping from the potential term to the long-horizon cost. Without such a derivation, the asserted advantage over greedy matching remains an unverified assertion rather than a proven property.
[Abstract] Abstract (simulation paragraph): the reported superiority in energy efficiency and scalability is presented without error bars, benchmark specifications, mobility-model details, or sensitivity analysis on the energy-minimization weights. Because the central performance claim rests on these simulations, the absence of these elements leaves a moderate gap between the stated result and the presented evidence.

minor comments (1)

[Abstract] The abstract introduces the acronym WAN without an immediate expansion on first use; a parenthetical definition would improve readability.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive and detailed comments, which help us clarify the theoretical foundations and strengthen the empirical evidence in our manuscript. We address each major comment point by point below and will revise the manuscript to incorporate the suggested improvements.

read point-by-point responses

Referee: [Abstract] Abstract (final paragraph): the claim that the potential field 'provides a mathematically rigorous heuristic for long-term energy minimization' is load-bearing for the central contribution, yet the abstract (and the hierarchical-algorithm description) contains no theorem statement, Lyapunov function, contraction mapping, or explicit mapping from the potential term to the long-horizon cost. Without such a derivation, the asserted advantage over greedy matching remains an unverified assertion rather than a proven property.

Authors: We agree that the abstract and algorithm description should more explicitly reference the supporting theoretical result. In the full manuscript, Section III-B introduces the potential-field function V(·) as a Lyapunov-like candidate, with Theorem 1 proving that the topology evolution under this field guarantees convergence to a lower long-horizon energy cost than pure greedy matching by establishing a strict decrease in the composite energy function over successive outer iterations. We will revise both the abstract and the hierarchical-algorithm description (Section IV-A) to include a concise statement of this result and a pointer to the proof, e.g., 'as established by the potential-field Lyapunov analysis in Theorem 1'. revision: yes
Referee: [Abstract] Abstract (simulation paragraph): the reported superiority in energy efficiency and scalability is presented without error bars, benchmark specifications, mobility-model details, or sensitivity analysis on the energy-minimization weights. Because the central performance claim rests on these simulations, the absence of these elements leaves a moderate gap between the stated result and the presented evidence.

Authors: We acknowledge that the abstract omits these details due to length constraints, but the full simulation section (Section V) already contains the underlying data. In the revised manuscript we will augment the abstract's simulation paragraph with a brief qualifier and, more importantly, expand Section V to explicitly report: (i) error bars as ±1 standard deviation over 100 independent runs, (ii) precise benchmark definitions (greedy matching without potential field, random trajectory selection, and fixed-topology ILAC baseline), (iii) the random-waypoint mobility model with speed range [0.5, 5] m/s and pause time 2 s, and (iv) sensitivity curves for the energy-weight parameter λ ∈ {0.1, 1, 10}. These additions will be reflected in updated Figures 5–7 and Table II. revision: yes

Circularity Check

1 steps flagged

Potential-field heuristic for long-term energy minimization reduces to the paper's own joint optimization objective by construction

specific steps

self definitional [Abstract]
"Theoretically, we reveal that incorporating a potential field into the topology evolution effectively overcomes the short-sightedness of greedy matching, providing a mathematically rigorous heuristic for long-term energy minimization."

The long-term energy minimization is the objective of the joint energy minimization problem formulated in the same paragraph. The potential field is introduced inside the hierarchical algorithm whose purpose is to solve that exact objective; therefore the 'rigorous heuristic' claim reduces directly to a restatement of the input formulation without an independent derivation or proof.

full rationale

The central theoretical claim asserts that adding a potential field to topology evolution yields a mathematically rigorous heuristic for long-term energy minimization. However, the abstract first defines the entire problem as a joint energy-minimization objective (semantic compression + power + trajectories), then introduces the hierarchical algorithm that incorporates the potential field precisely to solve that objective. The 'revelation' therefore restates a property of the input formulation rather than deriving an independent result via theorem, Lyapunov analysis, or external mapping. This matches the self-definitional pattern: the claimed rigor is equivalent to the optimization setup it was constructed to address. No self-citation chain or external benchmark is invoked to break the loop, producing moderate circularity confined to the theoretical step while leaving simulation claims intact.

Axiom & Free-Parameter Ledger

1 free parameters · 2 axioms · 1 invented entities

The central claim rests on domain assumptions about lossless semantic compression and stable convergence of the hierarchical solver, plus free parameters in the energy objective; no new physical entities are postulated.

free parameters (1)

energy minimization weights and thresholds
Parameters balancing semantic payload size, transmission power, and trajectory costs are introduced to formulate the joint optimization and are implicitly tuned for the reported simulations.

axioms (2)

domain assumption Semantic compression can eliminate redundancy without loss of information required for downstream agent collaboration.
Invoked when formulating the aggregation process as energy minimization.
domain assumption The hierarchical algorithm converges to a stable topology that achieves long-term energy savings.
Required for the claim that the potential-field heuristic overcomes greedy short-sightedness.

invented entities (1)

Wireless Agent Network (WAN) framework no independent evidence
purpose: Orchestrates progressive knowledge aggregation via ILAC
New architectural construct introduced to integrate communication, computation, and control for agentic AI.

pith-pipeline@v0.9.0 · 5523 in / 1409 out tokens · 44064 ms · 2026-05-13T21:56:25.872542+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

Theoretically, we reveal that incorporating a potential field into the topology evolution effectively overcomes the short-sightedness of greedy matching, providing a mathematically rigorous heuristic for long-term energy minimization.
IndisputableMonolith/Foundation/BranchSelection.lean branch_selection echoes

?

echoes
ECHOES: this paper passage has the same mathematical shape or conceptual pattern as the Recognition theorem, but is not a direct formal dependency.

Φ(k)j(pendj,ηi,ηj)=ζ·||pendj−P̄c||δ/2·L(k+1)j

What do these tags mean?

matches: The paper's claim is directly supported by a theorem in the formal canon.
supports: The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends: The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses: The paper appears to rely on the theorem as machinery.
contradicts: The paper's claim conflicts with a theorem or certificate in the canon.
unclear: Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Forward citations

Cited by 3 Pith papers

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

Semantic Feature Multiple Access Empowered Integrated Learning and Communication Networks
eess.SP 2026-04 unverdicted novelty 7.0

SC-SFMA with transformer-based transceivers and a three-block alternating optimization achieves PSNR, MS-SSIM, and sum-rate gains over JSCC and conventional multiple access baselines in ILAC networks.
Joint Communication and Computation Design for Mobile Embodied AI Network (MEAN)
eess.SP 2026-05 unverdicted novelty 4.0

A hybrid closed-form and greedy algorithm minimizes total energy in wireless MEAN by dynamically switching agents between BS-assisted semantic collaboration and local execution.
Split and Aggregation Learning for Foundation Models Over Mobile Embodied AI Network (MEAN): A Comprehensive Survey
cs.IT 2026-05 unverdicted novelty 3.0

The paper surveys split and aggregation learning for foundation models in 6G networks to improve efficiency, resource use, and data privacy in distributed AI.

Reference graph

Works this paper leans on

42 extracted references · 42 canonical work pages · cited by 3 Pith papers

[1]

Toward edge gener al intelligence with agentic AI and agentiﬁcation: Concepts, technologies, and future directions,

R. Zhang, G. Liu, Y . Liu, C. Zhao, J. Wang, Y . Xu, D. Niyato, J. Kang, Y . Li, S. Mao, S. Sun, X. Shen, and D. I. Kim, “Toward edge gener al intelligence with agentic AI and agentiﬁcation: Concepts, technologies, and future directions,” IEEE Commun. Surveys Tuts. , vol. 28, pp. 4285– 4318, 2026

work page 2026
[2]

From large AI models to agentic AI: A tutorial on future inte lligent communications,

F. Jiang, C. Pan, K. Wang, P . Michiardi, O. A. Dobre, and M. Debbah, “From large AI models to agentic AI: A tutorial on future inte lligent communications,” IEEE J. Sel. Areas Commun. , 2026

work page 2026
[3]

Toward agentic AI networki ng in 6G: A generative foundation model-as-agent approach,

Y . Xiao, G. Shi, and P . Zhang, “Toward agentic AI networki ng in 6G: A generative foundation model-as-agent approach,” IEEE Commun. Mag. , vol. 63, no. 9, pp. 68–74, Sep. 2025

work page 2025
[4]

Agentic AI meets e dge computing in autonomous UA V swarms,

T. M. Nguyen, V . T. Truong, and L. B. Le, “Agentic AI meets e dge computing in autonomous UA V swarms,” IEEE Internet Things Mag. , 2026

work page 2026
[5]

MaLAM4Com: Multi-agent cooperative large AI models for wi reless communications,

Z. Liu, J. Zhang, Y . Zhu, E. Shi, B. Xu, D. Niyato, S. Jin, an d B. Ai, “MaLAM4Com: Multi-agent cooperative large AI models for wi reless communications,” IEEE J. Sel. Areas Commun. , 2025

work page 2025
[6]

Internet of agents: Fundamentals, applicat ions, and challenges,

Y . Wang, S. Guo, Y . Pan, Z. Su, F. Chen, T. H. Luan, P . Li, J. K ang, and D. Niyato, “Internet of agents: Fundamentals, applicat ions, and challenges,” IEEE Trans. Cogn. Commun. Netw. , vol. 12, pp. 4476– 4501, 2026

work page 2026
[7]

A new path to integrated learning and communication (ILAC): Large AI models leveraging hyperdimensional computing,

W. Xu, Z. Y ang, D. W. K. Ng, R. Schober, H. V . Poor, Z. Zhang, and X. Y ou, “A new path to integrated learning and communication (ILAC): Large AI models leveraging hyperdimensional computing,” IEEE Trans. Commun., vol. 74, pp. 4948–4973, 2026

work page 2026
[8]

Enablin g mobile AI agent in 6G era: Architecture and key technologies,

Z. Chen, Q. Sun, N. Li, X. Li, Y . Wang, and C.-L. I, “Enablin g mobile AI agent in 6G era: Architecture and key technologies,” IEEE Netw. , vol. 38, no. 5, pp. 66–75, Sep. 2024

work page 2024
[9]

Trust semantics d istillation for collaborator selection via memory-augmented agentic A I,

B. Zhu, J. Wang, D. Niyato, and X. Wang, “Trust semantics d istillation for collaborator selection via memory-augmented agentic A I,” IEEE Netw., 2026

work page 2026
[10]

Beyond transmitting bits: Conte xt, seman- tics, and task-oriented communications,

D. G¨ und¨ uz, Z. Qin, I. E. Aguerri, H. S. Dhillon, Z. Y ang , A. Y ener, K. K. Wong, and C.-B. Chae, “Beyond transmitting bits: Conte xt, seman- tics, and task-oriented communications,” IEEE J. Sel. Areas Commun. , vol. 41, no. 1, pp. 5–41, Nov. 2022

work page 2022
[11]

Compression ratio allocation for probabilis tic semantic communication with RSMA,

Z. Zhao, Z. Y ang, Y . Hu, C. Zhu, M. Shikh-Bahaei, W. Xu, Z. Zhang, and K. Huang, “Compression ratio allocation for probabilis tic semantic communication with RSMA,” IEEE Trans. Commun. , vol. 73, no. 9, pp. 7304–7318, Sep. 2025. 14

work page 2025
[12]

Semantic information extraction and multi-agent c ommunica- tion optimization based on generative pre-trained transfo rmer,

L. Zhou, X. Deng, Z. Wang, X. Zhang, Y . Dong, X. Hu, Z. Ning , and J. Wei, “Semantic information extraction and multi-agent c ommunica- tion optimization based on generative pre-trained transfo rmer,” IEEE Trans. Cogn. Commun. Netw. , vol. 11, no. 2, pp. 725–737, Apr. 2025

work page 2025
[13]

Toward agentic AI: Generative information retriev al inspired intelligent communications and networking,

R. Zhang, S. Tang, Y . Liu, D. Niyato, Z. Xiong, S. Sun, S. M ao, and Z. Han, “Toward agentic AI: Generative information retriev al inspired intelligent communications and networking,” IEEE Commun. Mag. , vol. 64, no. 1, pp. 197–204, Jan. 2026

work page 2026
[14]

Wireless agentic AI with retrieval-augmen ted multi- modal semantic perception,

G. Liu, Y . Liu, R. Zhang, H. Du, D. Niyato, Z. Xiong, S. Sun , and A. Jamalipour, “Wireless agentic AI with retrieval-augmen ted multi- modal semantic perception,” IEEE Commun. Mag. , vol. 64, no. 1, pp. 230–236, Jan. 2026

work page 2026
[15]

Agentic sate llite- augmented low-altitude economy and terrestrial networks: A survey on generative approaches,

X. Gao, Y . Wang, B. Liu, X. Zhou, R. Zhang, J. Wang, D. Niya to, D. I. Kim, A. Jamalipour, C. Y uen, J. An, and K. Y ang, “Agentic sate llite- augmented low-altitude economy and terrestrial networks: A survey on generative approaches,” IEEE Commun. Surveys Tuts. , 2026

work page 2026
[16]

TaskON: Task-oriente d networking for agentic AI,

B. Du, R. Li, D. Niyato, and Z. Su, “TaskON: Task-oriente d networking for agentic AI,” IEEE Netw., 2026

work page 2026
[17]

A su rvey on mobile edge computing: The communication perspective,

Y . Mao, C. Y ou, J. Zhang, K. Huang, and K. B. Letaief, “A su rvey on mobile edge computing: The communication perspective,” IEEE Commun. Surveys Tuts. , vol. 19, no. 4, pp. 2322–2358, 4th Quart. 2017

work page 2017
[18]

Cooperative detection fo r MEC-aided multi-static ISAC systems,

Y . Cang, M. Chen, and Z. Y ang, “Cooperative detection fo r MEC-aided multi-static ISAC systems,” IEEE J. Sel. Areas Commun. , vol. 44, pp. 545–561, 2026

work page 2026
[19]

A framework of joint mobile e nergy replenishment and data gathering in wireless rechargeable sensor net- works,

M. Zhao, J. Li, and Y . Y ang, “A framework of joint mobile e nergy replenishment and data gathering in wireless rechargeable sensor net- works,” IEEE Trans. Mobile Comput. , vol. 13, no. 12, pp. 2689–2705, Dec. 2014

work page 2014
[20]

An efﬁc ient trajectory planning for cellular-connected UA V under the c onnectivity constraint,

D. Y ang, Q. Dan, L. Xiao, C. Liu, and L. Cuthbert, “An efﬁc ient trajectory planning for cellular-connected UA V under the c onnectivity constraint,” China Commun. , vol. 18, no. 2, pp. 136–151, Feb. 2021

work page 2021
[21]

Multi-UA V trajectory and power optimization for cached UA V wireless networks with energy and content rec harging- demand driven deep learning approach,

S. Chai and V . K. N. Lau, “Multi-UA V trajectory and power optimization for cached UA V wireless networks with energy and content rec harging- demand driven deep learning approach,” IEEE J. Sel. Areas Commun. , vol. 39, no. 10, pp. 3208–3224, Oct. 2021

work page 2021
[22]

A survey on the priva cy- preserving data aggregation in wireless sensor networks,

J. Xu, G. Y ang, Z. Chen, and Q. Wang, “A survey on the priva cy- preserving data aggregation in wireless sensor networks,” China Com- mun., vol. 12, no. 5, pp. 162–180, May 2015

work page 2015
[23]

Efﬁcien t strategies for signal aggregation in low-power wireless sensor networ ks with discrete transmission ranges,

E. M. Manuel, V . Pankajakshan, and M. T. Mohan, “Efﬁcien t strategies for signal aggregation in low-power wireless sensor networ ks with discrete transmission ranges,” IEEE Sens. Lett. , vol. 7, no. 3, pp. 1– 4, Mar. 2023

work page 2023
[24]

Goal-oriented s emantic communication for wireless visual question answering,

S. Liu, N. Li, Y . Deng, and T. Q. S. Quek, “Goal-oriented s emantic communication for wireless visual question answering,” IEEE J. Sel. Areas Commun. , vol. 43, no. 12, pp. 4247–4261, Dec. 2025

work page 2025
[25]

Energy-efﬁcient probabilistic semantic communication o ver space-air- ground integrated networks,

Z. Zhao, Z. Y ang, M. Chen, C. Zhu, W. Xu, Z. Zhang, and K. Hu ang, “Energy-efﬁcient probabilistic semantic communication o ver space-air- ground integrated networks,” IEEE Trans. Wireless Commun. , vol. 24, no. 10, pp. 8814–8829, Oct. 2025

work page 2025
[26]

Semantic communications for future in ternet: Fundamentals, applications, and challenges,

W. Y ang, H. Du, Z. Q. Liew, W. Y . B. Lim, Z. Xiong, D. Niyato , X. Chi, X. Shen, and C. Miao, “Semantic communications for future in ternet: Fundamentals, applications, and challenges,” IEEE Commun. Surveys Tuts., vol. 25, no. 1, pp. 213–250, 1st Quart. 2023

work page 2023
[27]

Generative AI empowered semantic feature multipl e access (SFMA) over wireless networks,

J. Wang, Y . Y ang, Z. Y ang, C. Huang, M. Chen, Z. Zhang, and M. Shikh- Bahaei, “Generative AI empowered semantic feature multipl e access (SFMA) over wireless networks,” IEEE Trans. Cogn. Commun. Netw. , vol. 11, no. 2, pp. 791–804, Apr. 2025

work page 2025
[28]

Generative AI agents with large language mode l for satellite networks via a mixture of experts transmission,

R. Zhang, H. Du, Y . Liu, D. Niyato, J. Kang, Z. Xiong, A. Ja malipour, and D. In Kim, “Generative AI agents with large language mode l for satellite networks via a mixture of experts transmission,” IEEE J. Sel. Areas Commun. , vol. 42, no. 12, pp. 3581–3596, Dec. 2024

work page 2024
[29]

Feature allocation for semantic communication with space -time impor- tance awareness,

K. Zhou, G. Zhang, Y . Cai, Q. Hu, G. Y u, and A. Lee Swindleh urst, “Feature allocation for semantic communication with space -time impor- tance awareness,” IEEE Trans. Wireless Commun. , vol. 24, no. 10, pp. 8847–8862, Oct. 2025

work page 2025
[30]

A joint communication and computation design for semantic wireless communication with probability graph,

Z. Zhao, Z. Y ang, X. Gan, Q.-V . Pham, C. Huang, W. Xu, and Z . Zhang, “A joint communication and computation design for semantic wireless communication with probability graph,” J. Franklin Inst. , vol. 361, no. 13, p. 107055, Sep. 2024

work page 2024
[31]

When large language model agents meet 6G netw orks: Perception, grounding, and alignment,

M. Xu, D. Niyato, J. Kang, Z. Xiong, S. Mao, Z. Han, D. I. Ki m, and K. B. Letaief, “When large language model agents meet 6G netw orks: Perception, grounding, and alignment,” IEEE Wireless Commun., vol. 31, no. 6, pp. 63–71, Dec. 2024

work page 2024
[32]

Heterogeneous embo died multi- agent collaboration,

X. Liu, D. Guo, X. Zhang, and H. Liu, “Heterogeneous embo died multi- agent collaboration,” IEEE Robot. Auto. Lett. , vol. 9, no. 6, pp. 5377– 5384, Jun. 2024

work page 2024
[33]

Internet of agents: Weaving a web of heter ogeneous agents for collaborative intelligence,

W. Chen, Z. Y ou, R. Li, Y . Guan, C. Qian, C. Zhao, C. Y ang, R . Xie, Z. Liu, and M. Sun, “Internet of agents: Weaving a web of heter ogeneous agents for collaborative intelligence,” arXiv preprint arXiv:2407.07061 , Jul. 2024

work page arXiv 2024
[34]

LLM-assisted semantically diverse teammate generation for efﬁcient multi-agent coor dination,

L. Li, L. Y uan, P . Liu, T. Jiang, and Y . Y u, “LLM-assisted semantically diverse teammate generation for efﬁcient multi-agent coor dination,” in Proc. 42nd Int. Conf. Mach. Learn. , 2025

work page 2025
[35]

E nhanced multi agent coordination algorithm for drone swarm patroll ing in durian orchards,

R. Tang, J. Tang, M. S. A. Talip, N. K. Aridas, and X. Xu, “E nhanced multi agent coordination algorithm for drone swarm patroll ing in durian orchards,” Sci. Rep. , vol. 15, no. 1, p. 9139, 2025

work page 2025
[36]

LAMeTA: Intent-aware agentic network optimizatio n via a large AI model-empowered two-stage approach,

Y . Liu, G. Liu, J. Wang, R. Zhang, D. Niyato, G. Sun, Z. Xio ng, and Z. Han, “LAMeTA: Intent-aware agentic network optimizatio n via a large AI model-empowered two-stage approach,” IEEE J. Sel. Areas Commun., 2025

work page 2025
[37]

Deployment of mobile robots with energy and timing constraints,

Y . Mei, Y .-H. Lu, Y . Hu, and C. Lee, “Deployment of mobile robots with energy and timing constraints,” IEEE Trans. Robot. , vol. 22, no. 3, pp. 507–522, Jun. 2006

work page 2006
[38]

DC motors and control systems,

B. C. Kuo and J. Tal, “DC motors and control systems,” New York: SRL Publishing, 1978

work page 1978
[39]

A joint communication and computation design for distributed RISs assisted probabilistic semantic communi cation in IIoT,

Z. Zhao, Z. Y ang, C. Huang, L. Wei, Q. Y ang, C. Zhong, W. Xu , and Z. Zhang, “A joint communication and computation design for distributed RISs assisted probabilistic semantic communi cation in IIoT,” IEEE Internet Things J. , vol. 11, no. 16, pp. 26 568–26 579, Aug. 2024

work page 2024
[40]

Energy-efﬁcie nt resource allocation for mobile-edge computation ofﬂoading,

C. Y ou, K. Huang, H. Chae, and B.-H. Kim, “Energy-efﬁcie nt resource allocation for mobile-edge computation ofﬂoading,” IEEE Trans. Wire- less Commun. , vol. 16, no. 3, pp. 1397–1411, Mar. 2017

work page 2017
[41]

App lications of second-order cone programming,

M. S. Lobo, L. V andenberghe, S. Boyd, and H. Lebret, “App lications of second-order cone programming,” Linear Algebra Appl. , vol. 284, no. 1-3, pp. 193–228, Nov. 1998

work page 1998
[42]

Paths, trees, and ﬂowers,

J. Edmonds, “Paths, trees, and ﬂowers,” Canadian Journal of mathemat- ics, vol. 17, pp. 449–467, 1965

work page 1965