Multi-distribution functionals reduce to integrals of coincidence divergences
Monotonicity under data processing and additivity on independent products force every such functional to an integral over four strata
Information Theory
math.IT is an alias for cs.IT. Covers theoretical and experimental aspects of information theory and coding.
Monotonicity under data processing and additivity on independent products force every such functional to an integral over four strata
The reduction uses only n plus little-o-n oracle bits and equates decompression ratios to Kolmogorov complexity rates.
The quadratic coefficient a must meet a prime-dependent valuation threshold at every p^α exactly dividing N.
· “A Local Valuation Criterion for Quadratic-Permutation Interleaved Zadoff--Chu Sequences”
A single integral construction unifies two rival quantum Rényi divergence families and extends to general von Neumann algebras, with commut
Standard convex programs recover rank-r matrices from O(rn) heavy-tailed quadratic samples, no Gaussian assumption needed
· “Low-Rank Matrix Recovery via Heavy-Tailed Quadratic Sampling”
A cancellation trick binds non-relative, relative, and cross informational functionals into sharp bounds with Gaussian and Beta minimizers.
Poly-time algorithm succeeds for ρ² above Otter's constant whenever dimension is polylogarithmic
A complete contraction proof shows the messages settle uniquely for any positive start
· “On the Convergence of Belief Propagation for Multipath Data Association in Target Tracking”
A row-span constraint turns curve-decoding into a local property and removes the exponential loss
Relaxed density evolution turns edge probabilities into a deterministic loss; the resulting matrices win by 0.03–0.18 dB under min-sum.
· “Learning LDPC codes with quantized density evolution over relaxed protographs”
Joint orientation-and-beam design plus a fast GNN keep multiuser rates high even with few antennas.
The same Gaussian formula tells you the finest binning that still detects miscalibration
· “Why Constants Matter in Distribution Testing: From Uniformity to Calibration”
Centrality reduces constacyclic codes of length p^s to explicit left ideals that yield MDS Gray images
Closed-loop LQR, not sum rate, becomes the allocation objective for multi-sensor 6G loops
· “Deep Reinforcement Learning-Empowered Wireless Sensor Networking for 6G Closed-Loop Controls”
Superimposed pilots then match single-user accuracy and beat embedded pilots on spectral efficiency for sinc pulses
· “Multiuser Zak-OTFS on the Uplink with Superimposed Spread-Pilots”
MSRD diagonals settle the Etzion–Silberstein bound for extreme distances over every field
· “On the Etzion-Silberstein conjecture for block Ferrers diagrams”
Flow-matching diffusion learns the full distribution of angular radio maps, not just their average, enabling beam sweeps and likelihoodbased
· “RadioDiff-v2: Generative Angular Radio Maps for Multi-Beam Selection and Localization”
Estimating high-order entropy functionals needs only linear-in-α samples, proven optimal via quantum primitives that yield a purely class
· “Towards Minimax Estimation of High-Order Functionals by Quantum Arguments”
Adding a third rotational degree of freedom to a robot neck costs more energy but transmits less visual information — a bottleneck with real
· “Communicative Efficiency of Single vs. Multi-Axis Robot Neck Motion”
Adding a scalar-shift parameter to trace-based defining-set codes forces self-orthogonality and produces explicit four-weight families with
Fresh queries lift identification and quality ceilings exponentially, at the Chernoff rate
· “Computing with Stochastic Oracles in AI-Augmented Computation”
Injecting noise into the sensing Hamiltonian lets entangled networks estimate averages privately, without a trusted curator.
Clos interconnects cut peak fronthaul link loads 75% versus trees, keeping cell-free MIMO within reach of off-the-shelf hardware.
· “Rethinking Fronthaul Topologies for Cell-Free 6G Networks”
Optimal detection needs ~1/d samples; majority-vote needs ~1/d². The gap prices the cost of coarse local decisions.
· “Sample complexity bounds for the Jensen-Shannon divergence”
A succinct encoding lets you run breadth-first search in O(n) time using o(n) working bits and keep the tree for constant-time queries, plus
Two generalized Levenshtein models for insertion errors are proven to require exactly ceil((n+2)/2) channels for unique recovery, with full
· “Unique Insertion Error Patterns in Levenshtein's Reconstruction Problem”
A new upper bound on 5G delay violation probability accounts for decoding, feedback, and control-signal delays that prior models ignored,揭示了
· “Delay Violation Probability Modeling for 5G Systems with HARQ Operation”
A leader-follower game structure with Bayesian Q-learning detects malicious receivers with 69% accuracy while holding capacity near 2500Mbps
· “6G Sensing Security: Distributed Game-Theoretic RL for Urban Beamforming and Attacker Detection”
An aerial 5G platform can replace always-on macro-cells, letting both coverage and capacity cells sleep during low traffic.
· “HAPS as a Hypercell: Enabling Coverage and Capacity Carrier Shutdown in Cellular Networks”
Two geometric bounds yield concrete design rules for scattering network filters on low-dimensional data
· “Separation Capacity of Scattering Networks on Low-Dimensional Datasets”
Row codes purify residual errors; high-rate columns clean them without rate loss on any BMS channel.
· “From Bit to Block: Capacity Achievement via Product Coding”
The negative ELBO equals data entropy plus the KL between the true and learned reverse trajectories, making all noising processes share one
Matching lower bound closes the constant at 2Φ(−u*/2), proving Poisson approximation is tight for multinomial testing.
The usual constant-copy tester vanishes; learning costs Θ(n²/k) non-adaptively, so testing and learning match when memory is fractional
· “Optimal Stabilizer Testing and Learning with Limited Quantum Memory”
The architecture keeps beamforming performance close to full-PS sub-connected designs while lowering hardware needs in single- and multi-RF
Generalized rank weights over chain rings and poset weights over quasi-Frobenius rings inherit Singleton bounds and dualities.
The calculation over all 3.69 million cubic orbits also raises the covering radius lower bound to 408
· “The Weight Distribution of the Third-Order Reed-Muller Code of Length 2048”
C(u,a) constructions give explicit control over hull dimension and dual distance, producing 236 confirmed improvements for lengths up to 40.
· “Generalized Extended Codes with Applications in Entanglement-Assisted Qubit and Qutrit Codes”
Latent codes sent over idle channels replace full-image coding and still let the receiver finish the task quickly.
· “Low-Latency Task-Oriented Image Transmission with Opportunistic Spectrum Access”
The three-parameter family lies below an existing upper bound whenever both are applied to quantum states.
A left-d-regular expander mask reduces learned decoder values from mn to dn while preserving most reconstruction quality in language-model S
· “Expander Sparse Autoencoders: Parameter-Efficient Dictionaries for Mechanistic Interpretability”
The construction works for arbitrary lengths over any finite chain ring and yields the code rank plus exact conditions for MHDR and MDS via
· “On the structure of constacyclic codes over finite chain rings”
Indistinguishable SAT/UNSAT pairs force wide clauses in Resolution and push proof size toward 2^N.
· “Self-Referential K-SAT and the Finite Analogue of G\"odel's Incompleteness Theorem”
Maximal leakage from classical data to encoding limits any quantum task, with tight frames optimal in small dimensions.
Approximations preserve quantum properties while yielding regret bounds that improve sample efficiency and lower overhead in GP bandit optim
· “Balancing Expressivity and Learnability in Quantum Kernel Bandit Optimization”
Extending Cover's counting theory shows how manifold structure shapes classification capacity and generalization.
· “Function-Counting Theory for Low-Dimensional Data Structures”
Multimodal model predicts the field from geometry and motion to raise effective rate 5.67 percent and cut outage 8.35 percent on urban bench
Rank-deficient correlations let limited-RF-chain multiuser beamforming match digital performance
DRL jointly tunes beamforming and surface geometry to improve sensing while preserving communication quality.
SJB and LB schemes yield expressions and scaling laws showing K-factor impacts communication more than sensing.
· “Fundamental Limits of Random Downlink Integrated Sensing and Communication over Rician Channels”
The smallest balanced induced subgraph's densest part determines when exact recovery from a random graph becomes possible with high probabil
Reformulated inference cuts iterations and supports higher-order trackers that become zero-phase over full history.
Row-echelon transform yields scalar annulus sub-channels whose capacity is bracketed tightly by geometry and entropy-power arguments.
Four agents generate, optimize, evaluate and reflect on routes to raise energy transfer performance over single models.
Notebook-equipped agents reach stable coordination at high capacities; stateless agents collapse once vocabulary exceeds context limits.
· “From Signals to Structure: How Memory Architecture Drives Language Emergence in LLM Agents”
Pivotal reduction to a scalar raises the number of tokens required, with matching bounds in each regime.
The exact rate for the ρ-th moment of coset rank equals the unconstrained value minus ρ times one minus the code rate.
· “Guesswork Under Linear Constraints: Exact Exponent for Coset Decoding”
The scheme keeps error rates falling at high SNR where standard differential receivers plateau in varying channels
· “Decision Feedback Differential Detection for Reconfigurable Intelligent Surfaces”
A new absorbing Markov construction gives exact SMDP parameters for multi-threshold policies that minimize weighted AoII cost plus energy un
· “Dual-Regime Absorbing Markov Chain Theory in Remote Estimation: Age-Minimizing Push Policies”
Density evolution shows seeded decoding reaches the design-rate limit for equal-rate MN/HA families.
This exposes a privacy flaw in semantic comms; adversarial training widens the accuracy gap at the relay while preserving receiver performan
· “Semantic Leakage and Privacy Preservation in Relay-Assisted Semantic Communications”
Non-cooperating modules with hard boundaries allow practical scaling of 6G antenna arrays.
· “Trade-Offs in Decentralized Gigantic MIMO with Hard-Boundary Constraints”
The algorithm approximates exact marginalization over object partitions while using only a fraction of the computation required by direct en
· “Gaussian Belief Propagation for Tracking With Unresolved Measurements”
Review explains how sigma protocols for two coding problems become non-interactive signatures via Fiat-Shamir.
· “Digital signature schemes based on code equivalence and syndrome decoding from restricted errors”
Per-antenna orientation optimization with alternating algorithm outperforms fixed and array-wise schemes while meeting rate constraints.
· “Antenna Orientation Optimization for Rotatable Antenna-Enabled ISAC Systems”
Dual-rail systems with postselection surpass single-rail noise limits at kiloquop scales in simulations.
· “The limits of erasure-based postselection for quantum error mitigation”
Bounds tighten at low SNR and saturate at high SNR; LMMSE saturates based on spatial combining ratio.
· “Fundamental Limits of Quantized MIMO ISAC under Gaussian Signaling”
Extension of Chen-Ding method determines exact parameters via zero segments and gives refined choices for larger distances.
Hybrid neuroevolution and supervised training lets the system handle discrete phases and one-bit feedback while beating Kalman and particle
Port activation search in reconfigurable arrays lowers the highest sidelobe while keeping main beam width steady.
The network learns full channel vectors from partial observations, cutting detection error rates versus fixed-antenna baselines.
The number of binary label assignments realizable by a scattering network is controlled by its wavelet, layer, and pooling choices.
Model predictive control on the joint distribution cuts weighted age of incorrect information plus sampling costs over delayed erasure links
· “When and Which Sensor to Observe? Timely Tracking of a Joint Markov Source”
Adversary contaminates training with waveform to alter inference for one user while sparing the other in multiple access semantic system.
· “Wireless Backdoor Attack and Defense for Semantic Communications over Multiple Access Channel”
A network generates noise from the legitimate user's channel to interfere only with the eavesdropper over MIMO links.
· “Semantic Noise Aided Secure Image Transmission over MIMO Fading Channels”
Hybrid latent space captures environment, channels, and mobility to enable proactive SCD scheduling.
Covering-plus-sketching compression achieves the bound exactly when each coordinate is binary, but a gap remains for larger alphabets.
Variants that balance receiver error and secrecy rates deliver both reliability and security over THz channels.
· “Preprocessing for Physical-Layer Security in Wireless THz-Communication”
Auditors update a likelihood ratio after each query and halt when compliance or violation is supported, cutting unnecessary model interactio
Distances for repeated-root constacyclic codes of length 4p^s over R3 are explicit when Δ is non-square and via decompositions when square.