arXiv subjects

Arash Shahmansoori

Publications and source records attributed to Arash Shahmansoori.

7 recordsLinked to original sources

arXiv2026

The E$Δ$-MHC-Geo Transformer: Adaptive Geodesic Operations with Guaranteed Orthogonality

We present the E$Δ$-MHC-Geo Transformer, a novel architecture that unifies Manifold-Constrained Hyper-Connections (mHC), Deep Delta Learning (DDL), and the Cayley transform to obtain input-adaptive, unconditionally orthogonal residual connections. Unlike DDL, whose Householder operator is orthogonal only at $β\in \{0,2\}$, our Data-Dependent Cayley rotation $Q(x)=(I+(β/2)A(x))^{-1}(I-(β/2)A(x))$ preserves orthogonality for all $β$ and all inputs. To handle negation, an eigenvalue $-1$ case that Cayley provably excludes, we introduce the E$Δ$-MHC-Geo Hybrid, which combines Cayley rotation with Householder reflection via a learned operator-selection gate $X'=γ(X)Q(X)X+(1-γ(X))H_2(X)X$. A midpoint-collapse regularizer, $4γ(1-γ)$, encourages boundary gate decisions, where each selected component is orthogonal. In matched-parameter comparisons, with approximately 1.79M parameters per model and mean +/- standard deviation over 3 seeds, against four baselines including the concurrent JPmHC, E$Δ$-MHC-Geo achieves the best long-horizon stability, 1.9x over JPmHC and 3.8x over GPT; the best near-$π$ rotation loss, 4.5x over JPmHC on single-plane; strong norm preservation, with 0.001 mean deviation; and 0.96 negation cosine alignment in a diagnostic reflection probe, all with 33% fewer layers. While JPmHC's wider representation excels on pure rotation, its finite Cayley residual mixer excludes an exact $λ=-1$ operator and has no reflection branch, motivating our hybrid approach for accessing both connected components of $O(n)$.

cs.LG↗

arXiv2026

PRECEPT: Planning Resilience via Experience, Context Engineering & Probing Trajectories A Unified Framework for Test-Time Adaptation with Compositional Rule Learning and Pareto-Guided Prompt Evolution

LLM agents that store knowledge as natural language suffer steep retrieval degradation as condition count grows, often struggle to compose learned rules reliably, and typically lack explicit mechanisms to detect stale or adversarial knowledge. We introduce PRECEPT, a unified framework for test-time adaptation with three tightly coupled components: (1) deterministic exact-match rule retrieval over structured condition keys, (2) conflict-aware memory with Bayesian source reliability and threshold-based rule invalidation, and (3) COMPASS, a Pareto-guided prompt-evolution outer loop. Exact retrieval eliminates partial-match interpretation errors on the deterministic path (0% by construction, vs 94.4% under Theorem~B.6's independence model at N=10) and supports compositional stacking through a semantic tier hierarchy; conflict-aware memory resolves static--dynamic disagreements and supports drift adaptation; COMPASS evaluates prompts through the same end-to-end execution pipeline. Results (9--10 seeds): PRECEPT achieves a +41.1pp first-try advantage over Full Reflexion (d>1.9), +33.3pp compositional generalization (d=1.55), 100% $P_1$ on 2-way logistics compositions (d=2.64), +40--55pp continuous learning gains, strong eventual robustness under adversarial static knowledge (100% logistics with adversarial SK active; partial recovery on integration), +55.0pp drift recovery (d=0.95, p=0.031), and 61% fewer steps. Core comparisons are statistically significant, often at p<0.001.

cs.AI↗

arXiv2024

Concurrent Brainstorming & Hypothesis Satisfying: An Iterative Framework for Enhanced Retrieval-Augmented Generation (R2CBR3H-SR)

Addressing the complexity of comprehensive information retrieval, this study introduces an innovative, iterative retrieval-augmented generation system. Our approach uniquely integrates a vector-space driven re-ranking mechanism with concurrent brainstorming to expedite the retrieval of highly relevant documents, thereby streamlining the generation of potential queries. This sets the stage for our novel hybrid process, which synergistically combines hypothesis formulation with satisfying decision-making strategy to determine content adequacy, leveraging a chain of thought-based prompting technique. This unified hypothesize-satisfied phase intelligently distills information to ascertain whether user queries have been satisfactorily addressed. Upon reaching this criterion, the system refines its output into a concise representation, maximizing conceptual density with minimal verbosity. The iterative nature of the workflow enhances process efficiency and accuracy. Crucially, the concurrency within the brainstorming phase significantly accelerates recursive operations, facilitating rapid convergence to solution satisfaction. Compared to conventional methods, our system demonstrates a marked improvement in computational time and cost-effectiveness. This research advances the state-of-the-art in intelligent retrieval systems, setting a new benchmark for resource-efficient information extraction and abstraction in knowledge-intensive applications.

cs.IT↗

arXiv2023

Dynamic Recognition of Speakers for Consent Management by Contrastive Embedding Replay

Voice assistants overhear conversations and a consent management mechanism is required. Consent management can be implemented using speaker recognition. Users that do not give consent enrol their voice and all their further recordings are discarded. Building speaker recognition-based consent management is challenging as dynamic registration, removal, and re-registration of speakers must be efficiently handled. This work proposes a consent management system addressing the aforementioned challenges. A contrastive based training is applied to learn the underlying speaker equivariance inductive bias. The contrastive features for buckets of speakers are trained a few steps into each iteration and act as replay buffers. These features are progressively selected using a multi-strided random sampler for classification. Moreover, new methods for dynamic registration using a portion of old utterances, removal, and re-registration of speakers are proposed. The results verify memory efficiency and dynamic capabilities of the proposed methods and outperform the existing approach from the literature.

cs.SD↗

arXiv2019

Sparse Bayesian Multi-Task Learning of Time-Varying Massive MIMO Channels with Dynamic Filtering

Sparsity of channel in the next generation of wireless communication for massive multiple-input-multiple-output (MIMO) systems can be exploited to reduce the overhead in the training. The multitask (MT)-sparse Bayesian learning (SBL) is applied for learning time-varying sparse channels in the uplink for multi-user massive MIMO orthogonal frequency division multiplexing systems. In particular, the dynamic information of the sparse channel is used to initialize the hyperparameters in the MT-SBL procedure for the next time step. Then, the expectation maximization based updates are applied to estimate the underlying parameters for different subcarriers. Through the simulation studies, it is observed that using the dynamic information from the previous time step considerably reduces the complexity and the required time for the convergence of MT-SBL algorithm with negligible sacrificing of the estimation accuracy. Finally, the power leakage is reduced due to considering angular refinement in the proposed algorithm.

cs.IT↗

arXiv2018

Tracking Position and Orientation through Millimeter Wave Lens MIMO in 5G Systems

Millimeter wave signals and large antenna arrays are considered enabling technologies for future 5G networks. Despite their benefits for achieving high data rate communications, their potential advantages for tracking of the location of the user terminals are largely undiscovered. In this paper, we propose a novel support detection-based channel training method for frequency selective millimeter-wave (mm-wave) multiple-input-multiple-output system with lens antenna arrays. We show that accurate position and orientation estimation and tracking is possible using signals from a single transmitter with lens antenna arrays. Particularly, the beamspace channel estimation is formulated as two sparse signal recovery problems in the downlink and uplink for the estimation of angle-of-arrival, angle-of-departure, and time-of-arrival. The proposed method offers a higher sparse detection probability compared to the compressed sensing based solutions. Finally, a joint heuristic beamformer design and user position and orientation tracking approach are proposed based on initial estimation of channel parameters obtained in the training phase.

cs.IT↗

arXiv2017

Position and Orientation Estimation through Millimeter Wave MIMO in 5G Systems

Millimeter wave signals and large antenna arrays are considered enabling technologies for future 5G networks. While their benefits for achieving high-data rate communications are well-known, their potential advantages for accurate positioning are largely undiscovered. We derive the Cramér-Rao bound (CRB) on position and rotation angle estimation uncertainty from millimeter wave signals from a single transmitter, in the presence of scatterers. We also present a novel two-stage algorithm for position and rotation angle estimation that attains the CRB for average to high signal-to-noise ratio. The algorithm is based on multiple measurement vectors matching pursuit for coarse estimation, followed by a refinement stage based on the space-alternating generalized expectation maximization algorithm. We find that accurate position and rotation angle estimation is possible using signals from a single transmitter, in either line-of- sight, non-line-of-sight, or obstructed-line-of-sight conditions.

cs.IT↗