arXiv subjects

Sudipta Mondal

Publications and source records attributed to Sudipta Mondal.

18 recordsLinked to original sources

arXiv2026

Imaginarity-generating power of unitaries: A resource-theoretic approach

Imaginarity, stemming from the complex structure of quantum mechanics, has recently emerged as a fundamental resource, yet its dynamical generation remains largely unexplored. In this work, we introduce the notion of imaginarity-generating power (IGP) of unitary dynamics, which quantifies the ability of unitary operations to produce imaginarity from initially real quantum states. To quantify imaginarity, we employ a measure based on the Hilbert--Schmidt norm, which we show to be monotone under real unital operations. Within the framework of dynamical resource theories, we derive an exact expression for the purity-constrained IGP in arbitrary dimensions and show that, for pure real input states, it depends solely on intrinsic and experimentally accessible properties of the unitary. We further analyze its average behavior over ensembles of states with varying purity under both uniform and Hilbert--Schmidt distributions. We prove that it satisfies the essential properties of a valid resource monotone within the dynamical resource theory of imaginarity. We also characterize the unitaries that maximize the IGP and determine the corresponding bounds. Moreover, for Haar-random unitaries, we show that the IGP concentrates near its maximal value in high dimensions with small fluctuations, indicating that typical high-dimensional quantum dynamics are highly effective at generating imaginarity.

quant-ph↗

arXiv2026

Accelerating RL Post-Training Rollouts via System-Integrated Speculative Decoding

RL post-training of frontier language models is increasingly bottlenecked by autoregressive rollout generation, making rollout acceleration a central systems challenge. Many existing efficiency methods improve throughput by changing the rollout or optimization regime, for example, through off-policy execution, replay, or lower-precision generation. We study speculative decoding as a lossless acceleration primitive for RL rollouts that preserves the target model's output distribution. We implement speculative decoding in NeMo-RL with a vLLM backend, supporting both synchronous and asynchronous pipelines and enabling speculation during RL rollouts. This benefit is realizable across speculation mechanisms, such as pretrained MTP heads, small external draft models or even techniques such as Eagle3, which are traditionally applied after RL phase. This yields a deployment path for state-of-the-art speculative decoding inside RL training. In a reasoning post-training workload at 8B scale under synchronous RL, speculative decoding improves rollout throughput by 1.8x. Using a high-fidelity performance simulator, we project that combining speculative decoding with asynchronous RL yields up to 2.5x end-to-end training speedup at 235B scale.

cs.LG↗

arXiv2026

Hierarchies of Gaussian multimode entanglement from thermodynamic quantifiers

We develop a thermodynamic characterization of multimode entanglement in pure continuous-variable systems by quantifying the gap between globally and locally extractable work (ergotropy). For arbitrary pure multimode Gaussian states, we prove that the $2$-local ergotropic gap is a faithful entanglement monotone across any bipartition and constitutes a functionally independent upper bound to the Renyi-2 entanglement entropy. We further introduce the $k$-ergotropic score, the minimum $k$-local ergotropic gap, and show that it faithfully quantifies multimode entanglement across $k$ partitions. For pure three-mode Gaussian states, we derive its closed-form relation with the geometric measure for genuine multimode entanglement $(k=2)$, and total Gaussian multimode entanglement $(k=3)$. For systems with more than three modes, the $k$-ergotropic score becomes a functionally independent measure of multimode entanglement to the standard geometric measures. Our results reveal a direct operational hierarchy linking Gaussian multimode entanglement to work extraction under locality constraints, and provide a computable and experimentally accessible thermodynamic framework for characterizing quantum correlations.

quant-ph↗

arXiv2026

Hierarchies among genuine multipartite entangling capabilities of quantum gates

We classify quantum gates according to their capability to generate genuine multipartite entanglement (GME), using a hierarchy based on multipartite separable states. In particular, when a fixed unitary operator acts on the set of k-separable states, the maximal genuine multipartite entanglement content produced via that particular unitary operator is determined after maximizing over the set of k-separable input states. We identify unitary operators that are beneficial for generating high GME when the input states are entangled in some bipartition, although the picture can also be reversed, where such initial entanglement offers no advantage. We investigate the maximum entangling power of a broad range of unitary operators, encompassing special classes of quantum gates, as well as diagonal, permutation, and Haar-uniformly generated unitaries by computing generalized geometric measure (GGM) as a GME quantifier. Additionally, we observe a notable distinction in entangling power based on the nature of the input states: when maximization is restricted to separable states with real coefficients, the entangling power is lower than when the optimization is carried out over arbitrary separable states with complex coefficients, thereby highlighting the role of complex amplitudes in entanglement creation. Furthermore, we determine which unitary operators, along with their corresponding optimal inputs, yield output states with the highest achievable GGM.

quant-ph↗

arXiv2026

Classical capacities under physical constraints: More capacity with less entanglement

Current advancements in communication equipment demand the investigation of classical information transfer over quantum channels, by encompassing realistic scenarios in finite dimensions. To address this issue, we develop a framework for analyzing classical capacities of quantum channels where the set of states used for encoding information is restricted based on various physical properties. Specifically, we provide expressions for the classical capacities of noiseless and noisy quantum channels when the average energy of the encoded ensemble or the energy of each of the constituent states in the ensemble is bounded. In the case of qubit energy-preserving dephasing channels, we demonstrate that a nonuniform probability distribution based on the energy constraint maximizes capacity, while we derive the compact form of the capacity for equiprobable messages. We suggest an energy-constrained dense coding (DC) protocol that we prove to be optimal in the two-qubit situation and obtain a closed-form expression for the DC capacity. Additionally, we demonstrate a no-go result, which states that when the dimension of the sender and the receiver is two, no energy-preserving operation can offer any quantum advantage for energy-constrained entanglement-assisted capacity. We exhibit that, in the energy-constrained situation, classical-quantum noisy channels can show improved capabilities under entanglement assistance, a phenomenon that is unattainable in the unrestricted scenario.

quant-ph↗

arXiv2025

Efficient formulation of quantum network under amplitude damping noise: Highlighting benefits over its Pauli-twirled counterpart

At the heart of building a large-scale quantum internet lies the challenge of establishing long-distance entanglement using quantum repeaters, which mitigate direct transmission losses but introduce additional noise in the nodes via interactions with the environment and imperfect operations. This effect has typically been studied under a simplifying Pauli channel assumption. Our study focuses on distributing end-to-end entanglement in a homogeneous, repeater-based linear quantum network operating under a non-Pauli noise, specifically, amplitude damping noise, which we refer to as amplitude damping-affected quantum network (AQN). Unlike its twirled counterpart (TAQN), where the resulting state is fully Bell-diagonal with a single parameter, we prove that the AQN produces a block-diagonal state in the Bell basis with four parameters. We develop a method for the simulation of AQN, where we keep track of these four parameters of each entangled link, along with the number of times noise acts on it, i.e., its age, until it is consumed for swapping. Our results reveal that across diverse policies, including NESTING and SWAP-ASAP, AQN consistently outperforms TAQN in terms of both fidelity and average entanglement. The benefit is most significant in the low-probability regime of elementary link generation, highly relevant for near-term experiments. Notably, we also identify the coherence-time and link-probability regions where TAQN fails while AQN succeeds in distributing end-to-end entanglement.

quant-ph↗

arXiv2025

Imperfect Entangling Power of Quantum Gates

Achieving perfect control over the parameters defining a quantum gate is, in general, a very challenging task, and at the same time, environmental interactions can introduce disturbances to the initial states as well. Here we address the problem of how the imperfections in unitaries and noise present in the input states affect the entanglement-generating power of a given quantum gate -- we refer to it as imperfect (noisy) entangling power. We observe that, when the parameters of a given unitary are chosen randomly from a Gaussian distribution centered around the desired mean, the quenched average entangling power -- averaged across multiple random samplings -- exhibits intriguing behavior like it may increase or show nonmonotonic behavior with the increase of disorder strength for certain classes of diagonal unitary operators. For arbitrary unitary operators, the quenched average power tends to stabilize, showing almost constant behavior with variation in the parameters instead of oscillating. Our observations also reveal that, in the presence of a local noise model, the input states that maximize the entangling power of a given unitary operator differ considerably from the noiseless scenario. Additionally, we report that the rankings among unitary operators according to their entangling power in the noiseless case change depending on the noise model and noise strength.

quant-ph↗

arXiv2025

Hierarchy of entanglement detection criteria for random high-dimensional states

Entanglement is a cornerstone in quantum information science, yet detecting it efficiently remains a challenging task. Focusing on non-positive partially transposed (NPT) states, we establish a hierarchy among entropy-based, majorization, realignment, and reduction criteria for Haar uniformly generated random states in finite dimensions, analyzing their performance based on rank and subsystem dimension. We prove lower bounds on the rank of mixed quantum states beyond which the realignment and entropic criteria fail to detect entanglement. We evaluate the relative effectiveness of the considered detection methods using three key indicators -- fraction of detected states, mean detectable entanglement, and minimum required entanglement. Our results provide insights into the entanglement thresholds needed for reliable detection, showing that, beyond a certain level of entanglement, all criteria become equally powerful for low-rank states, while hierarchy among various criteria emerges with moderate to high ranks. Intriguingly, the proposed ordering among the considered criteria in qubit-qudit systems is different from that in higher dimensions. Additionally, we establish that the detection efficiency is influenced by the asymmetry in the subsystem dimensions, by illustrating how the realignment criterion behaves more efficiently than other detection methods when the difference between the subsystem dimensions is small.

quant-ph↗

arXiv2025

Recovery of resources through sequential noisy measurements

Noisy unsharp measurements incorporated in quantum information protocols may hinder performance, reducing the quantum advantage. However, we show that, unlike projective measurements which completely destroy quantum correlations between nodes in quantum networks, sequential applications of noisy measurements can mitigate the adverse impact of noise in the measurement device on quantum information processing tasks. We demonstrate this in the case of concentrating entanglement on chosen nodes in quantum networks via noisy measurements performed by assisting qubits. In the case of networks with a cluster of three or higher number of qubits, we exhibit that sequentially performing optimal unsharp measurements on the assisting qubits yields localizable entanglement between two nodes akin to that obtained by optimal projective measurements on the same assisting qubits. Furthermore, we find that the proposed approach using consecutive noisy measurements can potentially be used to prepare desired states that are resource for specific quantum schemes. We also argue that assisting qubits have greater control over the qubits on which entanglement is concentrated via unsharp measurements, in contrast to sharp measurement-based protocols, which may have implications for secure quantum communication.

quant-ph↗

arXiv2025

Disparity between multipartite entangling and disentangling powers of unitaries: Even vs Odd

We compare the multipartite entangling and disentangling powers of unitary operators by assessing their ability to generate or eliminate genuine multipartite entanglement. Our findings reveal that while diagonal unitary operators can exhibit equal entangling and disentangling powers, certain non-diagonal unitaries demonstrate an imbalance when acting on fully separable states, thereby extending the known disparity from bipartite systems to those with any number of parties. Counterintuitively, we construct classes of unitaries and their adjoints that display unequal entanglement generation capacities, behaving differently when applied to systems with an even number of qubits compared to those with an odd number. Further, we illustrate that this asymmetry can be simulated using physically realizable Hamiltonians: systems with an even number of qubits employ nearest-neighbor Dzyaloshinskii-Moriya (DM) interactions, while those with an odd number utilize a combination of Heisenberg and DM interactions. Additionally, we present a circuit composed of random noncommuting unitaries, constructed from alternating layers of two-qubit Haar-random gates, to illustrate the discrepancy in the entangling and disentangling capabilities of unitaries.

quant-ph↗

arXiv2024

More nonlocality with less incompatibility in higher dimensions: Bell vs prepare-measure scenarios

Connecting incompatibility in measurements with the violation of local realism is one of the fundamental avenues of research. For two qubits, any incompatible pair of projective measurements can violate Clauser-Horne-Shimony-Holt (CHSH) inequality for some states, and there is a monotonic relationship between the level of measurement incompatibility (projective) and the violation. However, in the case of two qutrits, we exhibit that the violation of the Collins-Gisin-Linden-Massar-Popescu (CGLMP) inequality responds non-monotonically with the amount of incompatibility; we term this more nonlocality with less incompatibility. Furthermore, unlike in the CHSH case, the maximally violating state in higher dimensions depends on the amount of measurement incompatibility. We illustrate that similar patterns can also be observed in an experimentally viable interferometric measuring technique. In such a measurement scenario, we provide an explicit example of incompatible (not jointly measurable) measurements that do not violate the CGLMP inequality for any shared quantum state. We extend our study of incompatibility in the prepare and measure scenario, focusing on quantum random access codes (QRACs). Surprisingly, we show that the monotonicity of average success probability with measurement incompatibility does not hold for higher dimensions, as opposed to two dimensions, even though the maximum probability of QRAC behaves monotonically with incompatibility.

quant-ph↗

arXiv2023

Duality between imperfect resources and measurements for propagating entanglement in networks

We propose a measurement-based entanglement propagation strategy for networks in which all nodes except two are initially occupied by a suitably chosen single-qubit system and the two nodes share a bipartite noisy entangled state. The connections between the sites are established using unsharp two-qubit measurements. When only a single node performs measurements, we refer to it as a unidirectional protocol while when both parts of the initial entangled states perform measurements, we call it a bidirectional scheme. When the measurement outcome is post-selected, we demonstrate that in the presence of a local amplitude damping channel acting on a single site, entanglement shareability, as measured by the monogamy score, of the resulting state after measurement can be higher for all values of the strength of the noise than that of the scenario without noise. We observe that irrespective of the channel, there exists a range of unsharpness parameter where a higher monogamy score may be obtained starting from the initial nonmaximally entangled states than from the initial maximally entangled state. We report that the effect of noise on the average monogamy score entered from the resource state may be reduced faster with the unidirectional protocol than with the bidirectional one.

quant-ph↗

arXiv2023

Interpretable Bangla Sarcasm Detection using BERT and Explainable AI

A positive phrase or a sentence with an underlying negative motive is usually defined as sarcasm that is widely used in today's social media platforms such as Facebook, Twitter, Reddit, etc. In recent times active users in social media platforms are increasing dramatically which raises the need for an automated NLP-based system that can be utilized in various tasks such as determining market demand, sentiment analysis, threat detection, etc. However, since sarcasm usually implies the opposite meaning and its detection is frequently a challenging issue, data meaning extraction through an NLP-based model becomes more complicated. As a result, there has been a lot of study on sarcasm detection in English over the past several years, and there's been a noticeable improvement and yet sarcasm detection in the Bangla language's state remains the same. In this article, we present a BERT-based system that can achieve 99.60\% while the utilized traditional machine learning algorithms are only capable of achieving 89.93\%. Additionally, we have employed Local Interpretable Model-Agnostic Explanations that introduce explainability to our system. Moreover, we have utilized a newly collected bangla sarcasm dataset, BanglaSarc that was constructed specifically for the evaluation of this study. This dataset consists of fresh records of sarcastic and non-sarcastic comments, the majority of which are acquired from Facebook and YouTube comment sections.

cs.CL↗

arXiv2023

Controlled transition to different proton acceleration regimes: near-critical density plasmas driven by circularly polarized few cycle pulse

We investigate the different facets of ion acceleration by a relativistically intense circularly polarized laser pulse interacting with thin near-critical density plasma targets. Our simulations establish that plasma density gradient and laser frequency chirp can be controlled to switch the interaction from the transparent to the opaque regimes of operation. This enables one to choose between a Maxwellian like ion energy distribution with a cut-off energy, in the relativistically transparent regime, or a quasi-monoenergetic spectrum, in the opaque regime. We subsequently demonstrate that a double-layer multi-species target configuration, can be effectively utilized for optimal generation of quasi mono-energetic ion bunches of a desired species. We finally demonstrate, the feasibility of generating mono-energetic proton beams with energy peak at $\mathcal{E}\approx20\sim40$ MeV with a narrow energy spread of $Δ\mathcal{E}/\mathcal{E}\approx18-28.6\%$ confined within a divergence angle of $\sim 175$ millirad at a reasonable laser peak intensity of $I_{0}\simeq 5.4\times 10^{20}\, \mathrm{W/cm^2}$.

physics.plasm-ph↗

arXiv2022

Ion acceleration with few cycle relativistic laser pulses from foil targets

Ion acceleration resulting from the interaction of 11 fs laser pulses of ~35 mJ energy with ultrahigh contrast (<10^-10), and 10^19 W/cm^2 peak intensity with foil targets made of various materials and thicknesses at normal (0-degree) and 45-degree laser incidence is investigated. The maximum energy of the protons accelerated from both the rear and front sides of the target was above 1 MeV. A conversion efficiency from laser pulse energy to proton beam is estimated to be as high as ~1.4 % at 45-degree laser incidence using a 51 nm-thick Al target. The excellent laser contrast indicates the predominance of vacuum heating via the Brunels effect as an absorption mechanism involving a tiny pre-plasma of natural origin due to the Gaussian temporal laser pulse shape. Experimental results are in reasonable agreement with theoretical estimates where proton acceleration from the target rear into the forward direction is well explained by a TNSA-like mechanism, while proton acceleration from the target front into the backward direction can be explained by the formation of a charged cavity in a tiny pre-plasma. The exploding Coulomb field from the charged cavity also serves as a source for forward-accelerated ions at thick targets.

physics.plasm-ph↗

arXiv2021

GNNIE: GNN Inference Engine with Load-balancing and Graph-Specific Caching

Graph neural networks (GNN) analysis engines are vital for real-world problems that use large graph models. Challenges for a GNN hardware platform include the ability to (a) host a variety of GNNs, (b) handle high sparsity in input vertex feature vectors and the graph adjacency matrix and the accompanying random memory access patterns, and (c) maintain load-balanced computation in the face of uneven workloads, induced by high sparsity and power-law vertex degree distributions. This paper proposes GNNIE, an accelerator designed to run a broad range of GNNs. It tackles workload imbalance by (i)~splitting vertex feature operands into blocks, (ii)~reordering and redistributing computations, (iii)~using a novel flexible MAC architecture. It adopts a graph-specific, degree-aware caching policy that is well suited to real-world graph characteristics. The policy enhances on-chip data reuse and avoids random memory access to DRAM. GNNIE achieves average speedups of 21233x over a CPU and 699x over a GPU over multiple datasets on graph attention networks (GATs), graph convolutional networks (GCNs), GraphSAGE, GINConv, and DiffPool. Compared to prior approaches, GNNIE achieves an average speedup of 35x over HyGCN (which cannot implement GATs) for GCN, GraphSAGE, and GINConv, and, using 3.4x fewer processing units, an average speedup of 2.1x over AWB-GCN (which runs only GCNs).

cs.AR↗

arXiv2019

Physical properties of RIr3 (R = Gd, Tb, Ho) compounds with coexisting polymorphic phases

The binary compounds GdIr3, TbIr3 and HoIr3 are synthesized successfully and found to form in macroscopic co-existence of two polymorphic phases: C15b and AuCu3-type. The dc magnetization and heat capacity studies confirm that C15b phase orders ferromagnetically, whereas the AuCu3 phase remains paramagnetic down to 2 K. The frequency dependent ac-susceptibility data, time dependent magnetic relaxation behavior and magnetic memory effect studies suggest that TbIr3 and HoIr3 are cannonical spin-glass system, but no glassy feature could be found in GdIr3. The critical behavior of all the three compounds has been investigated from the magnetization and heat capacity measurements around the transition temperature (TC). The critical exponents alpha, beta, gamma and delta have been estimated using different techniques such as Arrott-Noaks plot, Kouvel-Fisher plot, critical isotherm as well as analysis of specific heat data and study of magnetocaloric effect. The critical analysis study identifies the type of universal magnetic class in which the three compounds belong.

cond-mat.mtrl-sci↗

arXiv2010

A bright point source of ultrashort hard x-rays from laser bioplasmas

Micro and nano structures scatter light and amplify local electric fields very effectively. Energy incident as intense ultrashort laser pulses can be converted to x-rays and hot electrons more efficiently with a substrate that suitably modifies the local fields. Here we demonstrate that coating a plain glass surface with a few micron thick layer of an ubiquitous microbe, {\it Escherichia coli}, catapults the brightness of hard x-ray bremsstrahlung emission (up to 300 keV) by more than two orders of magnitude at an incident laser intensity of 10$^{16}$ W cm$^{-2}$. This increased yield is attributed to the local enhancement of electric fields around individual {\it E. coli} cells and is reproduced by detailed particle-in-cell (PIC) simulations. This combination of laser plasmas and biological targets can lead to turnkey, multi-kilohertz and environmentally safe sources of hard x-rays.

physics.optics↗