arXiv subjects

Chen Yuan

Publications and source records attributed to Chen Yuan.

At least 19 recordsLinked to original sources

arXiv2026

On Proximity Gaps for One-Point Algebraic Geometric Codes

Proximity gaps and correlated agreement underpin the soundness of interactive oracle proofs. We establish these properties for general one-point algebraic geometry (AG) codes, extending corresponding results for Reed-Solomon codes. For codes of length $n$ over $\mathbb F_q$ with designed distance $n-m$, write $ρ=m/n$ and fix $ρ\in(0,1)$. For polynomial curves of degree at most $\ell$ and any $0<\varepsilon<1$, we prove correlated agreement with multiplicative slack $\varepsilon$ and soundness $O(\ell(\ell+1)/(\varepsilon q))$ whenever $e\le(1-\varepsilon)(n-m)/2$, with no additional genus penalty. This gives asymptotically good AG code families from Garcia--Stichtenoth tower of unbounded length over a fixed field with soundness bounds independent of $n$, beyond the RS restriction $n\le q$. We also establish mutual correlated agreement up to radius $(1-\sqrtρ-η)n$, with soundness $O(\ell n/(η^3q))$ for $0<η<1-\sqrtρ$. Our half designed distance proof combines auxiliary divisors, simultaneous Berlekamp-Welch interpolation, and error-incidence counting to obtain a common polynomial representation without a Polishchuk-Spielman divisibility lemma. Finally, two counterexamples on the Garcia-Stichtenoth tower show limitations beyond half the minimum distance, at radii below the Johnson bound and with positive slack. The first requires a positive constant soundness lower bound independent of $q$ and $n$, precluding a uniform $O(1/q)$ guarantee throughout the Johnson regime. The second fixes the field and lets the relative gap above half distance tend to zero while ruling out correlated agreement for every soundness parameter $σ<1$, with positive limiting rate and relative error radius.

cs.IT↗

arXiv2026

List Size of Random Linear Codes: Optimality and the Limits of Average-Radius List Decoding

Guruswami, Li, Mosheiff, Resch, Silas and Wootters conjectured that, for every fixed prime power $q$ and $p\in(0,1-1/q)$, random linear codes of rate $1-h_q(p)-\varepsilon$ have list size $\frac{h_q(p)}{\varepsilon}\cdot(1+o(1))$ as $\varepsilon\to0$, both for list decoding and average-radius list decoding. The conjecture was known for $q=2$ and remained open for larger alphabets. We prove the conjecture for list decoding for every prime power $q$, with the sharper bound $L=\frac{h_q(p)}{\varepsilon}+O_{p,q}(1).$ In contrast, we disprove the conjecture for average-radius list decoding. We show that the required list-size leading constant is at least $p\left(1+\log_q\frac{(q-1)(1-p)}{p}\right)$, which is strictly larger than $h_q(p)$ whenever $q^p(1-p)>1$. Moreover, for $p=1-λ/q$ with fixed $λ>1$, the required leading constant can be as large as $\left(\frac{\logλ}{eλ}+o_q(1)\right)\frac{q}{\log q}.$

cs.IT↗

arXiv2026

Stochastic gravitational-wave background from self-interacting superradiant clouds

Gravitational-wave (GW) observations offer a powerful probe of new fundamental fields. One well-motivated source is black hole (BH)--boson cloud systems, in which an ultralight scalar field forms a cloud around a rotating BH via superradiance and emits long-lived nearly monochromatic GWs. In this work, we compute the stochastic GW background (SGWB) from such systems, extending previous work by including scalar self-interactions, and discuss its detectability with next-generation ground-based GW detectors. We find that self-interactions can suppress the SGWB and thereby relax existing LIGO-Virgo-KAGRA constraints inferred from null searches. Namely, the LIGO detectors at design sensitivity are insensitive to a SGWB produced by scalar fields with a decay constant $f_\mathrm{s} \lesssim 3\times 10^{17}$ GeV, independently of the scalar field mass. Looking ahead, we show that a moderately self-interacting cloud can still produce a detectable SGWB with next-generation detectors. Under conservative assumptions, for a decay constant $f_\mathrm{s} = 10^{17}$ GeV, the Einstein Telescope (ET) will be sensitive to scalar masses in the range $\sim[10^{-13.0},10^{-11.8}]$ eV, while Cosmic Explorer (CE) will be sensitive to scalar masses in the range $\sim[10^{-13.2},10^{-11.7}]$ eV. We also find that the minimum decay constants that still yield a detectable SGWB for ET and CE are $f_\mathrm{s}\sim 6\times10^{16}\,$GeV and $\sim 3\times10^{16}\,$GeV, respectively.

gr-qc↗

arXiv2026

Thresholds for Local Coordinate-Wise Linear Properties of Random One-Point AG Codes

Local coordinate-wise linear (LCL) properties provide a unified framework for studying list decoding, list recovery and other local properties of linear codes. We prove that random one-point algebraic geometry (AG) codes have the same LCL threshold rates as random linear codes under explicit alphabet and sampling conditions. More precisely, for a family of local profiles with random linear code threshold $R_{\mathcal P}$, a random one-point AG code avoids all such profiles below $R_{\mathcal P}-\varepsilon$ and contains one above $R_{\mathcal P}+\varepsilon$, with explicit failure bounds on both sides. Our proof also removes an auxiliary lower bound on the number of available rational places that arises in a direct adaptation of the Reed--Solomon argument and requires no additional genus-dependent rate gap. We also observe that the previously claimed above threshold proof for random Reed--Solomon codes does not establish the required probability lower bound. The containment relation used in that argument gives the probability comparison in the opposite direction from what is needed. We give a different above threshold argument, which in particular also yields an alternative proof in the genus zero case. Combining our threshold theorem with recent results for random linear codes gives a collection of new list decoding, average weight list decoding and list recovery guarantees for random one-point AG codes. We further obtain new correlated agreement and proximity gap results for polynomial curves. These results extend the LCL threshold framework from random linear and Reed--Solomon codes to random one-point AG codes.

cs.IT↗

arXiv2026

Constraining the Swift Memory Burden Effect with GW250114-like Events

Black hole spectroscopy allows to infer the properties of the remnant of a binary black hole coalescence. Motivated by the recent proposal that a black hole's information load can alter its classical response to small perturbations, an effect known as the swift memory burden (SMB), we develop a minimal phenomenological framework to analyze the ringdown of a binary black hole merger and confront it with the data from the GW250114 event. In particular, we model the SMB imprint through a suppression factor derived as an upper bound and adopt the least-suppressive form. We perform a Bayesian analysis combining the frequencies of the (220) and (440) quasi-normal modes and obtain a lower bound $\log_{10}p \gtrsim 2$, where $p$ controls how the gaps reopen when the black hole's master mode occupation departs from the critical value. Moreover, using a Fisher information matrix (high signal-to-noise ratio) approximation, we forecast the lower bound $\log_{10}p \gtrsim 3$ for a GW250114-like event observed with Cosmic Explorer or Einstein Telescope. Within this minimal, weak-effect framework, our results disfavour rapid gap reopening, shedding light on how the swift memory burden effect can be probed with current and next-generation detectors.

gr-qc↗

arXiv2026

FLAT: Resampling Image and Text into 1D Flexible-Length Aligned Transmodal Tokens for Retrieval and Generation

Traditional multimodal representation learning and generation are two stages: a contrastive or self-supervised visual encoder is trained first, followed by a separate downstream generative model. This setup bottlenecks generative performance behind frozen embeddings. To bridge this gap, we revisit joint multimodal representation learning and generation to produce linearly interpolatable embeddings that are directly consumable by generative decoders. We present FLAT (Flexible-Length Aligned Transmodal representations), a representation pre-training framework that jointly optimizes a shared multimodal encoder alongside downstream text-to-image (T2I) and image-to-text (I2T) decoders. By combining contrastive alignment with bidirectional cross-modal generative objectives, FLAT ensures its representations function as both discriminative semantic descriptors and generative conditions. Architecturally, FLAT maps visual and textual inputs into a unified continuous 1D sequence space, applying nested dropout over prefix-K tokens to enable dynamic output lengths. A single pre-training stage allows FLAT to perform cross-modal retrieval and generation across variable prefix K, achieving a T2I GenEval score of 71.1. Task-specific fine-tuning aligns model performance with state-of-the-art baselines: 83.1 GenEval on T2I generation; 40.5 BLEU-4 and 138.6 CIDEr on MS-COCO image captioning; and Recall@5 scores of 86.8 (I2T) / 75.8 (T2I) on MS-COCO alongside 98.3 (I2T) / 93.6 (T2I) on Flickr30K. Finally, qualitative evaluations demonstrate that FLAT representations natively support linear interpolation, latent space arithmetic, and zero-shot composed retrieval.

cs.CV↗

arXiv2026

Infrared Universality of Scalar Induced Gravitational Waves Beyond Second Order

Primordial black hole dark matter is formed from the collapse of enhanced primordial density fluctuations. The formation of primordial black holes is accompanied by scalar induced gravitational waves whose infrared scaling only depends on the equation of state, $w$, of the background at leading order. Establishing whether this characteristic scaling remains robust under corrections of arbitrary perturbative order, including in the presence of non-Gaussianities, requires a physical understanding of the infrared scaling. In this work, we provide such a physical interpretation. We find that the stress of sound waves with nearly equal frequencies produces a slow beat. Although the incoming waves oscillate rapidly, this slow beat continues to source the long wavelength tensor perturbation before its horizon entry, thus generating an infrared scaling that only depends on the expanding background. We show that higher order interactions generate freely propagating waves that finally lead to the same slow beat deep inside the horizon. Since non-Gaussianities would not change the speed of the freely propagating waves, they do not modify the infrared scaling. As the source with a slow beat has enough time to accumulate before the horizon entry of the infrared tensor mode, it records the background expansion and thus the infrared scaling could provide a robust probe of the cosmological equation of state.

astro-ph.CO↗

arXiv2026

Proximity Gaps for Gabidulin Codes and Applications

Proximity gaps are central to the soundness of interactive oracle proofs of proximity (IOPPs) and polynomial commitment schemes (PCSs). An $[n,k,d]$ linear code $C\subseteq\mathbb F^n$ has a $δ$-proximity gap with error $ε$ if, for every $u_0,u_1\in\mathbb F^n$, either all points on $\ell_{u_0,u_1}=\{u_0+αu_1:α\in\mathbb F\}$ are $δ$-close to $C$, or at most an $ε$ fraction are. Although proximity gaps for Hamming-metric codes are well understood, their rank-metric counterparts remain largely unexplored despite their applications in coding theory and cryptography. In this work, we study proximity gaps for linear rank-metric codes and their cryptographic applications. First, we show that every $[n,k,d]$ linear rank-metric code $C$ over $\mathbb F_{q^m}$ admits a proximity gap for every $δ\le(d-1)/(3n)$, with error at most $q^{e+1}/q^m$, where $e=\lfloorδn\rfloor$. For Gabidulin codes, we improve the gap to $(d-1)/(2n)$ with error $10q^{n-1}/q^m$. These two proximity gaps match those for general linear Hamming-metric codes and Reed--Solomon (RS) codes, respectively. We prove the $(d-1)/(2n)$ bound is tight by constructing an infinite family of constant-rate Gabidulin codes and affine lines $\ell_{u_0,u_1}$ on which a $1-o(1)$ fraction of points are $d/(2n)$-close to the code, while $u_1$ is at least $3d/(4n)$-far from it. At the $d/(3n)$ gap, we also give a counterexample establishing a lower bound on $ε$. As applications, we construct an IOPP for interleaved Gabidulin codes by adapting the Ligero IOPP for interleaved RS codes. We then adapt the Ligero-based PCS for ordinary polynomials to obtain a $q$-linearized polynomial commitment scheme. To our knowledge, this is the first PCS framework based on rank-metric error-correcting codes.

cs.IT↗

arXiv2026

Projected Constraints on Primordial Black Holes from Scalar-Induced Gravitational Waves with Taiji

Scalar-induced gravitational waves (SIGWs) provide a direct probe of the enhanced primordial curvature perturbations that may also produce primordial black holes (PBHs). We forecast the capability of the space-based gravitational-wave observatory Taiji to search for an SIGW background generated by a broken-power-law curvature spectrum. A signal-injection study is used to validate the analysis pipeline, after which a pure-noise realization is employed to derive projected upper limits on the curvature-spectrum parameters. We translate these limits into constraints on the PBH abundance using the nonlinear compaction function, critical collapse, and the joint Gaussian distribution of the compaction amplitude and curvature at its peak. The resulting projected $95\%$ upper limits on the PBH dark-matter fraction satisfy $f_{\mathrm{PBH}}^{95\%}<1$ over PBH masses from approximately $6.2\times10^{-18}\,M_\odot$ to $1.7\times10^{-8}\,M_\odot$. We compare the forecast with representative Hawking-evaporation and microlensing bounds. In part of the asteroid-mass interval, the projected Taiji limit is more restrictive than the current Subaru Hyper Suprime-Cam (HSC) microlensing constraint.

astro-ph.CO↗

arXiv2026

Random Gabidulin Codes Achieve List Decoding Capacity in the Rank Metric

Gabidulin codes, serving as the rank-metric counterpart of Reed-Solomon codes, constitute an important class of maximum rank distance (MRD) codes. However, unlike the fruitful positive results about the list decoding of Reed-Solomon codes, results concerning the list decodability of Gabidulin codes in the rank metric are all negative so far. For example, in contrast to Reed-Solomon codes, which are always list decodable up to the Johnson bound in the Hamming metric, Raviv and Wachter-Zeh (IEEE TIT, 2016 and 2017) constructed a class of Gabidulin codes that are not even combinatorially list decodable beyond the unique decoding radius in the rank metric. Proving the existence of Gabidulin codes with good combinatorial list decodability in the rank metric has remained a long-standing open problem. In this paper, we resolve the aforementioned open problem by showing that, with high probability, random Gabidulin codes over sufficiently large alphabets attain the optimal generalized Singleton bound for list decoding in the rank metric. In particular, they achieve list decoding capacity in the rank metric. Our work is significantly influenced by the recent breakthroughs in the combinatorial list decodability of Reed-Solomon codes, especially the work by Brakensiek, Gopi, and Makam (STOC 2023). Our major technical contributions, which may hold independent interest, consist of the following: (1) We initiate the study of ``higher order MRD codes'' and provide a novel unified theory, which runs parallel to the theory of ``higher order MDS codes'' developed by BGM. (2) We prove a natural analog of the GM-MDS theorem, proven by Lovett (FOCS 2018) and Yildiz and Hassibi (IEEE TIT, 2019), which we call the GM-MRD theorem. In particular, our GM-MRD theorem for Gabidulin codes are strictly stronger than the GM-MDS theorem for Gabidulin codes, proven by Yildiz and Hassibi (IEEE TIT, 2019).

cs.IT↗

arXiv2026

List Decoding of Reed-Solomon Codes and Folded Reed-Solomon Codes Over Galois Ring

List decoding of codes can be seen as the generalization of unique decoding of codes while list decoding over finite fields has been extensively studied, extending these results to more general algebraic structures such as Galois rings remains an important challenge. Due to recent progress in zero knowledge systems, there is a growing demand to investigate the proximity gap of codes over Galois rings in Yizhou Yao(2025). The proximity gap is closely related to the decoding capability of codes. It was shown in Eli Ben-Sasson(2020) that the proximity gap for RS codes over finite field can be improved to $1-\sqrt{r}$ if one consider list decoding instead of unique decoding. However, we know very little about RS codes over Galois ring which might hinder the development of zero knowledge proof system for ring-based arithmetic circuit. In this work, we first extend the list decoding procedure of Guruswami and Sudan to Reed-Solomon codes over Galois rings, which shows that RS codes with rate $r$ can be list decoded up to radius $1-\sqrt{r}$. Then, we investigate the list decoding of folded Reed-Solomon codes over Galois rings. We show that the list decoding radius of folded Reed-Solomon codes can reach the Singlton bound as its counterpart over finite field. We also extend the deterministic pruning method of Vikrant Ashvinkumar(2026) to Galois rings, showing how to prune the affine free module obtained from the linear-algebraic decoder and recover the candidate codewords. Finally, we improve the list size of our folded Reed-Solomon code to $O(1/\varepsilon^2)$ by extending recent work in Shashank Srivastava(2025) to Galois Rings. By developing the recent work of Yeyuan Chen(2025), we show that folded Reed-Solomon codes over Galois rings satisfy the relaxed generalized Singleton bound in the average-radius sense with optimal list size $O(1/\varepsilon)$.

cs.IT↗

arXiv2026

A Syndrome--Space Approach to Proximity Gaps and Correlated Agreement for Random Linear Codes and Random Reed--Solomon Codes

Proximity gaps and correlated agreement have become central tools in the analysis of interactive oracle proofs of proximity (IOPPs) and code-based SNARKs. Informally, a proximity-gap statement says that for a structured set of words -- such as an affine space, or a curve -- either all points are close to the code, or most are far from it. Such statements are essential in sampling-based proof systems, where a verifier queries only a few random locations on a structured object but must still obtain a global soundness guarantee. In Reed--Solomon-based proof systems, one would ideally like the proximity parameter to approach the information-theoretic limit $1-R$, since this is the largest possible radius for a rate-$R$ code and directly affects protocol efficiency. We establish a direct approach to proximity gaps and correlated agreement for random linear codes in the random parity-check-matrix model, without relying on list decoding of the proof. Our approach is based on a syndrome-space reformulation together with a witness-based reduction argument. It is conceptually different from the existing decoding-driven route for random linear codes, and it also leads to sharper parameters, including the optimal-up-to-$\varepsilon$ large-alphabet radius bound $ρ<1-R-\varepsilon$ for $q=Θ(n)$, as well as near-capacity bounds over constant alphabets with improved alphabet-size requirements. We apply the same syndrome-space reductions to random Reed--Solomon codes. This yields correlated agreement for random Reed--Solomon codes over affine spaces and polynomial curves up to radius $ρ\le 1-R-\varepsilon$, with field size $q\ge n\cdot 2^{O(\varepsilon^{-3})}$ for affine spaces and $q\ge n\cdot 2^{O_\ell(\varepsilon^{-3})}$ for degree-$\ell$ curves.

cs.IT↗

arXiv2026

Resonances as signatures of scalar clouds in eccentric extreme-mass-ratio inspirals

Ultralight scalars arise naturally in many extensions to the Standard Model and are compelling dark matter candidates. Around spinning black holes, dense scalar clouds could form through the conversion of rotational energy into particles via black hole superradiance. Extreme-mass-ratio inspirals (EMRIs) targeted by future space-based detectors will give us unparalleled access to the environments of massive black holes, allowing us to probe the presence of scalar clouds. We consider EMRIs around a Schwarzschild black hole and show that eccentricity induces a dense sequence of resonances in the scalar fluxes near the last stable orbit. These resonances arise only in a fully relativistic treatment, as they are intrinsically tied to the splitting between the azimuthal and radial orbital frequencies in the strong-field regime. By evolving the orbits adiabatically, we show that the resulting resonant transitions substantially enhance the exchange of energy and angular momentum between the EMRI and the scalar cloud, significantly amplifying the accumulated dephasing in the gravitational waveform relative to circular motion. Our results highlight the importance of eccentricity in shaping the observational signatures of EMRIs embedded in scalar clouds.

gr-qc↗

arXiv2026

Relativistic effects in extreme-mass-ratio inspirals within scalar clouds: Eccentric and inclined orbits

We study extreme-mass-ratio inspirals (EMRIs) evolving in a scalar cloud environment that may form through superradiant instabilities, using a fully relativistic perturbative framework that allows for eccentric and inclined orbits. EMRIs, consisting of a stellar-mass compact object inspiraling into a supermassive black hole, are key sources for space-based gravitational-wave detectors such as LISA. Previous relativistic studies of EMRIs in scalar clouds have been restricted to circular, equatorial motion. Here, instead, we focus on a Schwarzschild black hole background to incorporate eccentricity and orbital inclination. By computing the scalar energy and angular momentum scattered off to spatial infinity and absorbed at the event horizon, we show that orbital eccentricity can induce a dense spectrum of resonances near the last stable orbit, associated with strong relativistic apsidal precession. We further find that orbital inclination can significantly modify the orbital energy and angular momentum losses. In particular, we identify a critical inclination angle below which, at sufficiently small orbital radii, there is a net transfer of energy from the scalar cloud to the orbit. Moreover, for sufficiently large eccentricities, resonances associated with relativistic apsidal precession persist across the full range of inclinations, although their structure changes significantly between prograde and retrograde orbits. These results provide a foundation for future studies of EMRIs in scalar cloud environments on fully generic orbits around spinning black holes.

gr-qc↗

arXiv2026

One-Loop Tensor Power Spectrum from a Non-Canonical Spectator Field during Inflation

We compute the full one-loop corrections to the primordial tensor power spectrum in an inflationary scenario with a {non-canonical spectator field}, using the in-in formalism. We derive semi-analytic results for the scalar-sourced one-loop tensor spectrum and the effective tensor-to-scalar ratio, $r_{\mathrm{eff}}$. We consider two representative coupling functions: a localized Gaussian dip (Model G), which leads to moderate loop corrections, and a rapidly oscillatory coupling (Model O), which can yield much larger loop contributions. For Model G, we find a $\mathcal{O}(1)$ correction to $r_{\mathrm{eff}}$ while Model O can significantly enhance $r_{\mathrm{eff}}$ by several orders of magnitude (relative to the tree-level value). We further calculate the energy density of primordial gravitational waves. Assuming that primordial black holes with mass $10^{-12}M_{\odot}$ generated in this scenario, constitute all of the dark matter, we find that the results are several orders of magnitude lower than the sensitivities of Taiji/TianQin/LISA.

astro-ph.CO↗

arXiv2026

Explicit List-Decodable Linearized Reed-Solomon and Folded Linearized Reed-Solomon Subcodes

The sum-rank metric is the mixture of the Hamming and rank metrics. The sum-rank metric found its application in network coding, locally repairable codes, space-time coding, and quantum-resistant cryptography. Linearized Reed-Solomon (LRS) codes are the sum-rank analogue of Reed-Solomon codes and strictly generalize both Reed-Solomon and Gabidulin codes. In this work, we construct an explicit family of $\mathbb{F}_h$-linear sum-rank metric codes over arbitrary fields $\mathbb{F}_h$. Our construction enables efficient list decoding up to a fraction $ρ$ of errors in the sum-rank metric with rate $1-ρ-\varepsilon$, for any desired $ρ\in (0,1)$ and $\varepsilon>0$. Our codes are subcodes of LRS codes, obtained by restricting message polynomials to an $\mathbb{F}_h$-subspace derived from subspace designs, and the decoding list size is bounded by $h^{\mathrm{poly}(1/\varepsilon)}$. Beyond the standard LRS setting, we further extend our linear-algebraic decoding framework to folded Linearized Reed-Solomon (FLRS) codes. We show that folded evaluations satisfy appropriate interpolation conditions and that the corresponding solution space forms a low-dimensional, structured affine subspace. This structure enables effective control of the list size and yields the first explicit positive-rate FLRS subcodes that are efficiently list decodable beyond the unique-decoding radius. To the best of our knowledge, this also constitutes the first explicit construction of positive-rate sum-rank metric codes that admit efficient list decoding beyond the unique decoding radius, thereby providing a new general framework for constructing efficiently decodable codes under the sum-rank metric.

cs.IT↗

arXiv2026

Improvement of the Gilbert-Varshamov Bound for Linear Codes and Quantum Codes

The Gilbert--Varshamov (GV) bound is a central benchmark in coding theory, establishing existential guarantees for error-correcting codes and serving as a baseline for both Hamming and quantum fault-tolerant information processing. Despite decades of effort, improving the GV bound is notoriously difficult, and known improvements often rely on technically heavy arguments and do not extend naturally to the quantum setting due to additional self-orthogonality constraints. In this work we develop a concise probabilistic method that yields an improvement over the classical GV bound for $q$-ary linear codes. For relative distance $δ=d/n<1-1/q$, we show that an $[n,k,d]_q$ linear code exists whenever $\frac{q^{k}-1}{q-1}\;<\;\frac{c_δ\sqrt{n}\, q^{n}}{\mathrm{Vol}_q(n,d-1)}$, for positive constant $c_δ$ depending only on $δ$, where $\mathrm{Vol}_q(n,d-1)$ denotes the volume of a $q$-ary Hamming ball. We further adapt this approach to the quantum setting by analyzing symplectic self-orthogonal structures. For $δ<1-1/q^2$, we obtain an improved quantum GV bound: there exists a $q$-ary quantum code $[[n,\,n-k,\,d]]$ provided that $\frac{q^{2n-k}-1}{q-1}<\frac{c_δ\sqrt{n}\cdot q^{2n}}{\sum_{i=0}^{d-1}\binom{n}{i}(q^2-1)^i}$. In particular, our result improves the standard quantum GV bound by an $Ω(\sqrt{n})$ multiplicative factor.

cs.IT↗

arXiv2026

Efficient Sequential Recommendation for Long Term User Interest Via Personalization

Recent years have witnessed success of sequential modeling, generative recommender, and large language model for recommendation. Though the scaling law has been validated for sequential models, it showed inefficiency in computational capacity when considering real-world applications like recommendation, due to the non-linear(quadratic) increasing nature of the transformer model. To improve the efficiency of the sequential model, we introduced a novel approach to sequential recommendation that leverages personalization techniques to enhance efficiency and performance. Our method compresses long user interaction histories into learnable tokens, which are then combined with recent interactions to generate recommendations. This approach significantly reduces computational costs while maintaining high recommendation accuracy. Our method could be applied to existing transformer based recommendation models, e.g., HSTU and HLLM. Extensive experiments on multiple sequential models demonstrate its versatility and effectiveness. Source code is available at \href{https://github.com/facebookresearch/PerSRec}{https://github.com/facebookresearch/PerSRec}.

cs.IR↗