arXiv subjects
Publications and source records attributed to Yiming Chen.
Let $M_n$ be an $n\times n$ matrix with independent uniform sign entries. We prove that there exist absolute constants $C,c>0$ such that, for all sufficiently large $n$, \[ \mathbb{P}\!\left( \left|\operatorname{Per}(M_n)\right| \ge e^{-Cn}\sqrt{n!} \right) \ge 1-n^{-c}. \] This confirms, up to the exponential scale, the lower bound suggested by Tao and Vu.
Exotic spin-spin-velocity-dependent interactions, predicted in extensions of the Standard Model involving new bosonic fields, could resolve fundamental puzzles from dark matter to cosmic asymmetry. However, exploring these weak potential interactions at centimeter scales presents formidable challenges, primarily due to the overwhelming dominance of electromagnetic backgrounds that can easily obscure the weak exotic signals. Here, we utilize a levitated magnet force sensor with ultrahigh electron spin density to probe these interactions. We constrain two interactions individually through a designed spin source and a multi-layer magnetic shielding system that suppresses electromagnetic backgrounds. In this study, we constrain two types of interactions: the V_6 potential at force ranges from $10^{-3}$ m to $6 \times 10^{-2}$ m and the V_{14} potential at ranges greater than $10^{-3}$ m. Our measurements establish 95% confidence-level bounds of $|f_6| \leq 2.12 \times 10^{-13}$ and $|f_{14}| \leq 2.34 \times 10^{-23}$ at $\lambda = 1.6 \times 10^{-2}$ m, improving prior limits by up to 12 and 13 orders of magnitude, respectively. Our result demonstrates the levitated magnet as a highly sensitive probe for detecting new bosonic fields in extensions of the Standard Model.
This paper reconsiders intelligence from a vision-centered perspective and examines whether intelligence emerging from visual experience and learning may provide a pathway toward AGI. In the language domain, beginning with the introduction of the Transformer architecture, the GPT series has demonstrated transfer to unseen tasks through autoregressive language modeling on web-scale text combined with aggressive scaling. This raises a natural question, namely, what capabilities and forms of intelligence can emerge from visual modalities such as images, videos, and geometry? In this paper, we discuss whether visual intelligence can serve as a pathway toward AGI, referred to in this paper as visual general intelligence (VGI), by bringing together contributors from diverse standpoints and affiliations. Our aim is not to offer a single definition of visual intelligence, but to clarify the principles that computer vision should pursue in the AGI era, the visual input modalities, the benchmarks, the learning paradigms, and the relationship between vision, when taken as the core, and other modalities such as language.
The Schubert vanishing problem asks whether the single Schubert coefficients $c_{u,v}^w$ are zero. In this paper, we consider the non-vanishing problems of double Schubert coefficients $c_{u,v}^w(t)$ and triple Schubert coefficients $c_{u,v}^w(t;y)$. We show that the non-vanishing of $c_{u,v}^w(t;y)$ is completely determined by the non-vanishing of single Schubert coefficients. As a byproduct, we obtain the saturation property of the triple Littlewood--Richardson coefficients $c_{\lambda,\mu}^\nu(t;y)$. Moreover, we pose a conjecture asserting that the non-vanishing of $c_{u,v}^w(t)$ is also determined by the non-vanishing of single or triple Schubert coefficients. We prove a one-side inclusion of the conjecture. For the reverse inclusion, we show that the conjecture holds for the following three cases: the Pieri case, the separated descents case, and the inverse Grassmannian case.
Fuzzing Rust library APIs requires constructing well-typed, compilable call sequences that satisfy ownership rules, generic parameters, and trait bounds; existing tools ignore these constraints or use shallow heuristics, yielding low coverage. We present GRAFT, which extracts structured API information from Rust documentation, builds an API dependency graph via recursive generics-aware type matching, and uses topology-guided traversal plus LLM synthesis with compiler-error feedback to produce compilable fuzz targets. On 13 crates from crates.io, GRAFT achieves 80.75% macro-average API coverage at 96.19% compilation success, outperforming RULF and RPG by 4.76x and 2.43x, and reaching 1.41x the average API coverage of deepSURF on crates with unsafe-reaching APIs.
Levitated mechanical oscillators are emerging ultrasensitive sensors with tremendous potential in both applied and fundamental physics. Levitated ferromagnets, with internal spin noises rapidly averaged, promise ultrahigh magnetic sensitivity. Here, we demonstrate a milligram-scale diamagnetically levitated ferromagnet system operating at room temperature. Through optimized geometry and multi-channel dissipation control, we achieve a magnetic sensitivity of 23~fT$/\sqrt{\text{Hz}}$ at frequency of 100-Hz level. We anticipate that a ferromagnetic magnetometer with subfemtotesla sensitivity is within reach, after modest technical improvements. This platform establishes a high-performance magnetometer for biomagnetic field detection and beyond-standard-model force searches.
Backdoor attacks in multimodal contrastive learning (MCL) have garnered growing attention in recent years, as many downstream tasks critically depend on pre-trained MCL models. Existing detection-based defenses predominantly rely on the CLIPScore metric, under the assumption that poisoned pairs exhibit lower semantic similarity between the image and the caption. However, we identify two critical flaws remaining in existing methods: (1) the substantial overlap between CLIPScore distributions of benign and poisoned pairs undermines the reliability of this metric, and (2) fixed-threshold detection cannot provide statistical guarantees for ambiguous samples within overlapping regions. To overcome these limitations, we propose integrating conformal prediction (CP), a statistical framework that quantifies uncertainty through nonconformity scores (NCSs), to establish provable confidence bounds for detecting poisoned image-caption pairs. Building on CP, we introduce CASCADE, a novel two-stage Coarse-to-Fine Conformal Backdoor Detection framework. The coarse-grained stage uses cross-modality consistency to identify high-confidence benign and poisoned pairs. In the fine-grained stage, a reference set is constructed from high-confidence poisoned pairs, and instance-level NCSs based on text-space similarity are computed for each sample in the unidentified subset. These NCSs measure conformity to the poisoning distribution and enable precise identification of latent poisoned pairs within the unidentified subset. Extensive experiments on the large-scale CC3M dataset demonstrate that CASCADE achieves an average FPR of 5.79% at 100% TPR and an average AUROC of 0.9867 across diverse attacks, while remaining effective against adaptive attacks.
Large language models have made it possible to generate executable computer-aided design (CAD) programs from natural-language descriptions or images. However, existing methods represent modeling processes as backend-specific sequential scripts with implicit dependencies or as static geometry, making it difficult to simultaneously preserve construction history, stable topological references, and feature-level editability across different CAD systems. We present CADIR, an agent-friendly executable intermediate representation for CAD generation and cross-backend editing. Built on the OCCT geometry kernel via OCP, CADIR provides explicit, compositional modeling operations and fine-grained execution diagnostics. During program execution, CADIR records modeling operations, parameter dependencies, constraints, and topology selections in a construction graph. To enable reliable cross-backend reconstruction, we introduce Geometric Signature Matching, which identifies corresponding edges and faces despite parameter changes and backend differences, allowing adapters to reconstruct native editable feature histories in FreeCAD, SolidWorks, and Fusion 360. Building on this representation, we further propose a construction-graph retrieval method for text and image queries that supports both full-graph and subgraph retrieval, enabling agents to leverage complete models and modeling substructures. Extensive experiments demonstrate that CADIR achieves higher geometric fidelity and execution reliability than existing CAD representations, that construction-graph retrieval further improves model generation quality, and that cross-backend editing enables reliable model reconstruction and post-reconstruction editing across multiple CAD environments.
X-ray absorption spectroscopy (XAS) is central to understanding the local electronic and atomic structure of materials, yet most published spectra remain inaccessible to data-driven analysis because they are embedded in figures and described through fragmented textual context in the literature. Here, we use multimodal (image and text) literature mining to transform this dispersed knowledge into an AI-ready experimental data resource. We developed a scalable spectroscopy data digitization pipeline that identifies XAS figures in full-text articles, digitizes spectral curves, and links each spectrum to accompanying metadata on the measured edge and material. Applying this pipeline to the battery literature produced an open dataset of 13,740 XAS spectra, spanning 66 absorbing elements and diverse battery chemistries, with expert validation confirming accurate extraction of spectral and metadata information. By converting literature-embedded spectra into structured numerical data, this dataset provides a foundation for large-scale XAS analysis, cross-laboratory comparison, high-throughput characterization, and autonomous discovery of advanced materials.
In this paper, we establish a moderate deviation bound for Gaussian maxima and the variance asymptotics of the Sherrington-Kirkpatrick free energy at criticality based on Gaussian convexity. First, let $(X_1,\ldots,X_N)$ be centered Gaussian vector with $\operatorname{Var}(X_i)\leq 1$. Suppose that, for fixed $\alpha\in(0,\sqrt 2)$ and $\kappa>0$, $\mathbb{E}\max_iX_i\geq\alpha\sqrt{\log N}$ and $\mathbb{E}\max_iX_i+\kappa\sqrt{\log N}\leq\sqrt{2\log N}$. We prove that $$ \mathbb{P}\left(\max_iX_i\geq \mathbb{E}\max_iX_i+\kappa\sqrt{\log N}\right) \leq N^{-\kappa^2/(2-\alpha^2)+o(1)}. $$ This answers a question of Ding, Eldan and Zhai. The exponent is sharp, as witnessed by an equicorrelated Gaussian field. Second, for the Sherrington--Kirkpatrick model at the critical inverse temperature $\beta_c=1/\sqrt2$, we prove $$ \operatorname{Var}\bigl(F_N(\beta_c)\bigr)=\frac16\log N+O(1). $$ Our argument provides the variance asymptotics at the critical temperature from an entropy perspective, via a route distinct from that of Du and Huang. For the upper bound, we express the variance as an entropy under exponential tilting and identify this entropy with the Kullback--Leibler divergence of a Gaussian synchronization model. Its derivative is then bounded using the I-MMSE formula, information percolation, and estimates for the susceptibility of the critical Erd\H{o}s--R'enyi random graph. For the lower bound, we combine Gaussian convexity applied at the replica parameter with an estimate for inverse moments on the sphere and an identity relating GOE eigenvalue densities in consecutive dimensions.
We give a Graham-positive expansion for the product of two double Schubert polynomials indexed by two inverse Grassmannian permutations. Surprisingly, the nonzero structure constants are double Schubert polynomials in two disjoint sets of equivariant variables. We also give a positive expansion for the product of two single Schubert polynomials indexed by a $321$-avoiding permutation (e.g., a Grassmannian permutation) and an inverse Grassmannian permutation. Unexpectedly, the nonzero structure constants are Edelman--Greene coefficients.
We study a version of de Sitter static patch holography in which the Euclidean gravitational path integral, with an observer worldline included, is conjectured to compute a trace. Motivated by recent evidence for this conjecture from the sphere path integral, we test it further by inserting operators along the observer worldline and computing two-point functions and out-of-time-ordered four-point correlators (OTOCs). We extend an earlier OTOC calculation by Kolchmeyer and Liu using the shockwave formalism, incorporating both observer recoil and gravitational backreaction. We find that the OTOC conflicts with two basic properties of a trace in a Hilbert space: cyclicity and positivity. The signaling feature of the shockwave geometry gives rise to two distinct resummations of the perturbative eikonal expansion, which are interchanged by cyclicity. Positivity is violated by the fact that the leading perturbative contribution causes the (regularized) OTOC to increase.
Audio-language models compress a speech encoder's output through a Querying Transformer (Q-Former) connector before feeding it to a large language model. We identify two failures in this compression. The connector's output vectors collapse to a single direction, and different speakers produce nearly indistinguishable outputs, with paralinguistic cues such as speaker identity, gender, and prosody lost along the way. Our method, ORCA, reverses this collapse by splitting the queries into groups whose outputs are constrained to point in different directions. On SAKURA multi-hop reasoning, ORCA gains 26.4 points over an identically trained 4B baseline, reaching 75.2% (vs. 49.0% for the 8B Audio Flamingo-3). At the connector level, the same change cuts query redundancy by 12x and raises cross-speaker variance by 75x.
End-to-end ASR models transcribe in a single pass, leaving no room for the decoder to revisit hard inputs. We propose LatentASR, a parameter-efficient method that adds continuous latent test-time scaling to a frozen ASR backbone. Two small trainable modules drive it: a Latent Adapter that iteratively refines a few latent prefix positions through bounded, stabilized updates, and a Value Head that predicts whether extra computation will help and halts the loop early. The Qwen3-ASR-0.6B backbone stays fully frozen, and we train only ~4M extra parameters. We activate this loop with a deliberately small, diverse 500-utterance training set. Under this minimal-data regime, standard adaptation methods all regress: full fine-tuning, LoRA, and prompt tuning each increase WER. LatentASR is the only tested method that reduces WER on both clean benchmarks (FLEURS -2.54% and VoxPopuli -0.47% relative). The reductions are concentrated on intrinsically hard inputs. On accented and code-switched speech (ASCEND), LatentASR achieves a 16.0% relative CER reduction. Across 30 FLEURS languages (23,049 utterances), the multilingual WER decreases uniformly across resource tiers, confirming that the adapter generalizes without overfitting. Dynamic halting preserves most of the clean-set reduction at a fraction of the compute, skipping roughly half of all utterances at the entry gate. Our results show that a small, carefully chosen activation set can switch on test-time scaling inside a frozen ASR model without corrupting the model itself, converting fixed per-utterance compute into input-dependent compute where it is most needed.
Technical lectures mix Mandarin speech with English technical terms. These terms carry the core meaning of the lecture, yet they occupy few characters. Character error rate (CER) therefore hides their recognition failures. We study whether lecture context helps recognize these terms. We build a term-rich Mandarin AI/ML lecture benchmark, and we define term-centric metrics that measure technical-term recognition directly. We then propose a two-pass, reference-free decoding method. The first pass runs segment-only ASR. We extract the most frequent technical terms from the first-pass hypotheses, and we prompt the recognizer with this self-built glossary in the second pass. Across five ASR backbones, the first-pass glossary raises term recall for every model and holds or lowers CER on all five. On Breeze-ASR-25 it lifts term recall from 52.50% to 60.13% while lowering CER, and a hybrid that adds a small external term list reaches 62.05% recall and 82.73% term precision. Lecture context, recovered from the model's own output, is a practical signal for technical-term recognition. Term-centric evaluation exposes errors that CER misses.
We study the size of the largest component in Pittel's preferential attachment process without vertex growth. Starting from the empty graph on a fixed vertex set $[n]$, edges are added one by one with probabilities proportional to $(d_u+\alpha)(d_v+\alpha)$, where $d_u$ and $d_v$ are the current degrees of $u$ and $v$, and $\alpha>0$. Let $L_1$ denote the size of the largest component, and set $m_c:=\frac{\alpha n}{2(\alpha+1)}.$ We prove that if $m=m_c(1-\varepsilon), \varepsilon=\varepsilon(n)\to0, \varepsilon^3 n\to\infty,$ then \[ L_1=(1+o_p(1))\frac{2(\alpha+2)}{\alpha+1}\varepsilon^{-2}\log(\varepsilon^3 n) \] for every fixed $\alpha>0$. Moreover, the same asymptotic holds whenever $\alpha=\alpha(n)\to a\in(0,\infty]$. In particular, the constant $2(\alpha+2)/(\alpha+1)$ converges to the Erd\H{o}s--R\'enyi value $2$ as $\alpha\to\infty$. If $m=\left\lfloor \frac n2(1-\varepsilon)\right\rfloor$ and $\alpha\varepsilon\to\infty$, then \[ L_1=(2+o_p(1))\varepsilon^{-2}\log(\varepsilon^3 n). \] The subcritical asymptotics for \(L_1\) resolve the problem left open by Janson and Warnke. The upper bound argument relies on the fact that, after conditioning on the degree sequence, the graph can be treated through the corresponding configuration model, the lower bound follows from tree component asymptotics and a second moment argument.
In this paper, we improve the arbitrary Banach space \(n \log n\) bound of Ivanisvili--Volberg \cite{IvanisviliVolberg2022} for the second order projection bound to the order \(\sqrt{n}\) bound. Moreover, we study the lower Riesz estimate with the pointwise square gradient, and prove a fixed chaos characterization: on every fixed homogeneous Walsh chaos $H_k$, the dimension free estimate \[ \|\Delta^{1/2}f\|_{L^p(\Omega_n;X)} \lesssim_{p,k,X} \||\nabla f|_X\|_{L^p(\Omega_n)} \] holds for all $n$ if and only if $X$ has Rademacher type $2$. We also consider an exact tail space norm of the analytic paraproduct $T_\varphi g(z)=\int_0^z g(\zeta)\varphi'(\zeta)\,d\zeta$ on Banach valued \(H^\infty\) spaces. A matching lower bound of Volberg \cite{Volberg2024} \[ \|T_\varphi:H_d^\infty(\mathbb D;Y)\to H^\infty(\mathbb D;Y)\| \asymp_{\alpha,\varphi} d^{-\alpha} \] under a nondegenerate boundary singularity assumption is established.
Recommendation algorithm iteration is moving from an artisanal, engineer-bound process toward an industrialized research loop, but this transition remains blocked by a structural execution bottleneck: the idea-to-launch cycle still depends on human engineers to generate hypotheses, modify production code, launch A/B experiments, and attribute online results. Innovation therefore scales linearly with headcount rather than compounding with evidence, compute, and accumulated experimental knowledge. We present AgentX, a production-deployed multi-agent system that fundamentally restructures this production function. AgentX operates as a self-evolving development engine: it autonomously generates, implements, evaluates, and learns from recommendation experiments at a scale and pace that no manual workflow can sustain. The system orchestrates four tightly coupled stages in a closed loop. A Brainstorm Agent synthesizes evidence from historical experiments, system architecture, data analysis, and external research into ranked, executable proposals. A Developing Agent translates each proposal into production-ready code through repository-grounded generation and multi-dimensional reliability verification. An Evaluation Agent conducts safe online rollout with guardrail-vetoed A/B judgment, converting both successes and failures into structured knowledge assets. A Harness Evolution layer (SGPO) then distills execution trajectories into semantic-gradient updates that continuously sharpen the agents themselves -- making the system not merely automated, but self-improving.