Searcharxiv⌕ Search

arXiv subjects

Zhi Yang

Publications and source records attributed to Zhi Yang.

At least 19 recordsLinked to original sources
arXiv2026

Hierarchical Fast-Slow ReAct Agent for Zero-Shot Object-Goal Navigation

Zero-shot object-goal navigation (ZSON) requires a robot to find a named object category in a building it has never entered. The prevailing approach scores frontiers with a vision-language value map: every decision is another argmax over the map as it currently stands, and the evidence behind that score is discarded the moment it is taken. Systems that place a large vision-language model inside the perception-action loop typically query it on a fixed schedule from the current view alone; a room the robot walked through minutes earlier is never reconsidered, and a failed call has no defined fallback. We turn what the robot has already seen into the object of deliberation. Our hierarchical fast-slow agent leaves the value-map controller running at every step and writes a coordinate-anchored memory as it moves: a semantic grid of room types and confirmed object instances, together with a bounded store of pose-tagged keyframes. A VLM screens each candidate detection before it is written. A deliberative layer reads this memory in a bounded reason-retrieve-act loop. It wakes on structural events the reactive layer computes, reasons first over text, and recalls a first-person view only for candidates that text alone cannot separate. Per-invocation and per-run caps bound its calls, a call-free first tier resolves the most frequent stall, and any failure returns control to the reactive controller. Our system reaches 68.75% SR on HM3D v1 val and 47.29% on MP3D val, the highest success rate among the zero-shot methods compared here. Choosing among far frontiers by argmax instead of deliberating costs 3.40 SR points in a paired comparison over all 2000 HM3D episodes (95% CI [1.70, 5.05]); deliberating over every frontier does not recover them.

cs.RO↗
arXiv2026

TileSight: A First-Principles Tile-Centric Analytical GPU Performance Model from Cores to Clusters

Recent GPU programming frameworks such as Triton, TileLang, and CUDA Tile adopt tiles as first-class primitives, making tile-centric programming the prevailing approach for high-performance GPU kernels. Performance-analysis tooling has not followed: programmers still rely on coarse roofline bounds, opaque ML predictors, or post-hoc profilers to understand kernel execution. This gap is acute for modern AI workloads, where kernel fusion and distributed inference depend on tensor cores, CUDA cores, cache hierarchies, memory pipelines, and inter-GPU networks. We present TileSight, a tile-centric performance-modeling tool that elevates the tile from a programming primitive to an analysis primitive. Within a GPU core, TileSight models compute-memory pipeline overlap; across cores, it models the cache hierarchy; across GPUs, it models inter-node communication. All layers share the tile abstraction: the intra-tile layer expresses work as a resource vector spanning network, memory, and compute pipelines; the inter-tile layer schedules dependent and ordered actions to expose legal overlap and infers multi-level cache hit rates from tile reuse distance; and the cross-device layer maps remote tensor accesses to placements and routes them through an alpha-beta stage cost. On A100, H200, B200, and B6000, TileSight predicts single-GPU kernel latency with 12.35% pooled mean absolute percentage error (MAPE), outperforming state-of-the-art baselines and transferring better across architectures. Its L2 cache-hit-rate predictions are within roughly one percentage point of measurements on every GPU. At up to 32 GPUs, TileSight achieves 16.18% weighted MAPE (wMAPE) on fused distributed kernels and 13.52% wMAPE on end-to-end vLLM serving. In optimization, TileSight selects tile configurations competitive with strong vendor and expert baselines. TileSight will be open-sourced upon publication.

cs.DC↗
arXiv2026

PRISMR: Overcoming Parse Collapse in Multimodal Listwise Ranking via Parameterized Representation Internalization

Generative listwise ranking with Large Multimodal Models (LMMs) aims to capture global list context in a single forward pass, but its effectiveness degrades in long-context multimodal scenarios. We identify a recurring failure mode, parse collapse, where the autoregressive decoder produces fluent yet incomplete rankings by silently omitting candidates and terminating early. This failure stems from limited context utilization rather than simple formatting mistakes, making prompt engineering and constrained decoding insufficient. We propose PRISMR (Parameterized Representation Internalization for Semantic Multimodal Ranking), a framework that replaces transient in-context list processing with parametric structural conditioning. PRISMR uses a lightweight hypernetwork to encode multimodal candidates in parallel and generate item-specific LoRA weights, which are synthesized into an instance-specific adapter for a LMM. This paradigm enables more robust internalization of list structure while preserving the base model. We further introduce a large-scale multimodal review-ranking benchmark for evaluation. Experiments demonstrate that PRISMR substantially reduces parse collapse, improves listwise ranking performance, and transfers effectively across domains and instruction-tuned backbones.

cs.AI↗
arXiv2026

Systematic Study of Coupled-Channel Dynamics in Doubly Heavy Hadronic Molecules

Heavy Quark Spin Symmetry (HQSS) is widely use to predict heavy molecules by extending the effective interactions fitted from low-lying states to heavier sectors. In this work, we systematically investigate the reliability of this approach for higher double heavy tetraquarks by comparing a single-channel effective interaction (Scheme I) with an explicit coupled-channel dynamics framework (Scheme II). The interactions are obtained within one-boson-exchange potential model and fixed by fitting the $T_{cc}^+$ lineshape. Utilizing the complex scaling method and $T$-matrix pole analysis, we extract the possible poles in the $S$-wave $D^{(*)}D^{(*)}$, $\bar{B}^{(*)}\bar{B}^{(*)}$ and $D^{(*)}\bar{B}^{(*)}$ systems with $J^{P}=1^+$. We find that both schemes provide consistent descriptions of the lowest-lying state. This confirms isoscalar-dominated $T_{cc}$ as a predominant $DD^*$ molecule (binding energy $\sim$ 381 keV), and predicts an isoscalar deeply bound $T_{bb}$ state ($40-60$ MeV) and an isovector $T^\prime_{bb}$ resonance in the bottom sector, together with a virtual $T_{bc}$ state. In contrast, significant differences emerge for higher-lying states. The inclusion of explicit coupled-channel dynamics modifies the effective interaction and reshapes the pole structure. The states predicted as bound or resonant in the single-channel framework can be shifted far from the physical region or disappear. These results indicate that while single-channel descriptions are adequate for near-threshold states, an explicit treatment of coupled-channel dynamics is required for reliable predictions of excited doubly heavy tetraquarks.

hep-ph↗
arXiv2026

SkillGenBench: Benchmarking Skill Generation Pipelines for LLM Agents

As LLM agents are increasingly built around reusable skills, a central challenge is no longer only whether agents can use provided skills, but whether they can generate correct, reusable, and executable skills from repositories and documents. Existing benchmarks primarily evaluate the efficacy of given skills or the ability of agents to solve downstream tasks from raw context, but they do not isolate skill generation itself as the object of study. We introduce SkillGenBench, a benchmark for evaluating skill generation pipelines under a unified and controlled protocol. In SkillGenBench, a generator receives raw corpora and produces standardized skill artifacts, which are then executed under fixed harnesses and assessed with unified evaluation procedures. The benchmark covers two generation regimes: task-conditioned generation, where a task-specific skill is synthesized after the task is revealed, and task-agnostic generation, where a reusable skill library must be distilled before downstream tasks are known. It also spans two complementary procedural sources: repository-grounded instances, where procedures are distributed across code, configuration, and scripts, and document-grounded instances, where procedures and constraints must be distilled from long-form text. We provide standardized task specifications, pinned environments, and evaluation protocols centered on deterministic execution-based checks, supplemented by auxiliary signals for diagnosis. Experiments across a range of skill-generation methods and backbones show substantial performance variation, highlight the difficulty of reusable skill distillation, and reveal distinct failure modes in skill generation from software repositories versus long-form documents. SkillGenBench establishes a reproducible testbed for studying skill generation as an independent research problem in agent systems.

cs.AI↗
arXiv2026

DynaTrain: Fast Online Parallelism Switching for Elastic LLM Training

Modern large language model (LLM) training is inherently dynamic: resource fluctuations, RLHF phase shifts, and cluster elasticity continually reshape the optimal parallelism layout, posing a significant challenge to existing training frameworks built around a static execution model. We present DynaTrain, a distributed training system for sub-second, online reconfiguration across arbitrary multi-dimensional parallelism. At its core, we propose a Virtual Parameter Space (VPS) abstraction that unifies all distributed training states under one logical coordinate space, turning any parallelism configuration into a deterministic mapping and collapsing complex transition into manageable geometric intersections. On top of VPS, a state routing-and-transition layer executes rank-local transfers under a memory-aware, deadlock-free schedule, and an Elastic Device Manager overlaps new-world construction with ongoing training to mask topology-change cost. On dense and MoE models up to 235B parameters, DynaTrain reconfigures a 70B dense model in under 2s and a 235B MoE model in 4.36s, outperforming state-of-the-art checkpoint-based and elastic systems by up to three orders of magnitude while preserving correctness.

cs.LG↗
arXiv2026

EEXICC: An event generator for doubly heavy baryon production at $e^+e^-$ colliders

We present EEXICC, a Monte Carlo event generator designed to simulate the production of doubly heavy baryons ($\Xi_{cc}$, $\Xi_{bc}$, and $\Xi_{bb}$) via $e^+e^-$ annihilation. Based on nonrelativistic QCD effective theory, the generator calculates the process $e^{+}+e^{-}\rightarrow \Xi_{QQ'}+\bar{Q}'+\bar{Q}$ using an improved trace technique at the amplitude level, which greatly improves numerical efficiency compared with traditional squared-amplitude methods. EEXICC is developed in Fortran with a modular structure and is fully compatible with the PYTHIA framework, enabling convenient integration into complete event simulation workflows. The program supports both weighted and unweighted event generation, and its numerical reliability has been verified against existing theoretical results. EEXICC provides a flexible and robust tool for studying the properties of doubly heavy baryons at future high-luminosity and high-energy $e^+e^-$ colliders such as the CEPC and FCC-ee.

hep-ph↗
arXiv2026

A Full-Stack Performance Evaluation Infrastructure for 3D-DRAM-based LLM Accelerators

Large language models (LLMs) exhibit memory-intensive behavior during decoding, making it a key bottleneck in LLM inference. To accelerate decoding execution, hybrid-bonding-based 3D-DRAM has been adopted in LLM accelerators. While this emerging technology provides strong performance gains over existing hardware, current 3D-DRAM accelerators (3D-Accelerators) rely on closed-source evaluation tools, limiting access to publicly available performance analysis methods. Moreover, existing designs are highly customized for specific scenarios, lacking a general and reusable full-stack modeling for 3D-Accelerators across diverse usecases. To bridge this fundamental gap, we present ATLAS, the first silicon-proven Architectural Three-dimesional-DRAM-based LLM Accelerator Simulation framework. Built on commercially deployed multi-layer 3D-DRAM technology, ATLAS introduces unified abstractions for both 3D-Accelerator system architecture and programming primitives to support arbitrary LLM inference scenarios. Validation against real silicon shows that ATLAS achieves $\le$8.57% simulation error and 97.26-99.96\% correlation with measured performance. Through design space exploration with ATLAS, we demonstrate its ability to guide architecture design and distill key takeaways for both 3D-DRAM memory system and 3D-Accelerator microarchitecture across scenarios. ATLAS will be open-sourced upon publication, enabling further research on 3D-Accelerators.

cs.AR↗
arXiv2026

Atomic Trajectory Modeling with State Space Models for Biomolecular Dynamics

Understanding the dynamic behavior of biomolecules is fundamental to elucidating biological function and facilitating drug discovery. While Molecular Dynamics (MD) simulations provide a rigorous physical basis for studying these dynamics, they remain computationally expensive for long timescales. Conversely, recent deep generative models accelerate conformation generation but are typically either failing to model temporal relationship or built only for monomeric proteins. To bridge this gap, we introduce ATMOS, a novel generative framework based on State Space Models (SSM) designed to generate atom-level MD trajectories for biomolecular systems. ATMOS integrates a Pairformer-based state transition mechanism to capture long-range temporal dependencies, with a diffusion-based module to decode trajectory frames in an autoregressive manner. ATMOS is trained across crystal structures from PDB and conformation trajectory from large-scale MD simulation datasets including mdCATH and MISATO. We demonstrate that ATMOS achieves state-of-the-art performance in generating conformation trajectories for both protein monomers and complex protein-ligand systems. By enabling efficient inference of atomic trajectory of motions, this work establishes a promising foundation for modeling biomolecular dynamics.

q-bio.BM↗
arXiv2026

General Hamiltonian Approach to the $\mathbf{N}$-Body Finite-Volume Formalism: Extracting the $\mathbf{\omega}$ Resonance Parameters from Lattice QCD

We present a nonperturbative Hamiltonian framework (NPHF) to address the general $N$-body problem. This framework rigorously connects finite-volume spectra from lattice QCD to scattering observables from experiment. To demonstrate its applicability, we extract the resonance parameters of the $\omega$ meson by simultaneously analyzing the isoscalar $3\pi$ and isovector $2\pi$ systems. The Hamiltonian unifies single-particle $\omega$, two-particle $\rho\pi$, and three-particle $\pi\pi\pi$ dynamics within a single unitary formalism. Using leading lattice QCD spectra from the Chinese Lattice QCD Collaboration at $m_\pi$ = 208 and 305 MeV, we perform a fit in the isovector and isoscalar channels, accurately describe the lattice spectra and obtain robust determinations of the $\rho$ and $\omega$ pole positions. This work establishes a foundational approach for extracting resonance dynamics from finite-volume spectra. Given the ubiquity of three-body dynamics in exotic hadrons, halo nuclei, and neutron star matter, this general formalism holds broad relevance across particle, nuclear, and astrophysical physics.

hep-lat↗
arXiv2026

Property-driven Protein Inverse Folding With Multi-Objective Preference Alignment

Protein sequence design must balance designability, defined as the ability to recover a target backbone, with multiple, often competing, developability properties such as solubility, thermostability, and expression. Existing approaches address these properties through post hoc mutation, inference-time biasing, or retraining on property-specific subsets, yet they are target dependent and demand substantial domain expertise or careful hyperparameter tuning. In this paper, we introduce ProtAlign, a multi-objective preference alignment framework that fine-tunes pretrained inverse folding models to satisfy diverse developability objectives while preserving structural fidelity. ProtAlign employs a semi-online Direct Preference Optimization strategy with a flexible preference margin to mitigate conflicts among competing objectives and constructs preference pairs using in silico property predictors. Applied to the widely used ProteinMPNN backbone, the resulting model MoMPNN enhances developability without compromising designability across tasks including sequence design for CATH 4.3 crystal structures, de novo generated backbones, and real-world binder design scenarios, making it an appealing framework for practical protein sequence design.

cs.LG↗
arXiv2026

Autonomous Algorithm Discovery for Ptychography via Evolutionary LLM Reasoning

Ptychography is a computational imaging technique widely used for high-resolution materials characterization, but high-quality reconstructions often require the use of regularization functions that largely remain manually designed. We introduce Ptychi-Evolve, an autonomous framework that uses large language models (LLMs) to discover and evolve novel regularization algorithms. The framework combines LLM-driven code generation with evolutionary mechanisms, including semantically-guided crossover and mutation. Experiments on three challenging datasets (X-ray integrated circuits, low-dose electron microscopy of apoferritin, and multislice imaging with crosstalk artifacts) demonstrate that discovered regularizers outperform conventional reconstructions, achieving up to +0.26 SSIM and +8.3~dB PSNR improvements. Besides, Ptychi-Evolve records algorithm lineage and evolution metadata, enabling interpretable and reproducible analysis of discovered regularizers.

cs.CE↗
arXiv2026

Chain of Mindset: Reasoning with Adaptive Cognitive Modes

Human problem-solving is never the repetition of a single mindset, by which we mean a distinct mode of cognitive processing. When tackling a specific task, we do not rely on a single mindset; instead, we integrate multiple mindsets within the single solution process. However, existing LLM reasoning methods fall into a common trap: they apply the same fixed mindset across all steps, overlooking that different stages of solving the same problem require fundamentally different mindsets. This single-minded assumption prevents models from reaching the next level of intelligence. To address this limitation, we propose Chain of Mindset (CoM), a training-free agentic framework that enables step-level adaptive mindset orchestration. CoM decomposes reasoning into four functionally heterogeneous mindsets: Spatial, Convergent, Divergent, and Algorithmic. A Meta-Agent dynamically selects the optimal mindset based on the evolving reasoning state, while a bidirectional Context Gate filters cross-module information flow to maintain effectiveness and efficiency. Experiments across six challenging benchmarks spanning mathematics, code generation, scientific QA, and spatial reasoning demonstrate that CoM achieves state-of-the-art performance, outperforming the strongest baseline by 4.96\% and 4.72\% in overall accuracy on Qwen3-VL-32B-Instruct and Gemini-2.0-Flash, while balancing reasoning efficiency. Our code is publicly available at \href{https://github.com/QuantaAlpha/chain-of-mindset}{https://github.com/QuantaAlpha/chain-of-mindset}.

cs.AI↗
arXiv2026

QuantaAlpha: An Evolutionary Framework for LLM-Driven Alpha Mining

Financial markets are noisy and non-stationary, making alpha mining highly sensitive to backtest noise and regime shifts. While recent agentic frameworks improve automation, they often lack controllable multi-round search and reliable reuse of validated experience. To address these challenges, we propose QuantaAlpha, an evolutionary alpha mining framework that treats each end-to-end mining run as a trajectory and improves factors via trajectory-level mutation and crossover. QuantaAlpha localizes suboptimal steps for targeted revision and recombines complementary high-reward segments to reuse effective patterns, enabling structured exploration and refinement across iterations. During factor generation, it enforces semantic consistency across hypothesis, factor expression, and executable code, and constrains the complexity and redundancy of the generated factor to mitigate crowding. Extensive experiments on CSI 300 show consistent gains over strong baselines and prior agentic systems. Using GPT-5.2, QuantaAlpha achieves an IC of 0.0472 with ARR of 4.68% and MDD of 11.8%. Moreover, factors mined on CSI 300 transfer effectively to CSI 500 and the S&P 500, delivering about 40.28% and 19.1% cumulative excess return over four years, respectively, which indicates strong robustness under market distribution shifts.

q-fin.ST↗
arXiv2026

Spider-Sense: Intrinsic Risk Sensing for Efficient Agent Defense with Hierarchical Adaptive Screening

As large language models (LLMs) evolve into autonomous agents, their real-world applicability has expanded significantly, accompanied by new security challenges. Most existing agent defense mechanisms adopt a mandatory checking paradigm, in which security validation is forcibly triggered at predefined stages of the agent lifecycle. In this work, we argue that effective agent security should be intrinsic and selective rather than architecturally decoupled and mandatory. We propose Spider-Sense framework, an event-driven defense framework based on Intrinsic Risk Sensing (IRS), which allows agents to maintain latent vigilance and trigger defenses only upon risk perception. Once triggered, the Spider-Sense invokes a hierarchical defence mechanism that trades off efficiency and precision: it resolves known patterns via lightweight similarity matching while escalating ambiguous cases to deep internal reasoning, thereby eliminating reliance on external models. To facilitate rigorous evaluation, we introduce S$^2$Bench, a lifecycle-aware benchmark featuring realistic tool execution and multi-stage attacks. Extensive experiments demonstrate that Spider-Sense achieves competitive or superior defense performance, attaining the lowest Attack Success Rate (ASR) and False Positive Rate (FPR), with only a marginal latency overhead of 8.3\%.

cs.CR↗
arXiv2026

BinaryDemoire: Moir\'e-Aware Binarization for Image Demoir\'eing

Image demoir\'eing aims to remove structured moir\'e artifacts in recaptured imagery, where degradations are highly frequency-dependent and vary across scales and directions. While recent deep networks achieve high-quality restoration, their full-precision designs remain costly for deployment. Binarization offers an extreme compression regime by quantizing both activations and weights to 1-bit. Yet, it has been rarely studied for demoir\'eing and performs poorly when naively applied. In this work, we propose BinaryDemoire, a binarized demoir\'eing framework that explicitly accommodates the frequency structure of moir\'e degradations. First, we introduce a moir\'e-aware binary gate (MABG) that extracts lightweight frequency descriptors together with activation statistics. It predicts channel-wise gating coefficients to condition the aggregation of binary convolution responses. Second, we design a shuffle-grouped residual adapter (SGRA) that performs structured sparse shortcut alignment. It further integrates interleaved mixing to promote information exchange across different channel partitions. Extensive experiments on four benchmarks demonstrate that the proposed BinaryDemoire surpasses current binarization methods. Code: https://github.com/zhengchen1999/BinaryDemoire.

cs.CV↗
arXiv2026

EvoFSM: Controllable Self-Evolution for Deep Research with Finite State Machines

While LLM-based agents have shown promise for deep research, most existing approaches rely on fixed workflows that struggle to adapt to real-world, open-ended queries. Recent work therefore explores self-evolution by allowing agents to rewrite their own code or prompts to improve problem-solving ability, but unconstrained optimization often triggers instability, hallucinations, and instruction drift. We propose EvoFSM, a structured self-evolving framework that achieves both adaptability and control by evolving an explicit Finite State Machine (FSM) instead of relying on free-form rewriting. EvoFSM decouples the optimization space into macroscopic Flow (state-transition logic) and microscopic Skill (state-specific behaviors), enabling targeted improvements under clear behavioral boundaries. Guided by a critic mechanism, EvoFSM refines the FSM through a small set of constrained operations, and further incorporates a self-evolving memory that distills successful trajectories as reusable priors and failure patterns as constraints for future queries. Extensive evaluations on five multi-hop QA benchmarks demonstrate the effectiveness of EvoFSM. In particular, EvoFSM reaches 58.0% accuracy on the DeepSearch benchmark. Additional results on interactive decision-making tasks further validate its generalization.

cs.AI↗
arXiv2026

RLPO: Residual Listwise Preference Optimization for Long-Context Review Ranking

Review ranking is pivotal in e-commerce for prioritizing diagnostic and authentic feedback from the deluge of user-generated content. While large language models have improved semantic assessment, existing ranking paradigms face a persistent trade-off in long-context settings. Pointwise scoring is efficient but often fails to account for list-level interactions, leading to miscalibrated top-$k$ rankings. Listwise approaches can leverage global context, yet they are computationally expensive and become unstable as candidate lists grow. To address this, we propose Residual Listwise Preference Optimization (RLPO), which formulates ranking as listwise representation-level residual correction over a strong pointwise LLM scorer. RLPO first produces calibrated pointwise scores and item representations, then applies a lightweight encoder over the representations to predict listwise score residuals, avoiding full token-level listwise processing. We also introduce a large-scale benchmark for long-context review ranking with human verification. Experiments show RLPO improves NDCG@k over strong pointwise and listwise baselines and remains robust as list length increases.

cs.IR↗