Search arXiv⌕ Search

arXiv subjects

Zhichao Chen

Publications and source records attributed to Zhichao Chen.

At least 19 recordsLinked to original sources
arXiv2026

Drafting Agents, Not Just Tokens: Speculative Recursive Self-Improvement for Agentic Systems

Improving an agent through successive revisions requires balancing exploration and evaluation under a limited budget. A narrow search can miss useful sequences of edits, while fully evaluating every intermediate version often exhausts the budget before those sequences yield a verified improvement. Inspired by the draft--verify principle of speculative decoding, we propose Speculative Recursive Self-Improvement (SpecRSI), a two-level framework coupling inner-loop agent search with outer-loop improver learning. SpecRSI drafts successors before their predecessors are fully evaluated, while adopting only fully evaluated, admissible versions. In the inner loop, Verification-Aware Tree Search (VATS) allocates the budget between further edits and outstanding checks using the predicted verified return of a continuation. In the outer loop, Group Verification Optimization (GVO) updates the improver from completed searches while keeping the real evaluator fixed. We evaluate SpecRSI on scientific optimization (including active queue management (AQM) for network infrastructure), coding, and general agent-system benchmarks. Under matched evaluation-call budgets, SpecRSI improves scientific terminal gain over Janus by 60.0%--96.3% and area under the budget curve by 32.1%--91.6%, while achieving higher means across all 50 scientific backbone--task pairs and outperforming MetaRSI-v1 by 1.30--2.70 percentage points (pp) in system score across twenty backbones. On instance-level comparisons, it further improves PyBaMM terminal gain by 142.6% and SWE-Pro success by 6.0 pp at 7.0% and 9.2% lower token cost, respectively. These results demonstrate the efficacy of the proposed approach.

cs.LG↗
arXiv2026

Sign-coherence and tropical sign pattern for rank $3$ real cluster-cyclic exchange matrices

The sign-coherence about $c$-vectors was conjectured by Fomin-Zelevinsky and proved by Gross-Hacking-Keel-Kontsevich for integer skew-symmetrizable case. We prove this conjecture associated with $c$-vectors for rank 3 real cluster-cyclic skew-symmetrizable case. Simultaneously, we establish their self-contained recursion and monotonicity. Then, these $c$-vectors are proved to be roots of certain quadratic equations. Based on these results, we prove that the corresponding exchange graphs of $C$-pattern and $G$-pattern are $3$-regular trees. We also study the structure of tropical signs and equip the dihedral group $\mathrm{D}_6$ with a cluster realization via certain mutations.

math.CO↗
arXiv2026

Log-concavity and unimodality of cluster monomials in fan coordinates of type $A_n$

We prove that every cluster monomial of type $A_n$ is log-concave and unimodal with respect to its Laurent expansion in a fan initial cluster. We derive a coefficient formula for laminar interval products arising from compatible diagonals in the polygon model. In even rank, we reduce log-concavity in fan coordinates to log-concavity along integer lines. In odd rank, we use convolution to handle the coefficient sums arising from the fan substitution. This gives partial affirmative answers to the log-concavity conjecture of Chen-Huang-Sun and the unimodality conjecture of Chen. Our result also complements Shen's study on the supports of products of canonical basis elements for cluster X-varieties of type $A_n$.

math.RT↗
arXiv2026

FreDF: Learning to Forecast in the Frequency Domain

Time series modeling presents unique challenges due to autocorrelation in both historical data and future sequences. While current research predominantly addresses autocorrelation within historical data, the correlations among future labels are often overlooked. Specifically, modern forecasting models primarily adhere to the Direct Forecast (DF) paradigm, generating multi-step forecasts independently and disregarding label autocorrelation over time. In this work, we demonstrate that the learning objective of DF is biased in the presence of label autocorrelation. To address this issue, we propose the Frequency-enhanced Direct Forecast (FreDF), which mitigates label autocorrelation by learning to forecast in the frequency domain, thereby reducing estimation bias. Our experiments show that FreDF significantly outperforms existing state-of-the-art methods and is compatible with a variety of forecast models. Code is available at https://github.com/Master-PLC/FreDF.

cs.LG↗
arXiv2026

Time-o1: Time-Series Forecasting Needs Transformed Label Alignment

Training time-series forecasting models poses unique challenges in loss function design. Most existing approaches adopt temporal mean squared error, but this study reveals two critical limitations: (1) it ignores the presence of label autocorrelation, which biases it from the true label sequence likelihood; (2) it involves excessive number of tasks, which complicates optimization, especially for long-term forecasting. To address these issues, we introduce Time-o1, a transform-enhanced loss function for time-series forecasting. The central idea is to transform the label sequence into decorrelated components with discriminated significance. Models are then trained to align the most significant components, thereby effectively mitigating label autocorrelation and reducing task amount. Experiments demonstrate that Time-o1 achieves state-of-the-art performance and is compatible with various forecast models. Code is available at https://github.com/Master-PLC/Time-o1.

cs.LG↗
arXiv2026

DistDF: Time-Series Forecasting Needs Joint-Distribution Wasserstein Alignment

Training time-series forecasting models requires aligning the conditional distribution of model forecasts with that of the label sequence. The standard direct forecast (DF) approach resorts to minimizing the conditional negative log-likelihood, typically estimated by the mean squared error. However, this estimation proves biased when the label sequence exhibits autocorrelation. In this paper, we propose DistDF, which achieves alignment by minimizing a distributional discrepancy between the conditional distributions of forecast and label sequences. Since such conditional discrepancies are difficult to estimate from finite time-series observations, we introduce a joint-distribution Wasserstein discrepancy for time-series forecasting, which provably upper bounds the conditional discrepancy of interest. The proposed discrepancy is tractable, differentiable, and readily compatible with gradient-based optimization. Extensive experiments show that DistDF improves diverse forecasting models and achieves leading performance. Code is available at https://anonymous.4open.science/r/DistDF-F66B.

cs.LG↗
arXiv2026

Quadratic Direct Forecast for Training Multi-Step Time-Series Forecast Models

The design of learning objectives is central to training time-series forecasting models. Existing learning objectives such as mean squared error mostly treat each future step as an independent, equally weighted task, which leads to the following two challenges: (1) they overlook the label autocorrelation effect among future steps, leading to biased learning objectives; (2) they fail to set heterogeneous task weights for different forecasting tasks corresponding to varying future steps, limiting the forecasting performance. To fill this gap, we propose a novel quadratic-form weighted learning objective, addressing both issues simultaneously. Specifically, the off-diagonal elements of the weighting matrix account for the label autocorrelation effect, whereas the non-uniform diagonals are expected to match the preferred weights of the forecasting tasks with varying future steps. On this basis, we propose a Quadratic Direct Forecast (QDF) learning algorithm, which trains the forecast model using the adaptively updated quadratic-form weighting matrix. Experiments show that our QDF effectively improves the performance of various forecast models, achieving state-of-the-art results. Code is available at https://github.com/Master-PLC/QDF.

cs.LG↗
arXiv2026

Deep Time-Series Forecasting in 10 Years: A Survey

Autocorrelation is a common property of time-series, where each observation is dependent on its predecessors. In deep time-series forecasting, it raises two central challenges: (1) designing backbone architectures to model autocorrelation in history sequences, and (2) devising loss functions to model autocorrelation in label sequences. Recent studies have made strides in tackling these challenges, but a systematic survey examining both aspects remains lacking. To bridge this gap, this paper reviews deep time-series forecasting from an autocorrelation modeling perspective, offering two contributions beyond existing surveys. First, it introduces a taxonomy that jointly covers both backbone architectures and loss functions, whereas prior surveys provide limited coverage of the latter. Second, it analyzes the motivations and insights underlying the surveyed literature from a unified autocorrelation perspective, providing a holistic overview of the field's evolution. Additional resources and details are available at https://github.com/Master-PLC/Awesome-TSF-Papers.

stat.ML↗
arXiv2026

Unbiased Reward Modeling from Implicit Feedback for LLM Alignment

Despite the success of reinforcement learning from human feedback (RLHF), existing reward modeling methods largely rely on explicit feedback, which is costly to collect and difficult to scale. This work studies implicit reward modeling, learning reward models from implicit user feedback, such as clicks, copies and skips. While scalable and cost-effective, implicit feedback poses two key challenges: It lacks definitive negative samples, which makes standard positive-negative classification methods inapplicable; It suffers from selection bias, where responses have heterogeneous propensities to elicit feedback, which further obscures definitive negative samples. To address these challenges, we propose ImplicitRM, which learns unbiased reward models from implicit feedback. It stratifies training samples into four latent groups using a stratification model and derives a likelihood-maximization objective that is theoretically unbiased, thereby addressing both challenges. Experiments across diverse LLM backbones and benchmark datasets validate that ImplicitRM learns accurate reward models from implicit feedback and improves performance on downstream RLHF tasks.

cs.CL↗
arXiv2026

Rubrics on Trial: Evolving Rubrics from a Single Query via Synthetic Pairwise Evidence

Rubric evolution offers a promising approach to improving the quality of rubrics generated by large language models (LLMs). Central to this process is rubric comparison, which identifies the better of two rubrics and guides the direction of evolution. However, accurate rubric comparison is difficult, which presents two challenges. (1) It should reflect downstream task performance, which is essential for assessing rubric utility but often prohibitively expensive to evaluate. (2) It should discourage unnecessary criteria, which increase verification costs and may dilute the influence of essential criteria. To address these challenges, we introduce Rubrics on Trial, a multi-agent framework that evolves rubrics by comparing synthetic response pairs. To address challenge 1, the framework compares synthetic responses that satisfy the respective rubrics, providing a proxy for downstream performance without training a separate policy for each rubric. To address challenge 2, it assesses the necessity of a candidate criterion by independently generating high-quality alternative responses that violate it and comparing them with edited versions that satisfy it. A rubric is favored when it improves response quality in both comparisons, and the resulting comparison signal is further incorporated for rubric evolution. Extensive experiments demonstrate that Rubrics on Trial improves the quality of generated rubrics and leads to better downstream task performance.

cs.CL↗
arXiv2026

Integrated yoctosecond-precision timing detector

Precise timing detection is essential for exploring ultrafast phenomena in fields ranging from free-electron lasers to ultra-high-power laser facilities. However, achieving simultaneous high resolution and large dynamic range remains a fundamental challenge, and state of the art systems are often constrained by their physical size, power requirements, and limited scalability. Here we introduce an integrated dual electro optic sub cycle timing detector (DEST) that overcomes these limitations. In a proof of principle measurement, the device resolves timing jitter as small as 11 yoctoseconds (ys, $10^{-24}$ s) at 1 MHz--equivalent to the transit time of light across two protons--while maintaining an unambiguous measurement range of 6.15 ps and a dynamic range exceeding 155 dB. The core detection unit is miniaturized to chip scale dimensions of 18 mm x 2 mm x 1 mm on a thin film lithium niobate platform, ensuring inherent stability and immunity to environmental disturbances. Moreover, the architecture naturally lends itself to massive parallelization through array integration, with the potential to push timing precision to the sub 10 rontosecond (rs, $10^{-27}$ s) level within a 1 $m^2$ footprint. This combination of extreme sensitivity, wide dynamic range, compact size, and scalability opens new avenues for detecting previously inaccessible weak signals, including those from gravitational waves, quantum vacuum fluctuations, and beyond.

physics.optics↗
arXiv2026

Unfolding the Interdisciplinary Complexities of Climate Science: Fuxi-Climate Foundational Model

Climate research and decision-making require integrating evidence across physical processes, socio-economic dynamics and policy responses. Large language models (LLMs) have been explored for accessing and synthesizing climate knowledge, but their ability to support structured interdisciplinary reasoning is still limited. Here we present the Fuxi-Climate Foundation Model (CFM), a climate-specialized LLM designed to support consistent reasoning across domains. CFM maintains more stable analytical behavior as interdisciplinary complexity increases, whereas performance in other models becomes more variable. On expert-designed climate transition tasks, CFM produces more structured analyses that explicitly address trade-offs and uncertainty, achieving 45% trade-off coverage and 47.27% uncertainty-aware reasoning. These results indicate that CFM can support more realistic analysis of climate risks and transition pathways, and provide a basis for agent-based systems to explore complex policy and decision scenarios. The model is openly available at https://huggingface.co/SII-yuning/cfm.

cs.CY↗
arXiv2026

Mutation-preserving generalized cluster algebras and Laurent mutation invariants

We introduce mutation-preserving generalized cluster algebras, for which the generalized cluster mutation in each direction is independent of the seed in the mutation equivalence class. We classify all irreducible generalized cluster algebras with this property. Then, a Markov-type Diophantine equation $x^2+y^2+z^2+2yz=kxyz$ is studied, which has a structure of the mutation-preserving generalized cluster algebra. We prove that positive integer solutions exist if and only if $1\leq k\leq 5$ and determine all their orbits under the associated generalized cluster mutation groups. In particular, multiple orbits occur for each $k=1,3$, whereas the solutions form a single orbit for each $ k=2,4,5$. We then classify all generalized Markov Laurent mutation invariants. As an application, a conjecture proposed by Chen-Li is proved, showing that every Laurent mutation invariant of irreducible sign-equivalent cluster algebras is essentially a polynomial in the corresponding basic invariant.

math.RA↗
arXiv2026

Battery-Swapping Station Operation Under Forecast Uncertainty: A Scenario-Based Stochastic MPC Framework

Battery-swapping stations (BSSs) can shorten electric-vehicle energy replenishment while using centrally managed battery inventories as flexible grid-connected storage. Realizing both benefits requires the station to schedule charging, grid discharge, and swapping service before future customer demand and electricity prices are known. This paper develops a forecast-aware rolling-horizon operating framework for this problem. A lightweight DLinear model predicts 24-hour price and demand trajectories, Stein variational gradient descent quantifies their uncertainty through representative scenarios, and a two-stage stochastic model predictive controller converts those scenarios into station decisions. The controller accounts for service shortfall, terminal readiness, a protected service buffer, and electrochemical degradation without assuming perfect future information. The application contribution is an implementable controller that coordinates the station's mobility-service and energy-storage roles. The methodological contribution is a modular forecast-to-control interface that separates the operational value of mean-forecast accuracy from that of uncertainty representation. In a 120-day closed-loop evaluation, DLinear-SVGD SMPC achieves the lowest cost among the implementable controllers. Relative to deterministic DLinear MPC, it reduces final cost by 1.2\% and service-shortfall hours by 80.7\%, with 99.10\% of the evaluated hours free of shortfall.

math.OC↗
arXiv2026

Rethinking the Flow-Based Gradual Domain Adaptation: A Semi-Dual Optimal Transport Perspective

Gradual domain adaptation (GDA) aims to mitigate domain shift by progressively adapting models from the source domain to the target domain via intermediate domains. However, real intermediate domains are often unavailable or ineffective, necessitating the synthesis of intermediate samples. Flow-based models have recently been used for this purpose by interpolating between source and target distributions. Notably, their training typically relies on sample-based log-likelihood estimation, which can discard useful information and thus degrade GDA performance. The key to addressing this limitation is constructing the intermediate domains via samples directly. To this end, we propose an Entropy-regularized Semi-dual Unbalanced Optimal Transport (E-SUOT) framework to construct intermediate domains. Specifically, we reformulate flow-based GDA as a Lagrangian dual problem and derive an equivalent semi-dual objective that circumvents the need for likelihood estimation. However, the dual problem leads to an unstable min-max training procedure. To alleviate this issue, we further introduce the entropy regularization to convert it into a more stable sequential optimization procedure. Based on this, we propose a novel GDA training framework and provide theoretical analysis in terms of stability and generalization. Finally, extensive experiments are conducted to demonstrate the efficacy of the E-SUOT framework.

cs.LG↗
arXiv2026

Geometric structures of $G$-fans associated with rank $3$ cluster-cyclic exchange matrices

In this paper, we investigate the geometric structures of $G$-fans associated with rank $3$ real cluster-cyclic exchange matrices. In this class, a simple recursion for tropical signs was found, which enables us to study the detailed properties of $c$-, $g$-vectors. We introduce two kinds of upper bounds of the $G$-fans. The first one is the global upper bound, which comes from a hyperbolic surface containing all $g$-vectors after an initial mutation. The second one is the local upper bound, which reflects the internal separateness structure. As applications, we prove that there is no periodicity among $g$-vectors, and we completely determine the sign of $g$-vectors. We also prove the monotonicity of $g$-vectors under the minimum assumption. Moreover, we show that the three global upper bounds can be simplified to a single uniform upper bound.

math.CO↗
arXiv2026

Real $C$-, $G$-structures and sign-coherence of cluster algebras

We generalize the theory of integer $C$-, $G$-matrices in cluster algebras to the real case. By a skew-symmetrizing method, we can reduce the problem of skew-symmetrizable patterns to skew-symmetric patterns. In this sense, the sign-coherence of a more general real class called of quasi-integer type can be inherited directly from that of integer $C$-, $G$-matrices proved by Gross-Hacking-Keel-Kontsevich. However, the sign-coherence of real $C$-, $G$-matrices does not always hold in general. For this purpose, we classify all the rank $2$ case and the finite type case via the Coxeter diagrams. We also give two conjectures about the real exchange matrices and $C$-, $G$-matrices. Under these conjectures, the dual mutation, $G$-fan structure and synchronicity property hold. As an application, the isomorphism of several kinds of exchange graphs is studied.

math.RT↗
arXiv2026

Efficient, Validation-Free Intrinsic Quality Estimation for Large-Scale Face Recognition Datasets

We propose Intrinsic Quality (IQ), a validation-free metric designed to estimate the inherent potential of face recognition (FR) datasets to produce high-performance models without the need for full-scale training. IQ integrates two components: (i) a Neighbor-Consistency Score that quantifies local identity label agreement via nearest neighbors, and (ii) Global Representation Subspace Complexity (Effective Rank, ER), which captures the underlying embedding geometry and dataset diversity. IQ allows for rapid evaluation using lightweight proxy models or data subsets, facilitating dataset diagnosis and curation prior to resource-intensive full-scale training. We describe an experimental protocol tailored to clean, noisy, and mixed-quality FR datasets, and outline evaluation methodologies to validate IQ's predictive power for downstream performance.

cs.CV↗